Encyclopedia of

Plant Physiology
New Series Volume 15 A

Editors
A. Pirson, G6ttingen
M.H. Zimmermann, Harvard
Inorganic
Plant Nutrition
Edited by
A. LaucWi and RL. Bieleski

Contributors
C.l Asher L. Beevers R T. Besford R.L. Bieleski
P. Boger E.G. Bollard H. Bothe D. Bouma G.D. Bowen
F.C. Cannon C.C. Delwiche 1 Dobereiner W.M. Dugger
D.G. Edwards lB. Ferguson T.l Flowers RC. Foster
W.H. Gabelman G.C. Gerloff R.H. Hageman A. Lauchli
U. Liittge D. Marme H. Marschner 1 Moorby M.G. Pitman
A. Pollard A. Quispel A.D. Robson R Roth A.D. Rovira
G. Sandmann lA. Schiff W.R. Ullrich D. Werner
R.G. WynJones M.G. Yates

With a Foreword by E. Epstein

With 131 Figures

Springer-Verlag
Berlin Heidelberg NewYork Tokyo 1983
Professor Dr. A. LXUCHLI
University of California
Department of Land, Air and Water Resources
Hoagland Hall
Davis, CA 95616jUSA

Dr. R.L. BmLESKI

Department of Scientific and Industrial Research
Division of Horticulture and Processing
Private Bag
Auckland
New Zealand

ISBN-13 :978-3-642-68887-4 e-ISBN-13: 978-3-642-68885-0

DOl: 10.1007/978-3-642-68885-0

Library of Congress Cataloging in Publication Data. Main entry under title: Inorganic plant nutrition.
(Encyclopedia of plant physiology; new ser., v. 15) 1. Plants-Nutrition. I. Liiuchli, A. (Andre), 1933-.
II. Bieleski, R.L. (Roderick Leon), 1931-. III. Series. QK711.2.E5 n. s., vol. 15 581.1 s [581.1'3] 83-9861
[QK867] ISBN-13 :978-3-642-68887-4 (U.S.).
This work is subject to copyright. All rights are reserved, whether the whole or part of the material is
concerned, specifically those of translation, reprinting, re-use of illustrations, broadcasting, reproduction
by photocopying machine or similar means, and storage in data banks.
Under §54 of the German Copyright Law where copies are made for other than private use, a fee is
payable to "Verwertungsgesellschaft Wort" Munich.
© by Springer-Verlag Berlin-Heidelberg 1983
Softcover reprint of the hardcover 1st edition 1983
The use of registered names, trademarks, etc. in this publication does not imply, even in the absence
of a specific statement that such names are exempt from the relevant protective laws and regulations
and therefore free for general use.

2131/3130-543210
Foreword

The first book bearing the title of this volume, Inorganic Plant Nutrition, was
written by D.R. HOAGLAND of the University of California at Berkeley. As
indicated by its extended title, Lectures on the Inorganic Nutrition of Plants,
it is a collection of lectures - the JOHN M. PRATHER lectures, which he was
invited in 1942 to give. at Harvard University and presented there between
April 10 and 23 of that year - 41 years before the publication of the present
volume. They were not "originally intended for publication" but fortunately
HOAGLAND was persuaded to publish them; the book appeared in 1944.
It might at first blush seem inappropriate to draw comparisons between
a book embodying a set of lectures by a single author and an encyclopedic
volume with no less than 37 contributors. But HOAGLAND'S book was a compre-
hensive account of the state of this science in his time, as the present volume
is for ours. It was then still possible for one person, at least for a person
of HOAGLAND'S intellectual breadth and catholicity of interests, to encompass
many major areas of the entire field, from the soil substrate to the metabolic
roles of nitrogen, potassium, and other nutrients, and from basic scientific topics
to the application of plant nutritional research in solving problems encountered
in the field. Thus despite HOAGLAND'S admittedly more personal approach,
a comparison of these two books serves to drive home the enormous progress
that has been made in this science in the intervening years - and draws attention
to areas that have lagged behind the broad surge of advancement of the field
as a whole.
The first and most pervasive impression to emerge from a perusal of the
two books is the effect that new techniques have had on the progress of inorganic
plant nutrition, and the conviction that dazzling prospects are opening up by
further advances in research tools and their application to this science. Just
before America's entry into World War II HOAGLAND, in collaboration with
the late, great P.R. STOUT, was among the very first investigators to apply
radioisotopes of essential inorganic nutrients to the study of plant nutrition,
in particular, of ion transport. I well remember the early days of radioactive
tracers when, as a graduate student in that laboratory, I was in on the resump-
tion of that work in the years immediately following World War II. We did
our own target chemistry on material irradiated in the cyclotron up "on the
hill." Our Geiger-Muller tubes were hand-made in the laboratory. I even had
to determine the half-life of one of the radioisotopes I used, manganes,e-52,
which had not been determined, and was obliged to do the job with a quartz
fiber electroscope because our G.-M. counters were so unstable in their perfor-
mance from day to day.
HOAGLAND relished this revolutionary innovation, but he did not live to
see the enormous developments that the use of radioisotopes, and of stable
VI Foreword

isotopic tracers, especially nitrogen-15, would make possible in every branch

of plant nutrition, from field studies to biochemical investigations. In the present
volume nearly all chapters bear witness to the utility of isotopic tracers as
powerful tools in the investigation of every aspect of this science.
Virtually simultaneous with the advent of tracer methodology was the devel-
opment of chromatographic and electrophoretic techniques in all their variety.
Their effect has been most pronounced in biochemical investigations. In the
present context, they have played a large role in the advances in our understand-
ing of the metabolism of nitrogen, phosphorus, and sulfur.
Biochemical events occur in space and time. The traditional approach of
biochemists has been to disregard the former by disrupting the cell and studying
the reactions between biochemical entities as a function of time - hence the
emphasis on kinetics in classical biochemistry. Plant nutritionists cannot disre-
gard space: membranes, sites, compartments, transport, immobilization, seques-
tration, gradients, and fluxes are of the very essence in inorganic plant nutrition.
Hence plant nutritionists have had to deal with structure, at levels ranging
from the whole plant to subcellular entities. It is in studies of spatial relation-
ships, especially in tissues and cells, that new technological tools-of-the-trade
have made possible advances undreamed of 40 years ago. Electron probe X-ray
microanalysis, laser probe analysis, ion probe mass analysis, and still other
methods for pinpointing the distribution and localization of nutrient and other
elements represent a whole new armamentarium at the disposal of the plant
nutritionist. These and other localization techniques are already creating a revo-
lution in inorganic plant analysis, yielding information entirely beyond the scope
of traditional analytical techniques. Ion-sensitive microelectrodes are in this
kit of new tools. Methods for the isolation of membranes, protoplasts, vacuoles,
and other cellular organelles are also increasingly useful in sorting out the traffic
patterns of inorganic nutrients (and other elements) in space and time.
One of HOAGLAND'S best-known contributions to whole-plant nutrition was
his development, with his long-time collaborator T.e. BROYER, of what came
to be known as "Hoagland solution," or rather, solutions, because there are
two, the second containing ammonium as well as nitrate. These solutions are
not in principle different from other such solutions that had been in use since
the middle of the 19th century. In all of them, most nutrients are present
in concentrations far in excess of their concentrations in typical soil solutions,
to enable plants to grow for long periods by drawing on a large supply of
nutrients contained in conveniently small volumes of solution.
HOAGLAND was aware that plants can grow well at much lower concentra-
tions of nutrients in the medium, and do so in nature. But progress in developing
the technology of culture media automatically maintained at predetermined,
realistically low concentrations has been slow and sporadic. There is now, how-
ever, a much wider awareness of the limitations of high-concentration, Hoag-
land-type nutrient solutions, and an awareness as well that the technology for
the development of sophisticated, low-concentration solution culture systems
is "on the shelf." Progress so far in this field is recorded in the present volume.
It will be exciting to read the report on this subject in the next edition of
this work.
Foreword VII

Fundamental in any nutritional science is the question: what are the nu-
trients? In inorganic plant nutrition, the problem is to determine what elements
besides carbon, hydrogen, and oxygen are essential for the life of the plant.
Forty years ago much attention was paid to this topic, with HOAGLAND and
his collaborators in the van of the effort. The essentiality of molybdenum for
higher green plants was discovered in HOAGLAND'S Division of Plant Nutrition
and with his inspiration. Not long after his death in 1949, the impetus of this
and related work led to the discovery of the essentiality of chlorine in the
same laboratory. There followed the demonstration of the essentiality of cobalt
for legumes fixing nitrogen symbiotically, accomplished independently by two
groups, including the one founded by HOAGLAND.
As in the other instances discussed so far, it was the advent of new materials
and techniques that made these discoveries possible. The main material was
borosilicate glass, and the techniques were chemical procedures for purging
the experimental solutions of inadvertent contaminants. Since that time new
experimental and purification techniques have developed apace. They have been
used by animal nutritionists to establish the essentiality of a whole raft of
"new" inorganic nutrients: fluorine, silicon, vanadium, chromium, nickel, arse-
nic, and selenium. The present volume does not, however, record a comparable
interest in and effort toward the discovery of new plant micronutrients. On
the other hand, great progress has been made in our understanding of the
roles played by the known micronutrients, especially the heavy metals, in the
function of enzymes and electron transfer processes.
New techniques, new machines and new methodologies thus have greatly
advanced our knowledge of plant nutrition beyond its status 40 years ago. The
cornucopia of the chemist, the physicist, and the engineer will continue to pour
forth new tools that plant nutritionists will wield in the laboratory, the green-
house, and the field.
The present volume, however, delineates developments other than those that
depend largely or exclusively on new technology. Inorganic plant nutrition in
both its classical period (the 19th century) and its neoclassical development
(in the first half of the present century) bears the imprint of, mainly, continental
Europe and the British Isles, the United States, and Australia. The motivation
for pursuing it came almost exclusively from agriculture: plant nutritional re-
search dealt with beans, barley, wheat, corn, tobacco, alfalfa, soybeans, and
a few other economically important species. Considering the wide world and
its wealth of varied plant life, plant nutrition thus has had an alarmingly narrow
focus in terms of geography, ecosystems, and experimental plant materials.
The present volume documents a broadening of the intellectual scope of the
science of inorganic plant nutrition - almost entirely a development of the
last few decades. For the further advancement of our science this development
is no less significant than the new knowledge made possible by new techniques.
In addition, it is extending the application of plant nutritional knowledge to
agriculture everywhere, and to enterprises other than agriculture as well.
As for science per se, the processes of plant nutrition do more than nourish
the plant. They are instrumental in injecting huge ~mounts of nutrient and
other elements into the processes of global biogeochemical cycling. All elements
VIII Foreword

are involved to some extent, but the drain of phosphorus into the oceanic
sinks is the one of greatest concern. Unlike nitrogen, phosphorus is not recycled,
and its terrestrial supplies are ftnite.
Phosphate is of interest in another context. It is absorbed by most plants
in a symbiotic relationship with fungi. Mycorrhizae seem to be nearly universal
in terrestrial plants. Interest in mycorrhizae and their effects on the acquisition
of phosphate (and some other nutrients) is unlikely to rival that in nitrogen
fixation through the symbiosis between legumes and Rhizobium, but in due
time it may come close.
Mycorrhizae had long been known to occur in forest trees. The recent realiza-
tion that they are almost universal in terrestrial plants including crops has
been one impetus, among several, leading to a greater rapport between plant
nutritionists and ecologists than existed heretofore. Other factors in this interdis-
ciplinary rapprochement have been (a) the realization of the importance of
high concentrations of heavy metals in both natural and "manhandled" eco-
systems; (b) the recognition that nutrient deftciencies can playa major, often
the preponderant, role in the distribution of wild species, and (c) the under-
standing that agricultural plant nutritionists interested in salinity and ecologists
studying halophytes have much to learn from each other.
The response of plants to the adverse conditions of high metal concentra-
tions, low soil fertility, and salinity is a function of the genotype. The possibility
therefore exists of applying the concepts and techniques of physiological and
biochemical genetics to the study of the responses of plants to these edaphic
features. It is only in the last few decades that this possibility has begun to
be realized; even now, the entire subject is still in its pioneering phase. What
has been accomplished clearly demonstrates the potential of this approach,
and the prospects are enticing.
The broader concepts and interdisciplinary connections of inorganic plant
nutrition that have emerged in recent decades have profound implications for
the applications of this science in the service of mankind. Sophisticated plant
nutritional science will increasingly spread to areas where agriculture has re-
mained in a traditional mold, and be applied to crops it has so far largely
ignored, including tropical ones. Many tropical soils, especially in the humid
tropics, present vexing plant nutritional problems: they are often acidic, with
high concentrations of soluble aluminum, manganese, or both, and have an
inordinate propensity for ftxing phosphate in forms unavailable to plants. Other
nutritional problems, especially micronutrient deftciencies, are endemic in many
of them. There is no way in which the traditional methods for coping with
these problems - liming, use of amendments and fertilizers - can by themselves
cure these ills. It will be necessary to develop lines of crops better adapted
to these conditions than are the present varieties.
But it is by no means only in the tropics that a genetic appr9ach will have
to be taken. The crops of the agriculturally advanced countries have been select-
ed and bred to bestow on them desirable traits such as disease resistance, cold
hardiness, early maturation, and many more. They have also been bred for
high yield under nutritionally luxuriant conditions of ample fertilization. They
may thus have been selected inadvertently against effIciency in mineral nutrition,
Foreword IX

including avid absorption from soils of low fertility in respect to one or more
inorganic nutrients, rapid translocation of nutrients throughout the plant body,
effective storage when supplies are ample, and frugal metabolic utilization when
their concentrations in the tissue are relatively low. It is thus quite possible
that many of the successful crops of the temperature zone are genetically defec-
tive or crippled in their plant nutritional capabilities.
It is a reflection on our science that it is based in so large a measure on
findings from experiments with these plants. More attention than is evident
in this volume should be paid to the inorganic nutrition of wild plants - plants
that have been under the selective pressure of often scanty nutrient supplies
throughout their evolutionary history.
In the future it may no longer be possible, even in agriculturally advanced
countries, to apply fertilizer virtually ad lib, because of constraints in availabi-
lity, dollar cost, energy cost, and objections on environmental grounds. In many
agriculturally less advanced countries such profligate application of fertilizer
has never been affordable. These considerations lend force to the argument
that plant nutritionists ought to collaborate with geneticists and breeders to
develop nutrient-efficient genotypes of crops. They will take maximal advantage
of the natural fertility of the soil, and of what fertilizer may be applied out
of the bag. Plant nutritional studies of biologically productive wild plants can
provide guidelines as to what traits should be incorporated into these new,
nutrient-conserving varieties.
Salinity impairs agricultural productivity in arid and semi-arid regions where
crop production depends on irrigation. Like heavy metal toxicity and low ferti-
lity, this adverse condition has also been dealt with primarily by capital-inten-
sive, energy-intensive measures: reclamation, drainage, overirrigation, and other
technological operations. The existence of halophytes and of heritable variation
in salt tolerance even in crop species points the way to a genetic approach
to this problem as well: the development of salt-tolerant crops. This represents
yet another challenge for collaboration between plant nutritionists and plant
geneticist-breeders.
It may be that soils now considered marginal for agricultural production
can be used to good advantage for this purpose if crops genetically tailored
to cope with the soil conditions are developed. Still poorer soils may surely
be used for the production of energy crops or the cultivation of plants yielding
special materials such as lubricants and medicinals. The joint efforts of plant
nutritionists and breeders will be needed in these labors as well.
The editors of this volume have assembled an array of distinguished scientists
who collectively report a wealth of new knowledge derived from the imaginative
and skillful application of powei:ful new tools and methodologies, and who
convey the message that inorganic plant nutrition is broadening its intellectual
perspective and forging interdisciplinary linkages with other plant and environ-
mental sciences on a scale that has no precedent. Users of this volume whose
reading measures up to the work that authors and editors have put into it
will be amply rewarded, and spurred on in their efforts to advance still further
the science of inorganic plant nutrition.
EMANUEL EpSTEIN
Contents Part A

Introduction
A. LXUCHLI and R.L. BIELESKI 1

I. General Chapters of Inorganic Plant Nutrition

1.1 General Introduction to the Mineral Nutrition of Plants

H. MARSCHNER (With 11 Figures)

1 Introduction and Historical Resume . . . . . 5

1.1 Essential Mineral Elements - Plant Nutrients 5
1.2 Function of Essential Mineral Elements 6
1.3 Beneficial Mineral Elements 7
1.4 Recent Developments 9
1.4.1 Calcium 10
1.4.2 Potassium . 11
1.4.3 Phosphorus 13
1.4.4 Nitrogen 13
1.4.5 Copper . . 14
1.4.6 Chlorine 15
2 Uptake and Long-Distance Transport of Mineral Elements 16
2.1 Ion Concentration at the Root Surface, Role of the "Rhizosphere" 16
2.2 Long-Distance Transport in the Xylem . . . . . . . 18
2.2.1 From the Roots to the Shoot . . . . . . . . . 18
2.2.2 Into Fruits, Seeds and Storage Organs . . . . . 19
2.2.3 Retranslocation of Mineral Elements from Leaves 20
3 Calcium Nutrition of Higher Plants 22
3.1 Introduction . . . . . . . . . . . 22
3.2 Calcium Demand of Higher Plants 22
3.3 Calcium Uptake by the Roots 23
3.4 Long-Distance Transport of Calcium 23
3.4.1 Xylem Transport . . . . . . 23
3.4.2 Phloem Transport . . . . . . 26
3.4.3 Xylem Versus Phloem Transport 27
3.5 Role of Phytohormones and Growth Regulators 29
3.6 Conclusion and Outlook . . . . . . . . . . . 29
4 Mineral Nutrition and Physiology of Yield Formation - Sink-Source
Relationship ....................... 30
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . 30
4.2 Effect of Mineral Nutrition on Phytohormone Level and Sink
Formation . . . . . . . . . . . . . . . . . . . . . 31
4.3 Effect of Mineral Nutrients on Fertilization . . . . . . 33
4.4 Source-Sink Interactions in Relation to Mineral Nutrition 34
5 Environmental Aspects of Mineral Nutrition 37
5.1 Introduction . . . 37
5.1.1 Nitrogen 38
5.1.2 Heavy Metals . . . . . . . . . 39
XII Contents Part A

5.2 Heavy Metal Toxicity . . . . . . . 39

5.3 Heavy Metals in the Food Chain . . 40
5.4 Heavy Metals in the SoiljPlant System 41
5.4.1 Content of Soils . . . . . . . 41
5.4.2 Soil Factors Affecting Heavy Metal Accumulation in Plants 43
5.4.3 Genotypic Differences in Heavy Metal Uptake . 44
5.4.4 Distribution Within the Plants and Their Organs 46
5.4.5 Heavy Metal Tolerance 48
5.5 Concluding Remarks 49
References . . . . . . . . . . 49

1.2 The Significance of Rhizosphere Microflora and Mycorrbizas

in Plant Nutrition
A.D. ROVIRA, G.D. BOWEN, and R.C. FOSTER (With 7 Figures)
1 Introduction . . . . . . . . . 61
2 Energy Supplies in the Rhizosphere 61
2.1 Exudates 61
2.2 Secretions . . . 62
2.3 Plant Mucilages 62
2.4 Mucigel . . . . 63
2.5 Lysates 63
3 Microbiology of the Rhizosphere 64
3.1 Populations of Micro-Organisms 64
3.2 Colonization of Roots by Micro-Organisms 64
4 Mathematical Modelling of the Rhizosphere 65
5 Microscopy of the Rhizosphere . . . . . . . 66
5.1 Light Microscopy . . . . . . . . . . . . 66
5.2 Scanning Electron Microscopy (S.E.M.) . . 66
5.3 Transmission Electron Microscopy (T.E.M.) 69
5.3.1 General Description . . . . . . . . 69
5.3.2 Origin and Fine Structure of Root Mucilage 70
5.3.3 Microbial Invasion of the Mucilage and the Formation of Mucigel 72
5.3.4 Functions of Root Mucilage and Mucigel . . . . 72
5.3.5 The Outer Rhizosphere . . . . . . . . . . . . 72
5.3.6 Invasion of the Root by Microorganisms 73
6 The Role of Rhizosphere Microorganisms in Plant Nutrition 73
6.1 Availability of Nutrients . . . . . . . . 74
6.1.1 Nutrient Release and Immobilization 74
6.1.2 Nitrification and Denitrification 74
6.1.3 Nitrogen Fixation . . . . . 74
6.1.4 Phosphate Availability 74
6.1.5 Minor Nutrients . . . . . . 74
6.2 Growth and Morphology of Roots 75
6.2.1 Root Length and Root Hairs 75
6.2.2 Proteoid Roots 75
6.3 Nutrient Uptake Processes . 76
6.4 Physiology and Development 76
7 Mycorrhizas . . . . . . . . . 76
7.1 Plant Responses to Infection 78
7.2 Mechanisms of the Response 79
7.2.1 Nutrient Availability . 80
7.2.2 Absorption Characteristics of the Root 80
7.2.3 Absorption by the Fungus Component 80
7.3 Energy Requirements of Mycorrhizas 82
7.4 Overview of Mycorrhizas . . . . . . . . 83
Contents Part A XIII

8 General Conclusions 84
References . . . . . 86

1.3 Modern Solution Culture Techniques

C.J. AsHER and D.G. EDWARDS (With 3 Figures)
1 Major Differences Between Solution Culture and Soil Culture 94
1.1 Mechanical Support . . . . . . . . . . . . . . . 94
1.2 Spatial Variation in Root Environment Parameters 95
1.3 Temporal Variation in Root Environment Parameters 97
1.3.1 Nutrient Depletion . . . . . . . . . . . . . 97
1.3.2 pH Shifts . . . . . . . . . . . . . . . . . 98
1.4 Root-Microorganism Interactions . . . . . . . . . 98
2 Uses and Limitations of Existing Solution Culture Methods 99
2.1 Non-Renewed or Intermittently Renewed Water Cultures and Sand
Cultures . . . . . . . . . . . . . . . . . . . . 99
2.1.1 Use in Teaching, Demonstration, and Diagnosis 99
2.1.2 Production of Roots for Ion Transport Studies 101
2.1.3 Nutrient Essentiality . . . . . . . . . . . . 101
2.1.4 Effects of Root Environment Parameters 101
2.1.5 Establishment of Critical Tissue Concentrations 105
2.1.6 Control of Plant Nutrient Status . . . . . . . 105
2.1.7 Study of Symbiotic Associations with Microorganisms 106
2.1.8 Commercial Crop Production 108
2.2 Mist Culture . . . . . . . . . . . . . . . . 108
2.3 Flowing Solution Culture . . . . . . . . . . 109
2.3.1 The Flow Rate Problem . . . . . . . . 109
2.3.2 Composition of Flowing Culture Solutions 110
2.3.3 Research Applications 111
2.3.4 Likely Future Developments . 113
2.3.5 Commercial Crop Production 114
3 Summary and Conclusions 115
References . . . . . . . . . . . . . 115

1.4 Diagnosis of Mineral Deficiencies Using Plant Tests

D. BOUMA (With 5 Figures)
1 Introduction 120
2 Plant Analysis . . . . . 121
2.1 Physiological Basis 121
2.2 Choice of Tissue .. 123
2.3 Factors Affecting the Relationship Between Nutrient Concentration and
Yield . . . . . . . . . . . . . . . . 124
2.3.1 Plant Development . . . . . . . . . 124
2.3.2 Effects of Changes in Age of Tissue 125
2.3.3 Plant Age and Critical Levels 126
2.3.4 Interactions Betweeri Nutrient Elements 127
2.3.5 Environmental Factors . . . . . . . 128
2.3.6 Other Factors Affecting Nutrient Composition 130
3 Physiological and Biochemical Approaches to Diagnosis 131
3.1 Introductory Remarks . . . . 131
3.2 Physiological Approaches . . . . . . . 131
3.2.1 Physiological Assessment . . . . . 131
3.2.2 Nutrient Stress . . . . . . . . . 132
3.2.3 Approaches Based on Photosynthesis 132
3.2.4 Other Approaches . . . . . . . . 133
XIV Contents Part A

3.3 Biochemical Approaches . . . . 135

3.3.1 Nitrogen and Molybdenum 135
3.3.2 Phosphorus . . . . . . . 136
3.3.3 Potassium and Magnesium 137
3.3.4 Iron and Manganese 138
3.3.5 Copper . . . . 139
3.3.6 Zinc . . . . . 140
4 Prospects for the Future 140
References . . . . . . . 141

1.5 Interactions Between Nutrients in Higher Plants

A.D. ROBSON and M.G. PITMAN (With 9 Figures)
1 Introduction ........................... 147
2 Interactions Between Nutrients in Monoculture ............ 152
2.1 Interactions Between Nutrients Affecting the Absorption of Nutrients 152
2.1.1 Interactions Occurring in the Soil . . . . . . . . . . . . . 152
2.1.2 Absorption from Solution at the Root Surface . . . . . . . . 156
2.2 Interactions Between Nutrients Affecting the Utilization of Nutrients
Within the Plant 160
2.2.1 Distribution . . . . . . . . . . . . . . . . . . . . . . . 161
2.2.2 Function . . . . . . . . . . . . . . . . . . . . . . . . 164
2.3 Complex Interactions Between Nutrients Involving Several Processes 167
2.3.1 Calcium/Aluminium/Phosphate . . . . . . 167
2.3.2 Zinc/Phosphate . . . . . . . . . . . . . 169
3 Interactions Between Nutrients in Mixed Communities 170
4 Conclusion 173
References . . . . . . . . . . . . . . . . . . . 173

1.6 Import and Export of Mineral Nutrients in Plant Roots

U. LUnGE (With 10 Figures)
1 Introduction: The Dual Role of Roots in the Evolution of Higher Land
Plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
2 Relations Between Structure and Transport Functions Along the Length of
Roots . . . . . . . . . . . . . . . . . . . . . . . . . . . 182
2.1 The Phenomenon of Variations in Transport Functions Along the
Length of Roots . . . . . . . . . . . . . . . . 182
2.2 Structure-Function Relations in Various Root Zones 183
2.2.1 The Root Surface 183
2.2.2 The Cortex 189
2.2.3 The Endodermis . 190
2.2.4 The Stele . . . . 193
3 Variations of Physiological Activities Along the Length of Roots 199
3.1 Growth, Differentiation and Hormonal Gradients . . . 199
3.2 Bioelectrical Fields Along Roots . . . . . . . . . . 200
3.3 Differences in Ion Transport Mechanisms Along Roots 201
4 Root-Shoot Interactions and Circulation in the Whole Plant 202
4.1 Some Examples Illustrating General Aspects of Circulation 202
4.2 Nitrogen, Sulphur and Phosphorus 203
5 Conclusion 204
References 204

1.7 Cycling of Elements in the Biosphere

c.c. DELWICHE (With 5 Figures)
1 The Sources of Plant Constituents 212
1.1 Soil and Atmospheric Sources 212
1.2 The Weathering Process . . . 212
Contents Part A xv
2 The Nature of Cycles 214
2.1 The Hydrologic Cycle . 214
2.2 The Sedimentary Cycle 215
2.3 The Magmatic Cycle 217
2.4 The Geobiological Cycles 217
3 The Nitrogen Cycle 219
3.1 Overall Cycle Features 219
3.2 Nitrification . . . 221
3.3 Denitrification . . 222
3.4 Nitrogen Fixation 223
3.5 Human Influences 224
4 The Sulfur Cycle , . . 225
4.1 Comparison with the Nitrogen Cycle 225
4.2 Microbial Oxidation 227
4.3 Sulfate Reduction 227
4.4 Patterns of Sulfur Movement 228
4.5 Human Influences 228
5 The Phosphorus Cycle 229
5.1 Oxidation and Reduction 229
5.2 Movement and Transport in the Biosphere 230
5.3 Human Influences . . . . . . . . . . 231
6 Other Elements . . . . . . . . . . . . . 232
6.1 Biological Cycling . . . . . . . . . . 232
6.2 The Special Significance of Iron and Aluminum 232
6.3 Hydrogen Ion . . . . . . 233
6.4 Characteristics of Sediments . . . . . . . 234
6.5 Passive Cycling . . . . . . . . . . . . . 235
6.6 Possibilities of Deficiency . . . . . . . . 235
7 "Open" Versus "Closed" Agricultural Systems 236
References . . . . . . . . . . . . . . . . . 237

II. Inorganic Nitrogen Nutrition

11.1 Physiology, Biochemistry and Genetics of Dinitrogen Fixation

H. BOTHE, M.G. YATES, and F.C. CANNON (With 3 Figures)
1 The Nitrogen-Fixing Organisms and the Nitrogenase Reactions 241
1.1 Introduction . . . . . . . . . . . . . . 241
1.2 Nitrogen Fixation by Free-Living Organisms 244
1.3 Symbiotic Nitrogen Fixation. 245
1.4 Substrates of Nitrogenase . . . . . . 247
2 Biochemistry of Nitrogen Fixation 248
2.1 Introduction . . . . . . . . . . . 248
2.2 Nomenclature of Nitrogenase Proteins 249
2.3 Physicochemical Properties of Nitrogenase Proteins 249
2.4 Metal Clusters in Nitrogenase Proteins . . . . . 251
2.5 EPR and Mossbauer Spectroscopy on the MoFe Protein 252
2.6 The FeMo Cofactor and the Fe Protein . . . . . 253
2.7 Nitrogenase Proteins in Photosynthetic Organisms . . . 254
2.8 The Mechanism of Nitrogenase Activity . . . . . . . 254
2.8.1 The Roles of the Two Proteins . . . . . . . . . 255
2.8.2 Evidence for Interaction of MgATP and MgADP with the MoFe
Protein . . . . . . . . . . 257
2.8.3 The Nature of the Active Site(s) 257
2.8.4 Pathways of N 2 -Reduction 258
3 Electron Transport to Nitrogenase 259
3.1 Introduction 259
3.2 Ferredoxins . . . . . . . . 260
XVI Contents Part A

3.3 Flavodoxins .................... 260

3.4 Electron Donors . . . . . . . . . . . . . . . . . . . 261
4 Mechanisms to Protect Nitrogenase Against Damage by Oxygen 263
4.1 In Free-Living Organisms . . . . . . . . . . . 263
4.2 The Heterocysts of Blue-Green Algae . . . . . . 264
4.3 The Role of Leghaemoglobin in Legume Nodules 265
5 Regulation of Nitrogenase Activity and Biosynthesis 265
5.1 Regulation of Nitrogenase Biosynthesis . . . . . 265
5.2 Regulation of Nitrogenase Activity . . . . . . . 267
6 The Hydrogenase-Nitrogenase Relationship . . . . . 268
7 The Molecular and Genetic Characterization of Nitrogen Fixation Genes 271
7.1 Introduction 271
7.2 The nif Genes . . . . . . . . . . 272
7.3 nifGene Products . . . . . . . . 273
7.4 Cloning of K. pneumoniae nifGenes 274
7.5 A Physical Map of nifGenes 275
7.6 Interspecies Homology of Nitrogenase Genes 276
References . . . . . . . . . . . . . . . . . . 276

11.2 Dinitrogen-Fixing Symbioses with Legumes, Non-Legume

Angiosperms and Associative Symbioses
A. QrnsPEL (With 7 Figures)
1 Introduction . . . . . . . . . . . . . . . . . . . . . 286
2 Description of the Main Symbiotic Dinitrogen-Fixing Systems 287
2.1 Associative Symbioses 287
2.2 Symbioses with Cyanobacteria . . . 287
2.2.1 Distribution . . . . . . . . 287
2.2.2 Description and Development 287
2.2.3 N2 Fixation (C 2H 2 Reduction) 288
2.3 Root Nodules with Actinomycetes: Actinorhizas 288
2.3.1 Distribution . . . . . . 288
2.3.2 Description ....... . 289
2.3.3 Infection and Development . . 291
2.3.4 N 2 Fixation (C 2H 2 Reduction) 291
2.4 Leguminous Root Nodules with Rhizobium 291
2.4.1 Distribution . . . . . . . . . . 291
2.4.2 Description ......... . 291
2.4.3 Infection and Nodule Development 292
2.4.4 N2 Fixation (C 2H 2 Reduction) 295
2.5 Non-Leguminous Root Nodules with Rhizobium 295
3 The Dinitrogen-Fixing Micro-Symbionts: Isolates and Cultures 295
3.1 Introduction ............. . 295
3.2 Cyanobacteria . . . . . . . . . . . . . . 296
3.3 Frankia, the Endophyte from the Actinorhizas 297
3.3.1 Isolation and Cultivation 297
3.3.2 Specificity . . . . . 298
3.3.3 Nutrient Requirements 298
3.3.4 Metabolic Activities . 299
3.4 Rhizobium . . . . . . . 299
3.4.1 Isolation and Description 299
3.4.2 Taxonomy . . . . . . 300
3.4.3 Metabolism ..... 300
3.4.4 N2 Fixation (C 2H 2 Reduction) 302
3.4.5 Genetics . . . . . . . . . . 303
4 Symbiotic Relations . . . . . . . . . 304
4.1 Chemotaxis and Rhizosphere Accumulation 304
4.2 Binding of Rhizobium to Root Hairs 305
Contents Part A XVII

4.3 Root Hair Deformation and Infection-Thread Formation 307

4.4 Cell Wall Degrading Enzymes . . . . . . . . . 307
4.5 The Role of Plant Hormones in Nodule Formation 308
4.6 Miscellaneous Problems 310
5 The N 2 -Fixing System . . . . . . . . 311
5.1 Introduction . . . . . . . . . . 311
5.2 Bacteroids . . . . . . . . . . . 312
5.3 The Bacteroid-Containing Plant Cells 314
5.4 Nitrogenase . . . . . . . . . . . 315
5.5 NH3 Assimilation . . . . . . . . 315
5.6 Oxygen Regulation and Leghaemoglobin 317
5.7 Hydrogen Production and Hydrogen Uptake 318
6 Root Nodules as Part of the Whole Plant 319
7 Concluding Remarks . 323
References . . . . . . . . . . . . . . . . . . 323

11.3 Dinitrogen Fixation in Rhizosphere and Phyllosphere Associations

J. DOBEREINER (With 2 Figures)
1 Introduction . . . . . . . . . . . . . . 330
2 Cbaracterization of Rhizocoenoses 330
2.1 Sugar Cane - Beijerinckia . . . . . . . 331
2.2 Paspalum notatum - Azotobacter paspa/i 332
2.3 Azospirillum Rhizocoenoses . . . . 332
2.3.1 Taxonomy of Azospirillum spp. 333
2.3.2 Root Infection . . . . . . . 334
2.3.3 Host Plant Specificity 336
2.3.4 Physiology of Azospirillum . . 337
2.4 Associations with Other N 2 -Fixing Bacteria 340
3 Agronomic Aspects . . . 341
3.1 Plant Genotype Effects 341
3.2 Environmental Effects 342
3.3 Inoculation . . . . . 342
4 Phyllosphere Associations 343
4.1 Microorganisms in the Phyllosphere 344
4.2 Nitrogen Fixation in the Phyllosphere 344
5 General Conclusion 344
References . . . . . . . . . . . . . . . 345

11.4 Uptake and Reduction of Nitrate: Bacteria and Higher Plants

L. BEEVERS and R.H. HAGEMAN
1 Introduction . . . . . . . . . . . . . . . . . . . . . 351
2 Available Nitrogen Sources . . . . . . . . . . . . . . . 352
2.1 Species Differences in Ammonium and Nitrate Utilization 352
2.2 Influence of Ammonium or Nitrate on Cation Uptake 353
2.3 Nitrate Uptake . . . . . . . . . . . . . . . . . . 354
2.4 Influence of Ammonium on Nitrate Uptake and Utilization 355
3 Nitrate Reduction . . . . . . . . . . . . . 356
3.1 Bacteria . . . . . . . . . . . . . . . . . . . . . 356
3.2 Dissimilatory Nitrate Reductase . . . . . . . . . . . 356
3.3 Assimilatory Nitrate Reduction in Bacteria . . . . . . 358
3.4 Characterization of Nitrate Reductase from Higher Plants 358
4 Molybdenum in Nitrate Reduction 360
5 Nitrite Reduction . . . . . . 361
5.1 Assimilatory Bacteria 361
5.2 Dissimilatory Bacteria 361
5.3 Nitrite Reductase in Plants 361
XVIII Contents Part A

6 Location of Enzymes of Nitrate Assimilation in Higher Plants 363

7 Provision of Reductant for Nitrate Assimilation in Higher Plants 363
8 Regulation of Nitrate Reductase in Higher Plants 364
8.1 Substrate . . 364
8.2 Hormonal 365
8.3 Molybdenum 365
8.4 Ammonium . 366
8.5 Light 366
8.6 Genetic . . . 366
8.7 In Vivo Controls 367
9 Concluding Thoughts 368
References . . . . . . 369

II.S Uptake and Reduction of Nitrate: Algae and Fungi

W.R. ULLRICH (With 4 Figures)
1 Introduction . . . . . . . . . . . . . . 376
2 Nitrate and Nitrite Reduction in Algae 377
2.1 Nitrate Reductase of Eucaryotic Algae . 377
2.2 Nitrate Reductase in Blue-Green Algae 380
2.3 Nitrite Reductase in Algae . . . . . . 380
2.4 Location of Nitrate and Nitrite Reduction in Algal Cells 381
2.5 Stoichiometry Between Nitrate Reduction and O 2 Exchange 381
3 Nitrate Uptake in Algae 382
3.1 General Remarks 382
3.2 Substrate Affinity . . . . . . 383
3.3 Light Dependence ..... 384
3.4 pH-Dependence . . . . . . 385
3.5 Dependence on Carbon Sources 385
3.6 Inhibition by Anions 386
3.7 Inhibition by Ammonia and Amino Compounds 387
3.8 Effect of Metabolic Inhibitors and Uncouplers 387
3.9 Stoichiometry Between the Uptake of Nitrate and that of Other Ions 388
3.10 Transport Mechanism . . . . . . . . . . . . . . . . 388
4 Nitrite Uptake in Algae . . . . . . . . . . . . . . . . . 389
5 General Remarks on Regulation of Nitrate and Nitrite Uptake 390
6 Uptake and Reduction of Nitrate and Nitrite in Fungi 391
References . . . . . . . . . . . . . . . . . . . . . . . . 393

III. Metabolism of Sulfur and Phosphorus

m.l Reduction and Other Metabolic Reactions of Sulfate

J.A. SCIDFF (With 6 Figures)
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 401
2 The Place of Sulfate Reduction in the Sulfur Cycle . . . . . . . . 402
3 Phylogenetic Distribution of Reactions Involving Sulfate Transfer and
Reduction . . . . . . . . . . . . . . . . . . . . . 403
4 Sulfate Uptake, Activation and Transfer . . . . . . . 404
5 Sulfate Reduction . . . . . . . . . . . . . . . . . 406
5.1 Detailed Reactions of the Two Assimilatory Pathways 408
5.1.1 The APS Pathway . . . . . . . . . . . . . 408
5.1.2 The PAPS Pathway . . . . . . . . . . . . . 410
5.2 Location of Sulfate Reduction in Tissues and Organs of Multicellular
Plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . 413
Contents Part A XIX

6 Speculations on the Origin and Evolution of Pathways of Sulfate

Reduction 413
References 416

m.2 Physiology and Metabolism of Phosphate and Its Compounds

R.L. BIELESKI and LB. FERGUSON (With 4 Figures)
1 Introduction . . . . . . . . . . . . . . . . . . . . 422
2 Uptake and Transport of Phosphate . . . . . . . . . 424
3 Efflux of Phosphate, and Aspects of Phosphate Deficiency 428
4 Phosphorus Compartments and Pools . . . . . . 431
5 The Form of Phosphorus in the Cell . . . . . . 433
6 Synthesis and Turnover of Phosphorus Compounds 440
7 Dynamics of Phosphate Use in the Plant 443
8 Conclusions 445
References . . . . . . . . . . . . . . 445

Author- and Subject Index (see Part B)

Contents Part B

IV. General Function of Inorganic Nutrients in Growth

and Metabolism
IV.l Genetic Basis ofInorganic Plant Nutrition
G.c. GERWFF and W.H. GABELMAN (With 2 Figures) 453
IV.2 Mineral Nutrition and Growth
J. MOORBY and R.T. BESFORD (With 24 Figures) . 481
IV.3 Proteins, Enzymes and Inorganic Ions
R.G. WYN JONES and A. POLLARD (With 5 Figures) 528
IVA The Enzymological Function of Heavy Metals and Their Role in Electron
Transfer Processes of Plants
G. SANDMANN and P. BOOER (With 4 Figures) . . . . . . . . . . . . . 563

V. Special Functions of Some Elements

V.l Calcium Transport and Function
D. MARME (With 2 Figures) . . 599
V.2 Boron in Plant Metabolism
W.M. DUGGER (With 4 Figures) 626
V.3 Sodium Versus Potassium: Substitution and Compartmentation
T.J. FLOWERS and A. LXUCHLI (With 6 Figures) . . . . . . . 651
VA Silica Metabolism
D. WERNER and R. ROTH (With 3 Figures) 682
V.5 Involvement of Unusual Elements in Plant Growth and Nutrition
E.G. BOLLARD (With 4 Figures) . . . . . . . . . . . .. . . . 695

VI. Synthesis and Outlook

R.L. BmLESKI and A. LXUCHLI (With 1 Figure) 745
Author Index 757
Subject Index 829
List of Contributors Part A and B

C.J. AsHER D. BOUMA

Dept. of Agriculture CSIRO
University of Queensland Division of Plant Industry
St. Lucia P.O. Box 1600
Queensland 4067/Australia Canberra, ACT, 2601/Australia

L.BEEVERS G.D. BOWEN

770 Van Vleet Oval CSIRO Division of Soils
Dept. of Botany and Microbiology Private Bag 2
University of Oklahoma Post Office
Norman, OK 73019(USA Glen Osmond
South Australia 5064/Australia
R. T. BESFORD F.C. CANNON
Glasshouse Crops Research Institute BioTechnica International, Inc.
Worthing Road 85 Bolten Street
Littlehampton Cambridge, MA 02140(USA
West Sussex BN16 3PU/
United Kingdom C.C. DELWICHE
University of California
R.L. BIELESKI Dept. of Land, Air and Water
Dept. of Scientific and Resources
Industrial Research Hoagland Hall
Division of Horticulture and Processing Davis, CA 95616(USA
Private Bag
Auckland/New Zealand J. DOBEREINER
EMBRAPA PNPBS, Km 47
Seropedica
P. BOGER
23460 Rio de Janeiro/Brazil
Lehrstuhl fiir Physiologie und
Biochemie der Pflanzen W.M. DUGGER
Universitat Konstanz Dept. of Botany and Plant
Postfach 5560
Sciences
D-7750 Konstanz/FRG
University of California
Riverside, CA 92521(USA
E.G. BOLLARD
Dept. of Scientific and D.G. EDWARDS
Industrial Research Dept. of Agriculture
Division of Horticulture and University of Queensland
Processing St. Lucia
Private Bag Queensland 4067/Australia
AucklandfNew Zealand
E. EpSTEIN
H. BOTHE University of California
Botanisches Institut Dept. of Land, Air and Water
Universitat Koln Resources
Gyrhofstr.15 Hoagland Hall
D-5000 Koln 41/FRG Davis, CA 95616(USA
XXII List of Contributors Part A and B

I.B. FERGUSON H. MARSCHNER

Division of Horticulture and Institut fUr Pflanzenerniihrung
Processing Universitiit Hohenheim
DSIR Postfach 700562
Private Bag D-7000 Stuttgart 70/FRG
AucldandjNew Zealand
J. MooRBY
T.J. FLOWERS Agricultural Research Council
School of Biological Sciences 160 Great Portland St.
University of Sussex London WIN 6DT/United Kingdom
Falmer
Brighton BNl 9QG/United Kingdom M.G. PITMAN
School of Biological Sciences, A12
R.C. FOSTER University of Sydney
CSIRO Division of Soils NSW 2006/Australia
Private Bag 2
Post Office A. POLLARD
Glen Osmond Dept. of Biochemistry and
South Australia 5064/Australia Soil Sciences
University College of North Wales
W.H. GABELMAN Bangor, Gwynedd
Department of Horticulture Wales/United Kingdom
University of Wisconsin
1575 Linden Dr. A. QUISPEL
Madison, WI 53706/USA Botanical Laboratory
Dept. of Plant Molecular Biology
G.C. GERLOFF University of Leiden
Dept. of Botany N onnensteeg 3
Birge Hall 2311 VL Leiden/The Netherlands
University of Wisconsin-Madison
Madison, WI 53706/USA A.D. ROBSON
Dept. of Soil Science and
Plant Nutrition
R.H. HAGEMAN Institute of Agriculture
Dept. of Agronomy University of Western Australia
University of Illinois Nedlands
1102 S Goodwin Ave. Western Australia 6009/Australia
Urbana, IL 61801/USA
R. ROTH
A. LXUCHLI Fachbereich Botanik
University of California der Phillipps-Universitiit Marburg
Dept. of Land, Air and Water Lahnberge
Resources D-3550 Marburg/Lahn/FRG
Hoagland Hall
Davis, CA 95616/USA A.D. ROVIRA
CSIRO Division of Soils
U. LUTTGE Private Bag 2
Institut fiir Botanik Post Office
Fachbereich Biologie (10) Glen Osmond
Technische Hochschule Darmstadt South Australia 5064/Australia
SchnittspahnstraBe 3
D-6100 Darmstadt/FRG G. SANDMANN
Lehrstuhl flir Physiologie und
D. MARME Biochemie der Pflanzen
Institut flir Biologie III Universitiit Konstanz
SchiinzlestraBe 1 Postfach 5560
D-7800 Freiburg/FRG D-7750 Konstanz/FRG
List of Contributors Part A and B XXIII

J.A. SCHIFF R.G. WYN JONES

Institute for Photobiology Dept. of Biochemistry and Soil
of Cells and Organelles Science
Brandeis University University College of North Wales
Waltham, MA 02254/USA Bangor, Gwynedd
Wales/United Kingdom
W.R. ULLRICH
Institut fUr Botanik M.G. YATES
Technische Hochschule Darmstadt Agricultural Research Council
SchnittspahnstraBe 3 Unit of Nitrogen Fixation
D-6100 Darmstadt/FRG University of Sussex
Falmer
D. WERNER Brighton BNl 9QGfUnited Kingdom
Fachbereich Botanik
der Philipps-U niversitiit Marburg
Lahnberge
D-3550 Marburg/Lahn/FRG
Introduction
A. LXUCHLI and R.L. BmLESKI

Initial plans for this volume were made in 1974 when manuscripts for Vol. 2A
and B on Transport in: Cells, Tissues and Organs were in preparation. In discus-
sions involving ULRICH LUnGE, MICHAEL PITMAN and A.L. the idea emerged
that a volume was needed in the Encyclopedia of Plant Physiology, New Series
to cover the field of Inorganic Plant Nutrition.
To treat the whole field of plant nutrition in one volume has meant that
we have been able to present only a selection of topics. It has not been possible
to have a full review of the literature pertaining to each nutrient element. The
organization of this volume therefore evolved from the concept that the broad
subject should be arranged functionally rather than by elements. Furthermore,
this volume complements Vol. 2, for many central functions of inorganic nu-
trients are intrinsically connected to their transport in the plant. Thus our au-
thors were asked not to attempt comprehensive reviews and exhaustive literature
surveys, but rather to integrate the wealth of information on inorganic plant
nutrition and incorporate personal viewpoints. .
This volume consists of two parts and is organized in a foreword, five main
sections, and a final chapter on synthesis and outlook. Section I aims at integra-
ting modern aspects of inorganic plant nutrition and incorporates various fea-
tures of nutrient transport at the soil-root interface, in the plant, and in the
biosphere. Section II is devoted to inorganic nitrogen nutrition and treats sym-
biotic dinitrogen fixation (including associative symbioses) as well as uptake
and reduction of nitrate. The physiology and metabolism of sulfate and phos-
phate (Sect. III) conclude Part A of the volume. In Part B, Section IV treats
the general function of inorganic nutrients in growth and metabolism and in-
cludes the genetic basis of inorganic plant nutrition. Finally, Section V highlights
novel ideas on the functions of some particular elements - from the nutrient
elements calcium and boron and sodium-potassium relationships to silica and
unusual elements.
We would like to thank all those who contributed to this volume. Our
primary thanks go to Professor PIRSON, one of the Series Editors, whose advice
on organization of the volume and continued interest in its progress have been
extremely valuable. We also owe thanks to Professor PIRSON for understanding
that it was impossible to keep the whole book manuscript to the expected
size of a regular volume in this series. Otherwise, the treatment of some topics
would have been very sketchy. We also thank the publishers for their support
in the production of this volume.
We thank the colleagues whose discussions in the early stages helped us
to establish a group of topics that, we hope, reveal something of the ferment
and vitality of present-day plant nutrition research. Thus we particularly thank
2 Introduction

E. EpSTEIN, A.R. FERGUSON, LB. FERGUSON, H. GREENWAY, U. LUTTGE and

M.G. PITMAN. Special thanks are due to A.R. FERGUSON for his careful and
thorough preparation of the subject index. Above all we thank the authors
of this volume for their patience and their understanding that the Editors had
to make demands for reorganization and modification of their chapters.
I. General Chapters
of Inorganic Plant Nutrition
1.1 General Introduction
to the Mineral Nutrition of Plants
H. MARSCHNER

1 Introduction and Historical Resume

1.1 Essential Mineral Elements - Plant Nutrients

The beneficial effect of adding mineral elements (e.g. plant ash, lime) to soils
to improve plant growth has been known in agriculture for more than 2000
years. Nevertheless, even 150 years ago it was still a matter of scientific contro-
versy as to whether mineral elements function as nutrients for plant growth.
It was mainly to the credit of JUSTUS VON LmBIG (1803-1873) that the scattered
information concerning the importance of mineral elements for plant growth
was collected and summarized, and that mineral nutrition of plants was estab-
lished as a scientific discipline. This achievement led to a rapid increase in
the use of mineral fertilizers. In Europe in particular large amounts of potash,
superphosphate and later inorganic nitrogen were used in agriculture and horti-
culture to improve plant growth.
LmBIG'S conclusion as to the essentiality of the mineral elements, nitrogen,
sulfur, phosphorus, potassium, calcium, magnesium, silicon, sodium and iron
was based on observations and speculations rather than on precise experimenta-
tion. This shaky foundation on which the "mineral element theory" was laid
was one of the reasons for the onset of a large number of studies at the end
of last century. Both sand and water culture experiments were carried out on
higher plants to establish essentiality of the mineral elements for plant growth
and their role in plant metabolism. Progress in this research was closely related
to developments in analytical chemistry (purification of chemicals, methods
for determination). This relationship is well reflected in the time scale of the
discovery of the essentiality of the micronutrients (Table 1).

Table 1. Discovery of the essentiality of micronutrients for

higher plants

Element Year Discovered by

Iron 1860 J. SACHS

Manganese 1922 J.S. McHAGUE
Boron 1923 K. WARINGTON
Zinc 1926 A.L. SOMMER and C.B. LIPMAN
Copper 1931 C.B. LIPMAN and McKINNEY
Molybdenum 1938 D.1. ARNON and P.R. STOUT
Chlorine 1954 T.e. BROYER et al.
6 H. MARSCHNER:

As proposed by ARNON (1950), it is generally agreed that the term essential

mineral element (=mineral plant nutrient) should be restricted to those elements
which fulfil the following criteria: (a) that there is a positive requirement of
the element for normal growth or reproduction, (b) that it is not possible to
replace the function of this element by other elements, (c) that it has a direct
or indirect action in plant metabolism. The term does not include those elements
which, for example, compensate the toxic effects of other elements. Thus, a
mineral element which only stimulates growth but does not satisfy the above
criteria cannot be defined as mineral plant nutrient. It can merely be described
as a "beneficial" element (see below).
Six macronutrients are now firmly established for all higher plants. However,
the possibility still exists that with improving analytical techniques new micronu-
trients may be revealed so that the present list of 13 mineral nutrients for
higher plants (Table 2) may yet be extended.
It is still difficult to generalize concerning the essentiality of mineral elements
for plant growth. This is particularly obvious when higher and lower plants
are compared. In some bacteria and fungi, for example, even calcium, boron
and chlorine are probably not essential. On the other hand, for higher plants
whether also sodium and silicon are essential or just "beneficial elements"
appears to be genotypically dependent (see below).

1.2 Function of Essential Mineral Elements

For mineral elements that are constituents of organic compounds which act
in plants as enzymes, co-enzymes, membrane constituents etc. at least some
of their function has been well defined. The reader is referred for detailed infor-
mation on nitrogen to Chapter II, on sulfur and phosphorus to Chapters 111.1
and 111.2, on potassium and calcium to Chapter V.l and on boron to Chapter
V.2, all this Volume!. A chapter dealing with micronutrients in general is to
be found in Chapter IV.4 of the first edition of Encyclopedia of Plant Physiolo-
gy, Vol. IV. Detailed information on plant nutrients may be also found in recent
text books such as EpSTEIN (1972), GAUCH (1972), HEWITT and SMITH (1975),
BERGMANN and NEUBERT (1976), MENGEL and KIRKBY (1978), MENGEL (1979).
For this general introduction it is adequate to sum up one or two typical
well-established functions of each mineral nutrient (Table 2). Table 2 also refers
to relevant chapters of this Volume. The mechanisms of these effects on enzyme
reactions are of differing nature: (a) Valency change as an electron transferring
agent (e.g. Fe), (b) Attachment of the substrate to the enzyme (e.g. Mg) or
(c) Change in molecular configuration of enzyme or substrate or both (e.g. K).
Most results on enzyme activation (Table 2 last column) are' based on in
vitro experiments using relatively high concentrations of inorganic ions. Some
doubt is left therefore as to whether in vivo ion concentrations are sufficiently
high in the cell compartments or at the sites of enzyme reaction (e.g. membranes)
to act as activators as in the in vitro experiments. This is of particular relevance
to the micronutrients. In addition, several morphological, physiological and
1 Further chapters cited in this chapter refer to the present volume, unless specified
1.1 General Introduction to the Mineral Nutrition of Plants 7

Table 2. The amounts and examples of the functions of mineral nutrients in higher
plants

Element J.lmol Constituent of essential Activator ( + ) or inhibitor

per ga organic compounds ( - ) of enzymes or enzyme
(examples) reaction

N 1000 Proteins, nucleic acid Nitrate reductase

NO; (+), NHt (-)
see Chap. 11.4
S 30 . Protein, sulfolipids (see Chaps. IV.3 and III)
P 60 Nucleic acids, co-enzymes ADPG-pyrophosphorylase
(-)
starch phosphorylase
(see Chap. 1II.2)
K 250 Unknown Starch synthase
(see Chaps. IV.3 and V.3)
Ca 125 Pectates Amylase (+), (see below and
Chap. V.l)
Mg 80 Chlorophyll Phosphorylation ( + )
RubP-carboxylase ( + )
Fe 2.0 Ferredoxin, cytochrome (see Chap. IV.4)
Mn 1.0 Unknown "Hill reaction" ( + )
Amino peptidases ( + )
Zn 0.3 Carbonic anhydrase Hexokinase ( + )
alkaline phosphatase Ribonuclease ( - )
(see Chap. IV.4)
Cu 0.1 Ascorbic acid oxidase (see Chap. IV.4)
plastocyanine
B 2.0 Unknown (see Chap. V.2)
CI 3.0 Unknown (see Chap. IV.4)
Mo 0.001 Nitrate reductase (see Chap. 11.4)

a Average concentration in dry matter sufficient for adequate growth (see for example
BERGMANN and NEUBERT 1976)

biochemical changes in higher plants caused by deficiencies of certain mineral

nutrients, micronutrients in particular, either cannot or can only indirectly be
causally related to the known functions of the essential mineral elements. Within
the last 20 years research activities have thus centered more on integrated meta-
bolic processes influenced by essential mineral elements rather than on purely
biochemical in vitro studies (see below).

1.3 Beneficial Mineral Elements

Mineral elements which either stimulate growth without being essential (see
definition above) or which are essential only for certain plant species or families
8 H. MARSCHNER:

are usually defined as beneficial elements. This applies particularly to sodium,

silicon and cobalt. The difficulties in this definition of a beneficial mineral
element are readily demonstrated in the following examples.

Sodium. The use of the term "natrophilic" and "natrophobic" plant species
indicates the wide range of response to sodium in the plant kingdom
(MARSCHNER 1971, JENNINGS 1976). Sodium is essential for certain halophytes
(BROWNELL and JACKMAN 1966), and is considered also as a micronutrient
at least for C4 plant species (BROWNELL 1979). In addition it has a beneficial
effect on growth of the so-called halophytic species such as Beta vulgaris, particu-
larly under conditions of low potassium supply (EL-SHEIKH et al. 1967,
MARSCHNER and POSSINGHAM 1975), but is either without effect or is even
harmful on growth of natrophobic species such as Phaseolus vulgaris (HAWKER
et al. 1974). This differentiation between families and species in response to
sodium, which is considered in more detail in Chapters IV.3 and V.3, opens
interesting possibilities into basic research on the compartmentation and func-
tion of mineral elements which are not constituents of organic compounds (e.g.
potassium). It is also of interest in applied research, since sodium is abundant
in most natural ecosystems and in particular in the irrigation water of arid
and semi-arid climates. This must be considered in relation to the fairly close
positive correlation between the beneficial effect of sodium on growth and the
"salt tolerance" of certain plant species. It is urgently necessary to obtain more
information on the role of sodium in plants in general and on its ability to
replace potassium in particular. This would allow a more straightforward strate-
gy in selection and breeding programmes for crop plants with high" salt toler-
ance" in order to adapt these plants to saline soils, as has been sucessfully
demonstrated by EpSTEIN et al. (1980).

Silicon. Silicon is an essential mineral element for certain lower plants (see
Chap. V 4). It is also well established as a beneficial mineral element for higher
plants, the monocotyledons in particular. In higher plants uptake of silicon
is closely related to the transpiration rate of the plants (JONES and HANDRECK
1965) as well as to the transpiration rate of individual organs (HunoN and
NORRISH 1974). It is deposited mainly in the cell walls of the epidermal layers,
and causes an increase in mechanical resistance. This can increase resistance
to certain fungal infections (YOLK 1958, FIDANOVSKI 1969). This deposition
of silicon also increases the rigidity of leaf blades and thus alters the leaf angle,
thereby increasing the efficiency of light utilization for photosynthesis, particu-
larly in dense plant populations (yOSHIDA et al. 1969, YUAN and CHANG 1978).
Silicon is particularly effective in counteracting manganese toxicity (yLAMIS
and WILLIAMS 1967). Growth stimulation by silicon is therefore often limited
to conditions of excessive manganese supply (HORST and MARSCHNER 1978a,
VORM and DIEST 1979). This effect of silicon seems to be either primarily the
result of restricted manganese uptake as in rice plants (YUAN and CHANG 1978,
VORM and DmsT 1979) or, as in other species such as barley (WILLIAMS and
VLAMIS 1957) and bean (HORST AND MARSCHNER 1978a), caused by more uni-
form distribution of manganese within the leaf tissue. These effects of silicon
1.1 General Introduction to the Mineral Nutrition of Plants 9

fully justify its classification as a beneficial mineral element. At least in Equi-

setum arvense, however, silicon is essential for growth (CHEN and LEWIN 1969).
There is also good evidence for the essentiality of silicon for rice (YUAN and
CHANG 1978). The surprising results of MIYAKE and TAKAHAsm (1978) showing
the essentiality of silicon also for tomato plants (Lycopersicon esculentum Mill)
needs to be critically re-examined. Unfortunately, these authors did not mention
the variety used for these studies, information which is essential for all studies
on mineral metabolism of higher plants in general (see Vol. 2/B.9, this Series)
and on mineral elements such as silicon in particular. The situation with silicon
seems to be similar to that of sodium, where classification as either an essential
or beneficial mineral element depends upon the genotypes. For further details
on silicon see Chapter VA.

Other Mineral Elements. The essentiality of other mineral elements such as

cobalt is well documented for certain higher and lower plants. For cobalt this
essentiality is coupled with the mode of nitrogen supply, as this metal is a
constituent of the vitamin B12 and thus essential for all higher and lower plants
which rely on N 2 -flXation for their nitrogen supply. For the same plant species
supplied with bound nitrogen either in inorganic or organic form, cobalt does
not have the characteristics of a beneficial mineral element. These special cases
of essentiality of cobalt and other mineral elements are discussed in detail in
Chapter V.5.

1.4 Recent Developments

Despite the early recognition of the essentiality of certain mineral elements

for plant metabolism and growth, the interest of plant physiologists and plant
biochemists in detailed studies on mineral nutrition was relatively small until
about 30 years ago. There were, of course, notable exceptions, including HOAG-
LAND and his school in California, but in general much greater attention was
paid to photosynthesis, respiration, or synthetic pathways of carbohydrates,
lipids and other organic components. After 1950, however, rapid development
started in research on mineral nutrition. This was mainly by the use of radioiso-
topes in short-term studies with excised roots on the ion uptake mechanism
and the approach of applying enzyme kinetics and the carrier concept to mem-
brane transport processes of inorganic ions in plant cells and tissues (EpSTEIN
and HAGEN 1952). This development was associated with a deeper understanding
of membrane transport of inorganic ions in plant cells and tissues (see Vol. 2 AI
B, this Series). During this period it was also realized that mineral elements
are not only excellent tools for studies of the mechanism or energetics of mem-
brane transport processes, but that essential mineral elements are also causally
involved in regulating biochemical and metabolic processes of fundamental im-
portance in the various compartments of living cells. Some of the more recent
important developments in studies relating to the functions of some essential
mineral elements in integrated systems such as plant cells or plant tissues are
presented below.
10 H. MARSCHNER:

Table 3. Effect of calcium chloride (10- 2 M) and benzyladenine (10- 7 MBA) on com
leaf segments kept for 4 days in darkness. (After POOVAIAH and LEOPOLD 1973 a, PoovAlAR
1979)

Treatment Chlorophyll Protein Hydraulic permeability

OD665 % initial half time in min for equilibrium
Control 0.21 58 0.8
Ca 0.32 76 2.5
BA 0.38 78 3.1
Ca+BA 0.66 90 8.0

1.4.1 Calcium
Although not a constituent of cell membranes, calcium ions are of fundamental
importance for membrane permeability and transport and the maintenance of
cell integrity (see Chap. V.1). The regulating effect of calcium at the plasmalem-
ma on influx and efflux of other ions and selectivity in ion uptake has been
one of the major research topics in ion uptake studies for more than 20 years
(see Chaps. I.6; I.7 and V.1). It is now accepted that ion uptake studies, for
example, with plant roots in the absence of externally supplied Ca2 +, are of
rather limited value.
The role of Ca 2 + for membrane stability, and thus compartmentation of
a single cell or a plant tissue, has consequences for several processes other
than ion transport. It affects the efflux rate of sugars from plant roots (SHAY
and HALE 1973) or leaf segments (HAWKER et al. 1974) into the external solution,
as well as the flux rate of endogenous respiratory substrates from the vacuole
to the respiratory enzymes in the cytoplasm (BANGERTH et al. 1972). An increase
in the calcium content of a tissue therefore decreases the respiration rate (FAUST
and KLEIN 1974). Typical features of senescence include an increase in the rate
of respiration together with enhanced permeability and thus a decline in com-
partmentation. Calcium is thus directly involved in the regulation of senescence
in addition to the role played by phytohormones such as ethylene or cytokinins
(Table 3).
These results show that senescence of the leaf segments was retarded by
calcium. A similar influence was found with cytokinin and the effect of both
substances was additive. Furthermore, in Phaseolus vulgaris the abscission of
leaf blades, caused by senescence of the pulvinar tissue, can be dramatically
delayed or even prevented by high Ca2+ concentrations (POOVAIAH and LEOPOLD
1973b).
The widespread occurrence of calcium-related physiological disorders, par-
ticularly in fruits (see below), has also initiated intensive research on the role
of Ca 2 + in fruit development and fruit ripening (for a comprehensive review
see BANGERTH 1979). Fruit ripening as a special case of senescence is correlated
with increased ethylene production. This is stimulated in calcium-deficient tissue
(FAUST and SHEAR 1969) and can be depressed by calcium treatment of the
fruit (LOUGHEED et al. 1979). According to MATTOO and LIEBERMAN (1977),
the causal connection between calcium-regulated membrane permeability and
1.1 General Introduction to the Mineral Nutrition of Plants 11

Table 4. Calcium content (~mol g-l dry weight) in non ripening

(rin) and normal Rutgers tomato pericarp tissues of different
stages of fruit development. (After POOVAIAH 1979)

Days after Soluble Ca Bound Ca

anthesis
rin Rutgers rin Rutgers

40 7.5 8.7 13.3 14.1

50 10.3 15.1 16.7 6.2
60 12.3 15.6 33.9 7.3

ethylene production and thus senescence of a tissue can be seen in the compart-
mentation of the enzyme system for ethylene synthesis in a cell wall-membrane
complex, i.e. outside the cytoplasm. A fall in calcium content of a fruit due
to "dilution" by growth (see below) or by a decrease in the concentration
of physiologically active calcium (for example by precipitation as oxalate, chela-
tion or compartmentation in the vacuole) can be considered as an essential
step in the regulation of phytohormone-dependent fruit ripening. Fruit ripening
certainly requires removal of part of the calcium from the middle lamella (bound
calcium). This causal involvement of calcium is well reflected in comparative
studies with a ripening and a non-ripening mutant (rin) of tomato (Table 4).
In the rin mutant the calcium content and particularly the bound fraction
continuously increased after anthesis whereas in Rutgers variety the total calci-
um content remained constant and the bound calcium declined. This decline
was correlated with an increase in polygalacturonase activity, the enzyme which
is responsible for decomposing pectates of the middle lamella and finally" soft-
ening" this tissue.
Calcium in the middle lamella is not only important for the structure of
the pectins itself, but also for the resistance of the tissue to fungal infections.
There is a strict negative correlation between the calcium content of the tissue
and the infection rate of lettuce leaves with Botrytis cinera Pers. (KRAUSS 1971)
or apple fruits with Gloeosporium spp. (SHARPLES and JOHNSON 1977). The
causality of this negative correlation is the result of either the inability of poly-
galacturonase to act on calcium pectates (BATEMAN and LUMSDEN 1965) or
the direct inhibition of this enzyme by calcium ions (CORDEN 1965). Plant cell
walls containing a high proportion of calcium pectates should therefore be
expected to show increased resistance to fungal infections. These various func-
tions of calcium, particularly in integrated systems, considered together with
the increase in occurrence of calcium-related physiological disorders i)1 crop
plants are reflected in the pronounced interest of crop physiologists in calcium
nutrition of plants (Sect. 3).

1.4.2 Potassium
With a few exceptions (halophytes), potassium is the most important inorganic
solute in plants. This explains at least in part the particular role of potassium
in osmoregulation and in water relations in plants as well as in enzyme activation
12 H. MARSCHNER:

Table 5. Effects of potassium supply on carbon dioxide assimilation in Medicago sativa.

(PEOPLES and KOCH 1979)

Substrate CO 2 diffusion Efficiency of photosystem Ribulosecarbox.

Resistance (s em -1) and CO 2 ftxation IlmolC0 2 mg- 1
(Ilffiol acceptor red mg- 1 chI.) protein h- 1

mMK+ Stomata Mesophyll System I System II

0.0 9.3 8.4 238 282 1.79

0.6 6.8 7.0 262 294 4.51
4.8 5.8 3.0 241 285 6.06

(Chaps. IV.3; V.3; see also Vol. II A/B, this Series, and a review by LXUCHLI
and PFLUGER 1978). The high mobility of potassium in transport through mem-
branes enables it also to act as counterion for anions both in short- and long-
distance transport (Chaps. 1.6 and 1.7) and in ATPase-stimulated proton effiux,
for example at the plasmalemma of root cells (Chap. 1.7). This proton-coupled
potassium transport has consequences for both the pH and osmotic pressure
in the cell and its compartments. The role- of potassium in the water balance
of higher plants can now be directly related to its function in regulating osmotic
pressure of the guard cells and thus stomatal movement (HUMBLE and RASCHKE
1971; HSIAO 1976, RAsCHKE 1979).
Potassium-mediated changes in turgor and thus induction of movements
are also responsible for the phytochrome-controlled movements, as for example
in the hypocotyl hook of mung bean (BROWNLEE and KENDRICK 1979) or the
nyctinasty in Albizzia (SATIER and GALSTON 1971) and the seismonastic reac-
tions in Mimosa (CAMPBELL et al. 1979).
The varied functions of potassium enable this element to influence complex
processes such as photosynthesis of higher plants at various stages. With potassi-
um deficiency even the CO 2 diffusion through the stomata can be limited, and
this is reflected in an increase in stomatal resistance (COOPER et al. 1967). The
increase in mesophyll resistance when potassium supply is limited appears how-
ever, to be primarily related to inhibited protein synthesis and in particular
to RubP-carboxylase synthesis (Table 5).
Incubating potassium-deficient leaves in potassium salt solution rapidly in-
creased RubP-carboxylase activity, an increase which could be prevented by
actinomycin (PEOPLES and KOCH 1979). This close correlation between potassi-
um supply and synthesis of an enzyme in the chloroplasts has also been demon-
strated with nitrate reductase in spinach chloroplasts (PFLUGER and WIEDEMANN
1977).
The stimulation of bound starch synthase activity from chloroplasts by po-
tassium (HAWKER et al. 1974) on the one hand and the enhanced export of
sucrose from leaf cells as "source" (DOMAN and GEIGER 1979) on the other
are further examples of the action of potassium in an integrated system. This
also applies to the effect of potassium on the N 2-fixation rate of legumes
(MENGEL et al. 1974) and growth rate of fruits and storage organs, i.e. yield
formation (see Sect. 4).
1.1 General Introduction to the Mineral Nutrition of Plants 13

The mineral nutritional status of plants can strongly affect the "quality"
of crop plants and its organs not only for human and animal nutrition, but
also as substrate for parasitic attack. When potassium supply is inadequate,
for example, the synthesis of highly polymerized compounds (starch, proteins)
is inhibited and the resulting plant composition is of lowered "quality" from
the agricultural and horticultural view point. However, the corresponding in-
crease in low molecular weight compounds (sugars, amino acids) in these plants
increases its "quality" for various parasites, fungi, bacteria or insects. As an
example, the close negative correlation between infection of soybean seeds with
Diaporthe sojae and the potassium level (CRITTENDEN and SVEC 1974) may be
cited. A comprehensive and more general review on this subject has been pub-
lished elsewhere (PERRENOUD 1977) with impressive examples showing the close
relationships between mineral nutrition of plants and plant diseases and plant
protection.

1.4.3 Phosphorus
Phosphorus is essential not only as a constituent of various organic compounds
(see Chap. III.2) but also inorganic phosphorus (PJ exerts important regulatory
functions in energy transfer. Another instructive example of its regulatory func-
tion has been described recently in chloroplasts where a concentration of approx-
imately 10 mM in the stroma almost completely inhibits starch synthesis (HELDT
et al. 1977). The control seems to be exerted mainly via the allosteric regulation
of ADPG-pyrophosphorylase by Pi' From these results obtained with chloro-
plasts one might speculate that in starch-storing seeds such as cereal grains,
phytin (Chap. II.2) might not only have the function as storage for phosphate
and inorganic cations (LOTI and BUTTROSE 1978) but also take part in the
regulation of starch synthesis. This aspect along with other, more indirect effects
of phosphorus, are discussed later (Sect. 3).

1.4.4 Nitrogen
Recent developments in research on inorganic nitrogen nutrition are character-
ized by particular attention to the various aspects of dinitrogen fixation (Chap.
II. 1, 2, 3), nitrate reduction and its regulation (Chap. 11.4) and the consequences
of intensity and sites (plant organs) of nitrate reduction for various other aspects
of mineral nutrition such as cation/anion balance (Chap. 1.6) or the interaction
between plant roots and their substrate (see next section). Ammonium as com-
pared to nitrate also seems to exert some distinct regulatory functions, not
only in induction of flower formation (see Sect. 4) but also, for example, in
chloroplasts by enhancing the direction of photosynthetic carbon flow towards
the Krebs cycle (MOHAMED and GUANAM 1979).
The main effect of both ammonium' and nitrate nitrogen on plant growth
and development (e.g. flowering and senescence), however, seems to be causally
related to direct interactions between nitrogen nutrition and phytohormones.
The close correlation between nitrogen supply (nitrate) and the formation of
cytokinins in the roots and their export to the shoot has been clearly demon-
14 H. MARSCHNER:

Table 6. Effect of nitrate supply to potato plants on exudation rate

and cytokinin export from the roots over a 24-h period. (SATTELMA-
CHER and MARSCHNER 1978)

Nitrogen Days Exudation Exudate

supply rate
(m124 h- 1) Concentration Amount of
of cytokinins cytokinins
(ngml-l) (ng 24 h- 1 )

+N 0 33 6 196
3 30 14 420
6 33 17 561
-N 3 13 2 26
6 17 1 17
+N a 9 22 6 132

a Renewed nitrate supply after 6 days - N

Table 7. Effect of nitrate supply on ABA content in different parts

of potato plants. (KRAuss 1978)
ng ABA g - 1 fresh weight
and ml- 1 respectively

Shoot Root Exudate

Beginning of experiment 12.6 0.3 0.1

After 6 days + N 12.2 0.5 0.1
After 6 days - N 23.6 4.1 4.4

strated in sunflower plants by WAGNER and MICHAEL (1971) and SALAMA and
WAREING (1979). This nitrogen-induced change in cytokinin export from roots
is quite impressive in both the speed and extent of the response to nitrogen
supply as shown in Table 6.
Nitrogen supply also affects other phytohormones, and abscisic acid (ABA)
in particular. However, the effect is the reverse of that of the cytokinins. Nitro-
gen deficiency increases ABA in leaves (GOLDBACH et al. 1975), roots and
exudate (Table 7).
The importance of these phytohormones in cell metabolism (e.g. protein-
synthesis, stomatal regulation) and developmental processes (e.g. flower initia-
tion, senescence) indicates the necessity of considering the nutritional status
of plants in general and of nitrogen in particular in the relation to hormonal
activity. This applies not only to research in the physiology of yield formation
(see Chap. IV.2), but also in developmental physiology or stress physiology
(GORING and THIEN 1979).

1.4.5 Copper
Typical symptoms of copper deficiency in higher plants such as development
of "pendula" forms in trees (OLDENKAMP and SMILDE 1966) or wilting of young
1.1 General Introduction to the Mineral Nutrition of Plants 15

Fig. 1 A, B. Effects of chlorine defi- 120 y = 42.1+3.10x

ciency on growth and photosynthesis of r = 0.765 A
o
sugar beet (Beta vulgaris L.). (TERRY __ J!IggD....g>I!![Q!....Y!!!.u~ __
1977) .5'} o

r
10
o

25

0
y = 189 - 0.23x
0 0 0 r =-0.067 B
.. .c"' 0 00 0
-8"'- 200
0
0

-a.u
.c
:0 IX)

. <9
01 0
oE 100 0 0
u
(5 0
- E
~.=!-
0 10 20 30 40 50 60
Leaf blade CI- concentration
til mole g-' dry weight)

leaves (RAHIMI and BUSSLER 1973a, GRAHAM 1976) are related to the role of
copper in polyphenol-oxidase activity and thus the synthesis of lignins. The
extent to which lignification is inhibited in copper-deficient tissue is correlated
with inadequate development of xylem vessels (RAmMI and BUSSLER 1973 b).
A test for lignins can thus be used as a rapid method for diagnosis of the
copper nutritional status of plants (RAHIMI and BUSSLER 1973c, see also Chap.
1.5). This role of copper also explains the positive correlation in cereals, for
example, between copper nutritional status, haulm stability and resistance to
lodging, and the interaction between copper and nitrogen fertilizer application
(VETTER and TEICHMANN 1968). Another typical symptom of copper deficiency,
non-viability of pollen grains and the consequences for yield formation, is dis-
cussed in Section 4.

1.4.6 Chlorine
In the absence of chlorine, deficiency symptoms and severe growth depressions
have been demonstrated in higher (BROYER et al. 1954, JOHNSON et al. 1957)
as well as in lower plants (MARTIN 1965). Whereas in studies with isolated
chloroplasts a role of chlorine in photo system II electron transport has been
confirmed (KELLEY and IZAWA 1978), results of TERRY (1977) with intact'sugar
beet plants contradict its direct involvement in photosynthesis in vivo, as only
growth but not uptake of CO 2 was affected by chlorine (Fig. 1).
The relatively high chlorine demand for optimal leaf growth in sugar beet
(>20 ~mol g-l dry matter) might indicate genotypic differences in demand
for chlorine, with perhaps positive correlation with the halophytic nature of
a species. The high demand for chlorine for sugar beet certainly opens possibili-
ties for more detailed studies concerning the functions of this mineral element
in higher plants.
16 H. MARSCHNER:

2 Uptake and Long-Distance Transport of Mineral Elements

2.1 Ion Concentration at the Root Surface, Role of the "Rhizosphere"

In the past, studies on ion uptake by roots have mainly concentrated on the
role of the free space, carrier mechanisms, ion selectivity, and occurrence of
single- or multiple-binding sites at the plasmalemma or the" dual mechanism"
of ion uptake. However, in the last decade increasing attention has been paid
to electrophysiological aspects (Chaps. 1.6 and 1.7), as well as to physiological
and anatomical gradients along the roots and their consequences for uptake
and transport of ions (Chap. 1.7).
Another field of new interest in ion uptake and mineral nutrition in general
is the consequences which arise at the root surface as the result for example
of either removal of ions (development of a "depletion zone" at the root surface)
or difference in cation/anion uptake and the corresponding induced change
in pH around the root (Chap. 1.6). These factors are of less importance in
a stirred nutrient solution, but can be dominant in ion uptake and in the mineral
nutrition of plants growing in solid substrates such as soils. An example for
the role of the source of nitrogen on pH in the rhizosphere and its consequences
for phosphorus uptake by the plants is shown in Table 8.
Changes in the pH at the root surface are the result of both different cation/
anion uptake (NHt >S01-, N0 3 >Ca2+) and corresponding changes in net
effiux of H+ and OH- /HC0 3 . Nitrate reduction in the roots also leads to
additional OH- or HC0 3 effiux according to the equation: N0 3 +8 H+ +
8 e- =NH3 +OH- +2 H 2 0 (KIRKBY and MENGEL 1967).
Additionally certain plant species and genotypes of particular species suffer-
ing from iron deficiency are able to decrease the pH of the substrate by increas-
ing proton effiux (BROWN 1978). In sunflower, for example, this iron deficiency-
induced decrease in the substrate pH is mainly restricted to the apical zone
and is correlated with anatomical changes in the roots (see Chap. 1.7) and
a marked increase in iron uptake (Table 9).
Supplying a sparingly soluble inorganic iron-III-compound (e.g. iron-III-
hydroxide) this iron-stress-induced pH decrease leads to the mobilization of
iron in the substrate with correspondingly periodical changes in both the iron
nutritional status of the plants and pH fluctuations in the substrate (VENKAT-
RAru and MARSCHNER 1972). These effects can be depressed in amplitude but
not prevented by an increase in phosphate concentration (Fig. 2).
The ability of a plant to increase the uptake rate of iron in response to
iron stress is causally related to an increase in the reducing capacity of its
roots (BROWN 1978, ROMHELD and MARSCHNER 1979). Both response phenom-
ena are typical for so-called iron-efficient plant species (see also Chap. IV.1).
This stress response offers an interesting insight into a regulatory mechanism
of ion uptake controlled by the nutritional status (" internal concentration")
(more details in Chap. 1.6 and 1.7).
In addition to the pH in the rhizosphere of the root-solid substrate interface,
ion concentration at the root surface can also be quite different from the average
concentration in the substrate. For mineral nutrients such as phosphorus and
1.1 General Introduction to the Mineral Nutrition of Plants 17

Table 8. Effect of soil pH and supply of either ammonium or nitrate salts on pH of

the bulk soil and rhizosphere soil and on phosphorus uptake by soybean plants. (After
RILEY and BARBER 1971)

Soil pH in the soil P uptake

without salts (mmol pot-I)
without plants Bulk soil Rhizosphere soil
(outside
of the rhizosphere)

NHt. NO; NHt NHt

5.2 4.98 5.43 4.71 6.60 0.27 0.16

6.3 5.90 7.00 5.60 7.05 0.16 0.09
6.7 6.64 7.01 6.25 7.19 0.14 0.07

Table 9. Effect of iron nutritional status of sunflower

plants (preculture ± Fe) on iron accumulation in various
root zones. (RoMHELD and MARSCHNER 1979)

Preculture cm from the root tip

0-1.5 (root tip) 1.5-3.5 3.5-5.5

59Fe content (/lmol g-I dry wt. 10 h -1)

+Fe 0.3 0.6 0.6
-Fe 21.8 2.5 1.7

pH

,."..~"" ''I<, £,'j>

/' \, . .//1f
Fig. 2. Interaction between 6 I
I ro-\
""..0 ,
(\ i
FeEDTA
x-x 0.2 mM P

J "'\, ;
source of iron and phosphate - - . 0 1.0 mM P

concentration in iron stress 5 'I I? \\ I inorgan. Fe-m

reactions (fluctuations in 'I
I \

\ \} J "--
I \ I x---"'0.2mMP
the pH of the nutrient solu-
4
\j 0----01.0 mM P
tion) of sunflower plants. .....\..~
(MARSCHNER et al. 1978) 1 •
5 10 15 days

potassium, the diffusion rate to the root surface usually limits uptake rate,
and a depletion zone can occur around the root (BREWSTER and TINKER 1970).
This may restrict the uptake of these ions to the apical region of the roots
regardless of a potentially high uptake rate in the basal regions (Chap. 1.7).
Both root growth in general and root hair formation in particular (BHAT and
NYE 1974) are therefore of overwhelming importance for mineral nutrition in
solid substrates. In contrast they are of much less significance in solution culture
where exhaustion zones are of limited importance.
18 H. MARSCHNER:

Ion concentration at the root surface can also be much higher than that
in the average substrate. This occurs when the uptake rate of an ion (e.g. calci-
um) is relatively low as compared to the mass flow of solution to the root
surface induced by transpiration (MENGEL et al. 1969). As result of this process
the availability of both mineral nutrients (e.g. by precipitation ofCa phosphates)
and water (e.g. saline soils) can be depressed. For plants growing in solid sub-
strates and transpiring at a high rate, the concentration of soluble salts at
the root surface can be several times higher than in the bulk substrate (SINHA
and SINGH 1976).
Results from studies in solution culture investigating the effect of high salt
concentrations, e.g. NaCl, on ion uptake and water status of plants cannot
therefore be extrapolated to solid substrate conditions even when average iso-
osmotic concentrations in both substrates are considered.
Ion concentration and ion availability at the root surface can also be affected
indirectly by microorganisms in the rhizosphere. This subject is of increasing
interest in mineral nutrition in general and especially in relation to the actual
and potential role of certain associations for N2 fixation (Chaps. 11.2 and 11.3)
and symbiosis for phosphate uptake (e.g. mycorrhiza by crop plants (Chap.
1.2).

2.2 Long-Distance Transport in the Xylem

2.2.1 From the Roots to the Shoot

Our understanding of long-distance transport from roots to shoots is based
mainly on analysis of xylem exudate from decapitated plants. The composition
of this exudate, however, does not necessarily reflect conditions occurring at
the sites of excretion of the ions into the root xylem, as resorption by the
surrounding xylem can induce considerable changes in ion concentration and
relation ofions in the xylem during long-distance transport (Chap. 1.7). Further-
more the contribution made by phloem in vivo to long-distance transport of
mineral nutrients from roots to shoots cannot be determined by this method.
Only indirect evidence is available for this in vivo transport, as shown in studies
comparing forms of nitrogen supply (MARTIN 1971). The possibility must also
be considered that diurnal fluctuations (day-night) occur in the relative impor-
tance of mineral element transport in the phloem from roots to shoots, as
has been demonstrated for phloem transport into fruits (HOCKING et al. 1978).
For some mineral nutrients such as phosphorus (DE JAGER 1979) and potassi-
um (BARBER 1979) the uptake rate by the roots seems to be controlled by
the shoot rather than by the roots (Table 10).
The mechanism of this feed-back effect from the shoots is not fully under-
stood as yet (Chap. 1.6). For phloem-mobile mineral elements such as phospho-
rus and potassium, the transmission of information concerning the nutritional
status of the shoot may be achieved by the intensity of retranslocation of these
mineral elements in the phloem to the roots. In plants where circulation of
potassium from the shoots into the roots as salts of organic acids is important
(Chaps. 1.6 and 1.7), a regulation of potassium uptake by the potassium content
1.1 General Introduction to the Mineral Nutrition of Plants 19

Table 10. Effect of K pretreatment of corn plants (split root experi-

ment) on K content of shoots and roots and V max for subsequent
K influx. (After BARBER 1979)

Percent of root K content (mmol g-l dry wt.) V max

exposed to K pmol cm- 1 S-l
Shoots Roots

100 2.05 1.48 15.8

50 1.64 1.41 28.0
25 1.10 1.28 33.8 .
15 1.05 1.41 36.8

of the shoots and thus intensity of potassium retranslocation would be a conse-

quence of this model. In the case of phloem-immobile mineral elements such
as calcium the removal from exchange sites of the xylem might offer another
type of control mechanism (see Sect. 3).

2.2.2 Into Fruits, Seeds and Storage Organs

The transpiration rate of a plant organ (e.g. water loss g-l dry wt.) is one
of the determining factors in the ratio of mineral element import via xylem
versus phloem. Compared to fully expanded leaves, the transpiration rate and
thus mineral element import via the xylem can be expected to be much lower
in developing leaves and particularly fruits, seeds and storage organs (e.g. potato
tubers).
During recent years PATE and co-workers (PATE et al. 1974a, b, PATE and
HOCKING 1978) have developed a method to calculate the relative importance
of xylem and phloem in the transport of mineral elements into the fruits of
various legumes. The question of whether uncontaminated phloem sap is ob-
tained by this method is discussed later (Sect. 3). In general (PATE and HOCKING
1978), mineral element import (% of total import) into legume fruits occurs
predominantly via the phloem: 80% C, N, S; 70%-80% P, K, Mg, Zn;
62%-66% Fe, Mn, Cu and approximately 30% Ca. However, for certain ele-
ments (e.g. Cu and Fe) considerable diurnal fluctuations occur in the ratio
ofxylem/phloem import (HOCKING et al. 1978), and the ratio may also be depen-
dent on the availability of particular mineral elements in the root medium.
This has been demonstrated for manganese (Table 11).
It is obvious from Table 11 that when manganese availability is low, xylem
import of manganese into fruits is small relative to phloem import, but becomes
important at high manganese availability in the rooting medium. Manganese
concentrations in the phloem sap below 0.5 ppm are associated with a marked
increase in occurrence of a "split seed" disorder, involving discoloration, split-
ting, and deformity of seeds (HOCKING et al. 1977).
In this legume fruit system no data are presented for the two micronutrients
boron and molybdenum. For boron at least there is evidence that this element
is quite mobile in the phloem of stems and leaves (OERTLI and RICHARDSON
20 H. MARSCHNER:

Table 11. Effect of manganese level in the rooting medium on plant growth,
manganese level in seeds (dry matter) and in transport fluid serving fruits of
Lupinus angustifolius. (After HOCKING et al. 1977)

Concentration Dry weight Mn content Manganese concentration

in rooting (g plant- 1) in seeds
medium (ppm)
(ppmMn) Xylem (ppm) Phloem (ppm)

0.1 7.1 0.56 0.18 0.56

0.15 12.6 0.66 0.19 0.92
0.60 13.8 1.08 0.36 1.28
20.0 14.6 11.65 5.88 6.94

1970) and that fruits and storage organs may be supplied mainly via the phloem.
The same is true for growing potato tubers where the influx of xylem fluid
is small or unimportant. Only calcium, and no other mineral element, including
boron, needs to be supplied directly to the tuber surface to maintain a high
growth rate (KRAUSS and MARSCHNER 1975). Similarly, during fruit development
of Arachis hypogaea and Trifolium subterraneum net influx of xylem fluid is
very low, but the boron demand for fruit and seed growth is completely met
by internal supply, most probably by phloem import (CAMPELL et al. 1975).

2.2.3 Retranslocation of Mineral Elements from Leaves

This process is particularly important (a) in perennials in autumn, (b) in both

perennials and annuals when uptake is inadequate (c) during reproductive stages
or storage organ growth. The extent of retranslocation also depends upon the
particular mineral element. Occurrence of deficiency symptoms on older leaves
(N, K, Mg, P) reflects a high retranslocation rate, whereas the occurrenCe of
deficiency symptoms on only the young leaves (Ca, Fe, Mn, Cu, Zn, Mo, B)
demonstrates insufficient retranslocation rates of these mineral elements. This
retranslocation from the older leaves has to take place in the phloem. It cannot
be concluded, however, that the occurrence of the deficiency symptoms on
younger leaves reflects inadequate phloem mobility of these mineral elements.
Retranslocation from the older leaves involves several steps: (1) "Mobilization"
within the leaf cells, (2) transport to the sieve elements, (3) phloem loading,
(4) mobility within the sieve tubes.
Phloem mobility is thus only one factor which determines the retranslocation
rate of a mineral element from the leaves.
Retranslocation from leaves is particularly important during seed develop-
ment, as in this growth stage the carbohydrate supply to the roots and thus
root activity and ion uptake generally decline rapidly. The extent of retransloca-
tion of certain mineral elements during fruit and storage growth depends upon
various factors, such as uptake rate by the roots and "sink-size", particularly
in relation to "source-size" (see Sect. 4). An example for the range ofredistribu-
tion is given in Table 12.
I.1 General Introduction to the Mineral Nutrition of Plants 21

Table 12. Retranslocation during the reproductive stage (percentage of the proportional
net loss from the time when maximum content was recorded) of mineral elements from
leaflets of the main axis of Lupinus albus, corresponding loss in dry weight 39%. (After
HOCKING and PATE 1978)

Element Loss in % Element Loss in % Element Loss in %

Nitrogen 73 Calcium 18 Manganese 50
Potassium 73 Iron 25 Copper 43
Phosphorus 75 Zinc 45 Sodium 4
Magnesium 57

The high retranslocation rate of N, P, K, and Mg from leaves is in accor-

dance with the occurrence of deficiency symptoms on (older) leaves insufficiently
supplied by the root system. Although somewhat lower than for the macronu-
trients, the rettanslocation rate of the micronutrients Fe, Zn, Mn and Cu (Table
12) is still surprisingly high, considering the fact that deficiency symptoms of
these elements occur only on young, growing parts (leaves), but not on fully
expanded leaves.
For the micronutrients only a certain fraction can be mobilized from fully
expanded leaves (LONERAGAN et al. 1976, LONERAGAN 1978). It appears that
when the shoot apex or fruits are inadequately supplied, the demand cannot
be transformed into a "signal" strong enough to bring about extensive mobiliza-
tion and retranslocation from fully expanded leaves. These leaves therefore
do not show micronutrient deficiency symptoms. In contrast, in the case of
the macronutrients nitrogen, potassium and magnesium, this "signal" seems
to be strong enough to induce exhaustion of the older leaves in these particular
elements.
A close correlation between concentration and retranslocation rate of miner-
al elements from older leaves has been clearly demonstrated by HILL et al.
(1978, 1979). In wheat, for example, during grain development, leaves with
a high copper content lost more than 70% of their copper, whereas for leaves
of copper-deficient plants the value was less than 20%. This proportion of
copper retranslocation can be increased, however, particularly from copper-
deficient leaves, for example, by shading (HILL et al. 1979). During senescence,
induced either by shading or by nitrogen deficiency (HILL et al. 1978), a further
fraction of copper is mobilized in these older leaves and is retranslocated. The
relatively high retranslocation rate of micronutrients during fruit development
(Table 12) is perhaps the result of leaf senescence. The extent of retranslocation
may also be causally involved in the development of sulfur deficiency symptoms
preferentially either on young or old leaves depending on the level of nitrogen
supply (LONERAGAN et al. 1976). This variability in extent of retranslocation
of a mineral element and its consequences for plant diagnosis are discussed
in more detail in Chap. 1.5.
The extent of remobilization of macro- and micro nutrients is attracting in-
creasing interest also in connection with selection and breeding of genotypes
of high "nutrient efficiency". Genotypes which are able to grow well on sub-
22 H. MARSCHNER:

strates (soils) with relatively low nutrient availability may differ not only in
the uptake and translocation rate of a particular mineral nutrient, but also
in a higher mineral nutrient effectivity at cellular level (compartmentation, solu-
bility etc.) and higher retranslocation rates from older to younger leaves and
to seeds and storage organs. (For further details see Chaps. IV.1 and IV.2
and a comprehensive review article by PATE 1976).

3 Calcium Nutrition of Higher Plants

3.1 Introduction

Calcium nutrition of higher plants is attracting increasing interest due to the

widespread occurrence of calcium deficiency or the so-called calcium-related
disorders which are particularly associated with intensive horticulture and agri-
culture. Typical examples are tipburn in lettuce, blackheart in celery, blossom
end rot in tomato and water melon or bitter pit in apple. For detailed informa-
tion the reader is referred to comprehensive review articles (SHEAR 1975, BAN-
GERTH 1979) and an International Symposium Report (SHEAR 1979).
Besides its economic -importance, this subject is also of interest for plant
physiologists, as these Ca-related disorders often occur on soils or substrates
high in available calcium, and thus are only in part caused by insufficient calcium
uptake. It is well established now that these disorders are primarily caused
by the limited capability of plants to regulate calcium distribution internally
in relation to the demand of the various parts and organs and particularly
of fast-growing leafy organs, tubers and fruits.

3.2 Calcium Demand of Higher Plants

There are considerable genotypical differences in the demand for calcium at

a cellular level, causally connected with differences in cation exchange capacity
(CEC) of the cell wall system (middle lamella) i.e. amount of free carboxylic
groups of pectins (polygalacturonic acid). Thus dicotyledons, as compared to
monocotyledons, are characterized by both higher CEC and also much higher
calcium demand. A typical example for these differences is presented in Table
13. In agreement with this, Ca-related disorders are typical for dicotyledons
but not for monocotyledons. '
As the calcium supply falls, the proportion of calcium bound in the cell
wall fraction (pectates) increases to more than 50% (MOSTAFA and ULRICH
1976, ARMSTRONG and KIRKBY 1979). In plant species with intensive oxalate
formation, the availability of calcium for the other functions, membrane stabili-
zation in particular (Chap. V.1) is further decreased (ARMSTRONG and KIRKBY
1979). Genotypical differences to calcium deficiency within a certain species
are therefore at least in part related to differences in oxalate formation, particu-
1.1 General Introduction to the Mineral Nutrition of Plants 23

Table 13. Effect of calcium concentration in nutrient solution (11M) on relative growth
rate of plants and calcium content (Ilffiol g-l dry wt.) in the shoots. (After LONERAGAN
et al. 1968 and LONERAGAN and SNOWBALL 1969)

Species Ca supply (11M)

0.8 2.5 10 100 1000

Relative growth rate (%)

Lolium perenne 42 100 94 94 93
Lycopersicon esculentum 3 19 52 100 SO
Calcium content <!lmol g-l dry wt.)
Lolium perenne 15.0 17.5 37.4 92.3 269.5
Lycopersicon esculentum 49.9 32.4 74.9 321.9 621.3

larly in certain plant parts such as the stem (BRUMAGEN and HIATT 1966) and
perhaps also the pedicels (LIEGEL 1970).
Besides this special case of oxalate formation, genotypical differences to
calcium deficiency within a certain species may also be related to differences
in uptake, long- and short-distance transport and/or compartmentation within
the cells (see below and Chap. IV.1).

3.3 Calcium Uptake by the Roots

The mobility of calcium is much greater in the apoplast than in the symplast.
In roots the uptake and radial transport of calcium into the stele is therefore
more intensive in the root tip regions (Chap. I.7). Calcium transport into the
shoot therefore seems to be more closely related to root growth rate (VISSER
et al. 1971, SCAIFE and CLARKSON 1978) than to the shoot growth rate (RApPER
et al. 1977). Certainly the classical approach of a general close relationship
between uptake of calcium and transpiration rate has to be reexamined (ARM-
STRONG and KmKBY 1979).

3.4 Long-Distance Transport of Calcium

3.4.1 Xylem Transport

Although the direction of water flux within the xylem determines the direction
of long-distance transport of calcium in this conducting system, the- transport
rate of calcium is only poorly correlated to the flux rate of water in the xylem.
The transport rate of calcium in the xylem is regulated more by exchange adsorp-
tion than by mass flow (FERGUSON and BOLLARD 1976, ISERMANN 1978). Thus
at high flux rates of water in the xylem the discrepancy to the transport rate
of calcium can be expected to be particularly wide (see Figs. 3 and 4). It is
also important to realize that this exchange adsorption is not restricted to the
24 H. MARSCHNER:

shoot apex
Fig. 3. Model for calcium trans-
location (... _ - --+ decreas-
ing rate) within the shoot. Mass
flow of water in the xylem (c:=>
= , decreasing rate). Fixed neg-
ative charges in cells walls (-)
and formation of new exchange
sites -, sites of IAA synthesis and
transport [IDIlnn-

fruit

old leaf

H H
H
~~~H H H
t:;;
H H

..

Fig. 4. Ratio between calcium accumula-

~O.1
.
'-
()
tion (~mol) and water loss (ml) in paprica
fruits (Capsicum annuum) during fruit
O~--~Q2--~O~A---O~.6~~Q~8---1~~--~1.2~
growth. (MIX and MARSCHNER 1976b)
dry weight (g/fruit)

xylem vessels, but can take place in the apoplast (Free Space) of the whole
stem, for example (VAN DE GEIJN et al. 1979). It may be supposed that the
mobilization of calcium from the stem in apple trees during the, early stages
of fruit growth (UiUDERS et al. 1976), for example, takes place at least in part
at the expense of this exchangeable fraction. Thus the transport of calcium
within the xylem can be controlled to a considerable extent by charge density
(dicotyledons> monocotyledons), concentration of cations other than calcium,
and the removal of calcium from the exchange sites by adjacent cells along
the conducting vessels and particularly in dividing and expanding tissues (e.g.
fruits, shoot apex, see Fig. 3).
1.1 General Introduction to the Mineral Nutrition of Plants 25

Table 14. Effect of environmental factors and growth rate on Ca and K content of
paprika fruits (MIX and MARSCHNER 1976a, b, MARSCHNER unpublished)

Treatment Ca K

(~ol, g-l dry wt. fruit)

Ca supply to the roots O.5mM 26.9 1315

5.0mM 33.2 1228
Relative humidity around 40% 55.4 1892
the fruits 90% 32.7 1918
Growth rate of the fruits 20.1 28.2 1772
(mg- 1 dry wt. increase per day) 29.9 20.7 1846
38.5 17.2 1813

3.4.1.1 Correlations with the Transpiration Rate

Inhibition of the transpiration rate decreases the calcium translocation to the
shoot apex, but not necessarily to the shoot in total (ARMSTRONG and KIRKBY
1979). Inhibition of transpiration rate of individual organs like fruits is also
correlated with a decrease in calcium translocation into these organs (WIERSUM
1966, Table 14).
Individual organs or plant parts with relatively large surface areas (e.g.
leaves in contrast to fleshy fruits) or those directly exposed to the free atmo-
sphere (e.g. outer leaves in contrast to inner leaves oflettuce) are preferentially
supplied with water from the xylem and thus also with calcium. As water avail-
ability in the shoot is depressed either by a high osmotic pressure of the soil
solution (ENDE et al. 1975) or by conditions which favor a high rate oftranspira-
tion of the shoot as a whole (KRUG et al. 1972), the net flux of water and
calcium within the xylem into fruits or inner leaves can either be temporarily
(light period) or permanently (saline substrates) restricted. Such restrictions are
associated with an increase in calcium-related disorders (KRUG et al. 1972, ENDE
et al. 1975). This transpiration-directed flux in the xylem of both water and
calcium can in part be compensated or superimposed by a high root pressure
and lead to influx of water and calcium via the xylem into these low-transpiring
organs and plant parts (PALZKlLL et al. 1976, BRADFIELD and GUTTRIDGE 1979).
Diurnal fluctuations in the water status of the shoot and its individual organs
(KRUG et al. 1972, WIEBE et al. 1977) are correlated with corresponding fluctua-
tions in the direction of the calcium transport within the xylem (KRUG et al.
1972, KRAUSS and MARSCHNER 1974, WIEBE et al. 1977).

3.4.1.2 Role of Binding Sites in the Apoplast

Regardless of the transpiration rate of the whole plant (WIENEKE and Fiilm.
1973a) or of individual leaves (KOONTZ and FOOTE 1966) or of fruits (Table
14) the translocation rate of calcium declines as th~ plants or organs mature
(Fig. 4). Two factors appear to be responsible for this (see Fig. 3): (a) In fruits
an increase in water influx via the phloem (" Druckstrom" according to MUNCH)
26 H. MARSCHNER:

depresses influx via the xylem, (b) in the case of fully developed leaves due
to a decline in removal of Ca2+ from the binding sites of the apoplast. This
removal of calcium depends mainly on storage as calcium salts in the vacuoles.
The storage capacity of vacuoles for soluble calcium salt (nitrate, malate etc.)
is, however, restricted for osmotic reasons. In the case of the formation of
calcium oxalate as a result of nitrate reduction (EGMOND and HOUBA 1970),
xylem transport of calcium into fully grown leaves might be maintained for
a longer period depending on nitrate reductase capacity. The effect of the form
of nitrogen supply (nitrate or ammonium) on calcium distribution within apple
shoots (SHEAR and FAUST 1970) might at least in part be the mechanism de-
scribed above. Removal of calcium from exchange sites in the apoplast might
also be a dominating factor for calcium transport into growing tissue such
as shoot apices, young leaves or fruits (Fig. 3). Cell division, and thus the
formation of new binding sites for calcium, should decrease calcium ion activity,
thus maintaining calcium transport within the apoplast towards this tissue. This
mechanism, of course, at least temporarily requires a net flux of solutes within
the xylem in the direction of this tissue. Such conditions can be demonstrated,
for example, at high root pressure (PALZKILL et al. 1976, BRADFmLD and GUT-
TRIDGE 1979). The importance of these binding sites in the apoplast for calcium
transport can be demonstrated by several examples. A decrease in CEC of
a tissue, for example as result of manganese toxicity (HORST and MARSCHNER
1978b) or application of TIBA (MARSCHNER and OSSENBERG-NEUHAUS 1977)
is correlated with a decrease in calcium transport into the tissue and the develop-
ment of calcium-deficiency symptoms. Even the calcium distribution within an
organ is correlated with the CEC. In flowers of Antirrhinum, for example, there
is a simultaneous increase in CEC, calcium content and CalK ratio from
stigma - style - ovary (KNIGHT and CROOKE 1973).

3.4.2 Phloem Transport

Compared to the xylem sap, the calcium concentration in the phloem sap is
very low, particularly in relation to potassium. In phloem sap of Ricinus stems
the following values have been obtained: 0.1-0.7 mM calcium as compared
to 50-100 mM potassium 01AN GOOR and WmRsMA 1974, MENGEL and HAEDER
1977). Considering the high pH (7-8) and the high phosphate concentration
of the phloem sap, the solubility of ionic calcium can be calculated to be lower
than 0.5 mM 01AN GOOR and WmRSMA 1974). According to RAVEN (1977),
however, concentrations of up to 3 mM calcium should theoretically be possible.
The calcium concentration might also depend upon complexing agents in the
phloem sap (BRADFmLD 1976).
Low calcium concentrations in the phloem sap (see also PATE Vol. 1, this
Series) and relatively high concentrations in the area of phloem strands
(WmNEKE 1979) respectively are presumably the result of either precipitation
of calcium (LXUCHLI 1967, WmNECKE and Ffurn. 1973b) or of a low membrane
permeability of the sieve tube cells to calcium. Also the existence of efflux
pumps for calcium at the sieve tube membranes is discussed (Chap. V.1, this
Vol.).
1.1 General Introduction to the Mineral Nutrition of Plants 27

Table 15. Estimated deliveries of mineral elements by

phloem and xylem and the proportions of actual intake
in fruits of Lupinus angustifolius L. attributable to vascular
transport within a 36-h period. (After HOCKING et al. 1978)

Element Delivery in lJIDol % of total

(actual)
Phloem Xylem intake

K 8.21 2.14 103.5

P 2.30 0.12 97.7
Ca 0.12 0.85 34.0
Fe 0.04 0.01 103.6

Calculations for phloem transport of calcium are usually based on analysis

of phloem sap collected by incision of the stem or removal of the distal end
oflegume fruits (PATE et al. 1978, PATE and HOCKING 1978). With this method,
however, "contamination" by calcium from the apoplast cannot be excluded.
In legume fruits, for example, calcium is highly mobile in the apoplast of the
pod tissue (MIX and MARSCHNER 1976c). The difficulties in calculating the
calcium transport into legume fruits based on xylem and phloem exudate analy-
sis are reflected in Table 15.
Another approach used in the determination· of calcium transport in the
phloem is by ringing plant stems. Such experiments have led to the conclusion
in apple shoots (PRIESTLEY 1976) and cotton stems (WIENECKE 1979) that calcium
is transported preferentially in the phloem. It has to be critically re-examined
whether such a severe disturbance of the integrity of the whole plant system
would not induce side effects as, for example, on calcium movement in the
apoplast.
At least in intact roots (MARSCHNER and RICHTER 1974) or stolons of
growing potato tubers (KRAUSS and MARSCHNER 1973) it is not possible to
observe long-distance transport of calcium within the phloem in the direction
of the apical meristems. Retranslocation of calcium after leaf application of
45Ca can be demonstrated only when high concentrations of calcium are applied
(RrNGOET et al. 1968) or when calcium is supplied combined with a complexing
agent directly to the leaf veins (MILLIKAN and HANGER 1969). On the other
hand, in studies with intact legume plants, between flowering and ripening a
substantial net loss of calcium occurred from the upper leaves, associated with
an increase in calcium content of the fruits (HOCKING and PATE 1978). It is
certainly an interesting aspect for further studies on phloem transport of calcium
as to whether retranslocation of calcium is in some way related to tissue'senes-
cence, as has been demonstrated with copper (HILL et al. 1978).

3.4.3 Xylem Versus Phloem Transport

Despite these contradictory results and differing views about the importance
of calcium transport in the phloem, there is general agreement that the calcium
concentration in the xylem is several times or even several orders of magnitude
28 H. MARSCHNER:

35 Fig. 5. Correlation between seed

basal weight and direction of 45 Ca trans-
port within bean fruits (Phaseolus
x vulgaris) 3 days after application
c
0 of 45Ca ( l ) on the dorsal vascular
15u strand in the middle of the fruit.
0
Vi (MIX and MARSCHNER 1976a)
c
~
0
u
([)
--.t
'[!. 10
•
apikal

60 70 80
D.M. mg/seed

higher than in the phloem. Consequently, plant organs or parts of plants mainly
supplied by the phloem stream are generally high in potassium but low in
calcium. In leafy tissue (e.g. lettuce), fleshy fruits (e.g. tomato), storage roots
(e.g. celery) or tubers (e.g. potato) for the average of the whole period of growth
the relative importance of the phloem/xylem influx ratio remains more or less
constant. Thus either direct uptake of calcium from the external medium, as
shown for peanut fruits (HALLOCK and GARREN 1968) or potato tubers (KRAUSS
and MARSCHNER 1975), or short-term periodical changes in the water influx
via the xylem (see Sect. 1.1) are essential to meet the calcium demand of these
growing tissues.
In fleshy fruits such externally regulated fluctuations of the xylem fluid
influx rate are superimposed by an internal regulation. During fruit growth,
the rate of calcium influx distinctly declines (WIENEKE 1974; Fig. 5). Several
factors are responsible for this decline: (a) change in the volume/surface area
ratio (b) a decline in cell division rate and thus formation of new binding
sites for calcium and (c) an increase in the relative importance of solute influx
via the phloem.
The ratio of solute influx of xylem/phloem also depends on the growth
rate of the fruit. Fast-growing fruits are therefore usually lower in calcium
and have a wider K/Ca ratio than slow-growing ones. This has been demon-
strated for tomato (WIERSUM 1966), apple (BANGERTH and MOSTAFAWI 1969)
and paprika (Table 14). In fleshy fruits at· the later stages of growth, even
a net efflux of calcium from the fruits seems to occur (WILKINSON 1968, TROMP
and OELE 1972, MIX and MARSCHNER 1976a).
In fruits with distinct phases of growth such as legume fruits, the onset
of rapid seed growth is correlated with a reversal from apical to basal direction
of net solute transport in the xylem of the dorsal vein of the pod (Fig. 5).
This change leads to a distinct decline in calcium transport into these fruits.
At these later stages of fruit growth, net influx of calcium either ceases or
is restricted to a low influx rate via the phloem, as supposed by PATE and
HOCKING (1978) or TROMP (1979).
1.1 General Introduction to the Mineral Nutrition of Plants 29

In Phaseolus vulgaris the developing seeds are supplied directly with calcium
formerly accumulated in the pod tissue (MIX and MARSCHNER 1976c), a way
of internal redistribution which presumably also occurs in other legume fruits
(HOCKING and PATE 1977). This uptake from the surrounding apoplast into
the growing seed is severely restricted (regulated) by the seed coat (MIX and
MARSCHNER 1976c), where most of this calcium is immobilized and is unavail-
able even for subsequent seed germination (HELMS and DAVID 1973).

3.5 Role of Phytohormones and Growth Regulators

Phytohormones and growth regulators also influence the long-distance transport

of calcium. WIENEKE et al. (1971) assume a direct depressing effect of gibberellins
on calcium transport to the shoot apex or fruits of apple. The most marked
effects on calcium transport are exerted by auxins. Evidence for this is the
distinct inhibitory effect of TIBA (2,3,5-triiodobenzoic acid) on calcium trans-
port into fruits (BANGERTH 1976) or to the upper parts of shoots (WIENEKE
et al. 1971) and the apical meristem in particular (MARSCHNER and OSSENBERG-
NEUHAUS 1977). As TIBA is known to be an inhibitor of basipetal auxin trans-
port (DE LA FUENTE and LEOPOLD 1972), inhibition of this transport from the
sites of synthesis, the apical meristem and developing seeds (NITSCH 1950),
can be expected to be correlated with a corresponding decrease in calcium
transport into these tissues and organs (Fig. 3). Causal interactions between
auxins and calcium mobility have been shown in other tissue (JANISTYN 1973).
The importance of seeds as sites of auxin synthesis for calcium transport
into developing fruits has been demonstrated by BANGERTH (1976). The effect
of boron sprays in increasing calcium transport into the fruits (DIXON et al.
1973) may relate to conditions where boron deficiency restricts seed development
and thus auxin production. Furthermore, competition for calcium between dif-
ferent plant parts (e.g. developing fruits and shoot apices) might be related
to a capability for auxin synthesis. Removal of shoot apices (QUINLAN and
PRESTON 1973) or inhibition of shoot growth by growth regulators (NAUMANN
1971) can therefore indirectly affect partitioning of calcium within the shoot
in favour of calcium transport into fruits.
The linkage between basipetal auxin transport and acropetal calcium trans-
port is still obscure. Auxins might increase mobility of calcium both in the
apoplast and in the phloem (BANGERTH 1976), or enhance formation of new
binding sites in the apoplast (MARSCHNER and OSSENBERG-NEUHAUS 1977). On
the other hand, supraoptimal concentrations of auxins and a correspondingly
particular high demand for calcium for elongation growth are assumed to be
the explanation for the negative correlations obtained between auxin levels and
calcium-related disorders in lettuce cultivars (COLLIER et al. 1979).

3.6 Conclusion and Outlook

Calcium transport into rapid-growing leafy tissues, tubers or fleshy fruits is

obviously characterized by mechanisms which restrict this transport. Consider-
30 H. MARSCHNER:

ing some of the effects of calcium - decreasing membrane permeability or inhibit-

ing auxin-induced cell elongation (POOVAIAH and LEOPOLD 1976) - it is necessary
that in rapidly expanding tissue and storage tissue in particular, the concentra-
tion of calcium is kept low to maintain high membrane permeability for solute
flux. In agreement with this, an artificial increase in calcium concentration of
fruits rapidly decreases their growth (MIX and MARSCHNER 1976a). Further-
more, "ripening" of fruits (onset of senescence) requires both an increase in
membrane permeability and a dissolving of the middle lamella, both processes
which are delayed or even prevented by high concentrations of calcium (see
Sect. 1).
Impressive crop yield increases have been achieved in recent years by improv-
ing agricultural and horticultural techniques together with progress made in
plant breeding. Many of those plant organs or plant parts which are considered
as "yield", however, are also often characterized genetically by restricted calci-
um influx. Increasing calcium-deficiency symptoms and economic loss due to
Ca-related disorders are the consequence. Although our present knowledge of
calcium transport and its regulation is still rather limited, the causal connection
directly with the water economy of a plant~ and its organs, and indirectly with
the phloem to xylem flux ratio into an organ, enables some control of calcium
distribution. The involvement of phytohormones and auxins in particular, en-
ables new approaches to be made. In the long run, to overcome Ca-related
disorders, it will certainly also be necessary to make more use of genotypical
differences within species in uptake, transport and distribution of calcium at
tissue and cellular level. These latter aspects are discussed in detail in Chap.
IV.1.

4 Mineral Nutrition and Physiology of Yield Formation -

Sink-Source Relationship

4.1 Introduction

In determining plant yields, as defined by the amount of either fresh or dry

matter or certain constituents (e.g. sugar, starch, protein) produced per unit
area, photosynthesis is the most important prerequisite. This holds true in partic-
ular for forage plants which show a close relationship between an increase
in photosynthetic active surface area and an increase in fresh and dry weight.
Here mineral nutrients determine yield, depending on their function as constitu-
ents of organic compounds, enzyme activators, osmoregulators etc. (see Chaps.
II-V).
In most crop species, however, only certain plant parts or organs (fruits,
roots and tubers for example) are considered as "yield". As the photosynthetic
activity of these plant parts or organs is generally low or even absent, they
act as a "sink" for assimilates produced in other plant parts (" source "). Increas-
ing evidence has been presented during the past 20 years.that under these circum-
1.1 General Introduction to the Mineral Nutrition of Plants 31

stances the sink itself exerts an important regulatory function and can be rate-
limiting for yield formation. This source-sink relationship is determined by
phytohormones in a more or less specific manner, and this has been demon-
strated in a number of cases. For further information the reader is referred
to a comprehensive review article by MICHAEL and BERINGER (1980).
It is the aim of this section to attract attention to the fact that mineral
nutrients can also exert a distinct regulatory function in this source-sink rela-
tionship. This may occur directly or indirectly by changing the phytohormone
balance.

4.2 Effect of Mineral Nutrition on Phytohormone Level and Sink Formation

Environmental factors such as day length and temperature affect flower forma-
tion (i.e. the initiation of seeds and fruits as sink). This occurs primarily by
changes in the phytohormone balance, as for example the ratio of gibberellic
acid (GA) to cytokinins (CYT). CYT playa particular role in flower initiation
(BRUINSMA 1977). As the main sites of CYT synthesis in higher plants are
the root tips (see review article of VAN STADEN and DAVEY 1979), all environmen-
tal factors - including supply of mineral elements - which affect root growth
are closely related to changes in the export of CYT towards the shoot. This
can be demonstrated for nitrogen (see below) and phosphorus (MENARY and
VAN STADEN 1976, SALAMA and WAREING 1979). The close correlation between
phosphorus supply, CYT production and flower formation in tomato is shown
in Table 16. From the results of Table 16, it may also be supposed that the
increase in spikelet initiation of wheat with increasing phosphorus supply
(RAHMAN and WILSON 1977) is causally connected with the CYT level. The
same applies to the close positive correlation which has been observed in apple
trees between phosphorus nutritional status at the time of flower differentiation
and the number of flowers and fruit yield per tree in the subsequent year (BOULD
and PARFITT 1973).
The nutritional factor which exerts the most distinct effect on phytohormone
level and flower formation seems to be nitrogen. Inorganic nitrogen supply
to the roots stimulates the production and export of CYT to the shoots (WAGNER
and MICHAEL 1971, SATTELMACHER and MARSCHNER 1978, see also Sect. 1).
In apple, ammonium-N as compared to nitrate-N is somewhat more effective
in this respect (BUBAN et al. 1978). It is questionable, however, whether the

Table 16. Effect of short-term phosphorus deficiency ( - P) on flowering and cytokinin

in stem exudate of tomato. (MENARY and VAN STADEN 1976)

-P +P +P + Kinetin a
No. flowers (first truss) 5 7 (16)
g callus flask - 1 b 0.78 1.12

a Direct application to the shoot

b CYT activity in stem exudate determined by tobacco callus test
32 H . MARSCHNER :

Table 17. Effect of ammonium versus nitrate supply on flowering of Jonathan apple
trees in the flowering growing season. (After GRASMANIS and EDWARDS 1974)

Treatment: Mean number of buds per tree Percentage of

duration and time flower buds
Emerged Flower

Continuous nitrate 35.3 12.8 38.7

8 weeks ammonium (Dec-Jan) 33.3 21.2 63 .7
24 h ammonium (Feb) 38.8 30.3 77.5

Fig. 6. Alternations in tuberization and "regrowth " of potato induced by fluctuations

in nitrogen supply to the plants. (KRAUSS and MARSCHNER 1976)

marked increase in flower formation induced already by a short-term supply

of ammonium-N (Table 17) can be explained solely in terms of CYT.
The nitrogen nutritional status of plants is also closely related to the level
of other phytohormones. Nitrogen deficiency or even short-term interruption
in nitrogen supply to the roots increases the export of abscisic acid (ABA)
from roots (KRAUSS 1978) and the ABA content of the shoot~ (GOLDBACH
et al. 1975). This is correlated with decrease in GA content of the shoots (RAJA-
GOPAL and RAo 1974). The interaction between nitrogen nutrition, CYT level
and flower initiation as described above might therefore be only one example
in a more complex system of regulation of developmental processes by nitrogen.
Regardless of this uncertainty in causal interpretations, application of nitrogen
at the time of flower initiation can be a quite effective tool in increasing the
number of flowers (TROBISCH and SCHILLING 1970) and in decreasing the yearly
fluctuations in yield of perennials (LOODERS and BUNEMANN 1970).
1.1 General Introduction to the Mineral Nutrition of Plants 33

Changes in phytohormone balance are also responsible for the regulatory

role of nitrogen in tuber initiation in potato. In this plant species, fluctuations
in nitrogen supply either as nitrate or as ammonium are closely <::orrelated
not only with corresponding changes in CYT and ABA export from the roots
but also with alterations in tuberization and" regrowth" of the tubers (Fig. 6).
The same effects resulting from nitrogen supply can be obtained by direct appli-
cation of phytohormones (ABA, GA) to the stolons and tubers respectively
(KRAuss and MARSCHNER 1976). This result further supports the idea of the
regulatory role of nitrQgen on induction of both flowers and tubers via a change
in the phytohormone balance. Changes in the CjN ratio are unlikely to be
the causative factor, as only nitrogen supplied to the roots and not that supplied
to the leaves is effective in regulation of tuberization (SATTELMACHER and
MARSCHNER 1979).

4.3 Effect of Mineral Nutrients on Fertilization

The number of seeds and/or fruits per plant or both can also be directly in-
fluenced by mineral nutrients. This is most clearly seen with copper and boron.
In cereals in particular copper deficiency affects the reproductive phase more
than the vegetative phase. The primary cause of failure in grain setting is the
non-viability of pollen of copper-deficient plants (Table 18). The synchronous
meiotic division of a large number of pollen mother cells and the corresponding
high localized copper demand are likely to be the primary reason for this symp-
tom of copper deficiency (GRAHAM 1975). This might be a useful tool for plant
breeding purposes to induce male sterility by a nutritional factor.
The necessity of boron for pollen tube growth (see Chap. V.2) is reflected
for example in a decrease in grain number per head in rice (GARG et al. 1979)
or even a total lack of fertilization in barley (SIMOJOKI 1972). As demonstrated
in cross-pollinating experiments (VAUGHAN 1977) the failure of seed formation
in maize suffering from boron deficiency is not due to lack of pollen viability
but caused by non-receptiveness of the silks. Dry matter production and its
distribution in maize as affected by boron supply (Fig. 7) provided an excellent
example of sink limitation. Increasing the boron supply led to an increase in
sink size (seed formation) correlated to a corresponding increase in photosyn-
thetic capacity of the leaves.

Table 18. Grain set by cross-pollination between Cu-deficient and

Cu-sufficient wheat plants. (GRAHAM 1975)

Cross No. of Total No.

grain
female x male set floats heads

-Cu x -Cu 0 73 3
+Cu x -Cu 2 76 3
-Cu x +Cu 47 157 7
34 H. MARSCHNER:

Fig. 7. Effect of boron supply to

maize plants (Zea mays L.) on the
production and distribution of dry
matter. (Data from VAUGHAN 1977)

Boron s\.Wly mg ~ plant

4.4 Source-Sink Interactions in Relation to Mineral Nutrition

In higher plants after fruit setting or tuber initiation, these newly formed storage
organs represent the dominating sink. However, other plant parts can also act
as competing sinks of different strengths. In annual plant species mainly the
roots are effective, whereas in perennials both roots and apical parts of vegeta-
tive tissues playa role as competing sinks.
Usually in crop species the dominance of the fruits in sink competition
is quite obvious and certainly a result of successful selection and breeding for
higher crop yield. Examples showing extreme competition have been presented
by LENZ (1970) for egg plants (Solanum melongena L.) and are demonstrated
for citrus (LENZ and DORING 1975). Increase in fruit number from 0 to 100
per citrus tree strongly depressed further growth of leaves, stems and roots.
The decrease in leaf area was correlated with a threefold increase in the transpi-
ration rate per unit leaf area, a decrease in stomatal diffusion resistance (see
also LENZ and WILLIAMS 1973) and a corresponding stimulation in net photosyn-
thesis (see also Chap. IV.2).
CYT produced in the roots is distributed within the shoots, depending mainly
on the transpiration rate of the various organs of the shoot (MICHAEL et al.
1969a, b, MONSELISE et al. 1978). As CYT is involved in both delaying senescence
and in increasing sink activity and capacity (e.g. number of cells per organ;
SEILER-KELBITSCH et al. 1974) the transpiration stream-directed distribution of
root-borne CYT may be considered as an important regulatory step in the
source-sink relationship (see also CARMI and KOLLER 1979). The higher content
of nitrogen and calcium in the leaves of plants with heavy fruit load and lower
leaf area (LENZ and BUNEMANN 1969) might therefore not only be a necessary
result of the increased transpiration rate, but might also be considered of imp or-
tance in delaying senescence (see Sect. 1) and maintaining the source. On the
other hand, partial defoliation favors CYT import into fruits (MONSELISE et al.
1978) and strongly stimulates the rate of fruit growth (sink activity) in seedless
tomato fruits which mainly rely on root-borne CYT (VARGA and BRUINSMA
1974). In cereals such as barley the beneficial effect of awns as transpiring
organs on the single grain weight (MICHAEL et al. 1969) is another example
for this type of regulatory mechanism.
The importance of root growth for CYT supply to the source (leaves) so
that it can compete more effectively with the sink becomes understandable
1.1 General Introduction to the Mineral Nutrition of Plants 35

Table 19. Effect of leaf application of growth sub-

stances on dry matter distribution in Daucus carota.
(LINSER et al. 1974)

Treatment g dry wt. per plant

Shoot Root Total

H 2O 3.2 10.9 14.1

CYT 7.3 8.8 16.1
GA 9.9 5.7 15.6
CCC 2.8 10.8 13.6

in the context of the whole plant. The well-known effect of late nitrogen applica-
tion to cereals causing delay in leaf senescence and increase in length of ripening
(MICHAEL et al. 1974, MICHAEL and BERINGER 1980) has to be considered in
relation to increased CYT export from the roots into the leaves. A similar
interaction might be assumed in potatoes, where a sudden increase in nitrogen
supply leads to a temporary "switching-off" of the tubers as a sink (KRAUSS
and MARSCHNER 1971).
The mode of nitrogen supply also presumably affects source-sink interac-
tions via the action of other phytohormones. Interruption of the nitrogen supply
to the roots, for example, also causes a marked increase in ABA content of
the root exudate as well as the leaves (GOLDBACH et al. 1975, KRAUSS 1978).
As ABA is very effective in inducing stomatal closure (ITA! and MEIDNER 1978)
and in decreasing the sink activity (WAGNER 1974), shortage of nitrogen can
effectively regulate the source-sink interactions primarily at the expense of the
source. ABA, however, is also quite phloem-mobile and can be transported
from the leaves into grains, i.e. the sink (GOLDBACH et al. 1977), where it seems
to be causally involved in termination of the storage process, presumably by
inhibition of RNA synthesis in the embryo (Hsu 1979).
Of the other phytohormones causally involved in the sink activity and af-
fected by mineral nutrition, data are available only for GA. A close positive
correlation occurs between GA activity in shoots and nitrogen supply to roots
(RAJAGOPAL and RAo 1974). The stimulation of shoot growth at the expense
of root growth, which is a typical response to nitrogen supply (MOTHES 1932,
GORING and MARDANOV 1976), is presumably mainly the result of a higher
sink activity of various shoot parts with high GA concentrations (GARROD
1974). In crop species where certain vegetative parts are considered as "yield",
such as storage roots of Daucus carota, this phytohormone-regulated sink com-
petition is quite obvious (Table 19). It is a well-known phenomenori that a
high nitrogen supply in the later stages of growth induces vigorous shoot growth
at the expense of storage root growth (e.g. in carrots or sugar beet). This effect
is presumably caused by a shift in the phytohormone balance as discussed above,
with corresponding effects on the source-sink-relationship and sink competi-
tions (MICHAEL et al. 1974).
In the source-sink relationship the important steps, regulated at least in
the sink by phytohormones (SETH and WAREING 1967, WAGNER 1974), are
36 H. MARSCHNER:

Table 20. Role of potassium and magne-

sium in sucrose accumulation in vacuoles nmol
isolated from red beet tissue. (DOLL et al. sucrose h- 1
1979)
4.9
+ 42.3
+ + 55.,5

Table 21. Time course of the potassium content (mmol g-l dry wt.) in petioles of two
tomato varieties. {After LINGLE and LORENZ 1969}

Variety Growing stage of the plants

6 true 3rd cluster 1st cluster 1st cluster 50%

leaves full bloom mature green fruit pink fruits ripe

Pearson 1.47 1.64 1.85 0.78 0.39

VF-13 L 1.47 1.41 1.58 0.54 0.10'

a Severe potassium deficiency symptoms

phloem loading at the source, long-distance transport in the sieve tubes and
unloading at the sink. The involvement of potassium in the sucrose-proton
cotransport at the loading site is still an open question (HUTCHINGS 1978, DOMAN
and GEIGER 1979). It seems more likely, however, that the favorable effect
of potassium on the export of sucrose from the leaves (MENGEL and HAEDER
1977) is primarily the result of a higher rate of sucrose release from the leaf
cells into the apoplast and thus an indirect support to phloem loading (DOMAN
and GEIGER 1979). A direct role of potassium on sucrose transport within the
sieve tubes has still to be elucidated (WILLENBRINK and SCHUSTER 1978). On
the other hand, a direct role for potassium has been demonstrated in the storage
cells itself (Table 20). This is presumably a reflection of a Mg2 + + K + stimulated
ATPase at the inner side of the tonoplast.
From the regulation of starch synthesis in chloroplasts by inorganic phos-
phate (PJ as demonstrated by HELDT et al. (1977) (see Sect. 1), it may be sup-
posed that a comparable effect also occurs in the amyloplasts of starch-storing
cells. The marked increase in formation of phytin in cereal grains in later stages
of grain growth may thus perhaps be regarded as necessary in order to maintain
a fairly constant concentration of Pi under conditions of decreasing grain water
content (MICHAEL et al. 1980), and thus avoid any increase in inhibition of
starch synthesis by Pi.
Sink activity of fruits or storage organs and thus final yield can also be
limited in quite another way, namely by the supply of mineral elements from
the leaves as source. An excellent example for this limitation has been shown
by LINGLE and LORENZ (1969) comparing tomato genotypes differing in ripening
pattern (Table 21). In the genotype which ripened more rapidly and uniformly,
severe potassium deficiency symptoms occurred in the leaves. This deficiency
could not be corrected by supplying a high concentration of potassium to the
roots, presumably because of the dominance of the fruits which were also acting
as sink for carbohydrates and thus decreasing root growth.
1.1 General Introduction to the Mineral Nutrition of Plants 37

A similar situation of source limitation caused by shortage of another plant

nutrient has been demonstrated by TROBISCH and SCIDLLING (1969, 1970). In
Sinapis alba, the developing seeds compete for nitrogen supplied from the leaves,
and seed growth and final yield are primarily determined by the pool size of
nitrogen in the vegetative parts. Consequently, additional nitrogen application
at flowering markedly increases seed number and yield. Source limitation by
shortage of nitrogen but not carbohydrates was also shown when the effect
of darkening the leaves was compared with that of removal of the leaves. Dark-
ening reduced seed weight by 20%, but removal of the leaves, which also
removed the source for nitrogen, depressed seed weight by nearly 50% (TRO-
BISCH and SCHILLING 1969).
A special type of source limitation can occur in legumes particularly with
the onset of seed growth, where a sink competition for carbohydrates arises
between seeds and root nodules. Seeds seem to be the dominating sink and
are responsible for the rapid fall in N 2 fixation in nodules which occurs with
the onset of seed growth (HERRIDGE and PATE 1977). Thus also in legumes
seed growth can be limited by the amount of nitrogen available in the leaves,
particularly when senescence is enhanced by the export of bound nitrogen from
the leaves (SINCLAIR and DEWIT 1976). Supplying inorganic nitrogen after flow-
ering to the roots can therefore also be quite a useful means of increasing
the protein content of legume seed (SCHILLING and SCHALLDACH 1966); leaf
spray during seed filling with nitrogen in combination with other nutrients
is particularly effective in increasing yield in soybean (RAMON GARCIA and
HANWAY 1976).
These examples showing the interactions between mineral nutrition and the
source-sink relationship clearly indicate that this particular area needs much
more attention in future research. This holds true especially for direct or indirect
effects of plant nutrients other than nitrogen on phytohormone levels. More
attention needs to be given to the consequences arising from restricted mineral
nutrient uptake by the roots in the period when the fruits or storage organs
are established as dominating sinks. Selection and breeding for genotypes with
a "high harvest index" and short periods for fruit growth or ripening (" filling
period" in cereals for example) might become more and more difficult, not
because of a limited source capacity for carbohydrates but rather by limitation
of mineral nutrients such as potassium, magnesium, nitrogen or phosphorus.
This limitation could be at the source and due to a lack of capacity (amounts)
or the capability for retranslocation, or it could be caused by the low activity
of roots for uptake during the period of fruit and storage organ formation.
For further details the reader is referred to Chap. IV.2.

5 Environmental Aspects of Mineral Nutrition

5.1 Introduction

Mineral nutrient cycles in terrestrial natural ecosystems are never entirely closed,
but only reach a quasi-equilibrium between loss and supply. Contributing factors
of this system include the atmosphere, ground and surface water, weathering
38 H. MARSCHNER:

of rocks and minerals, .and biological dinitrogen fixation. Man has influenced
this natural cycle of plant nutrients by removing mineral nutrients in the process
of crop production. In "classical agriculture" - which still occurs in many
developing countries where the population is mainly rural- the" loss" of miner-
al nutrients has been relatively small. With increasing urbanization, however,
this "loss" has markedly increased and has had to be compensated for by
the introduction of legumes in the crop rotation and an increase in the applica-
tion of mineral fertilizers (for more details see Chap. 1.8). These changes simulta-
neously have led to the "production" of large amounts of "waste material"
in urban areas and increases in mineral nutrient concentration in ground and
surface waters.
This development, considered together with the limited resources (e.g. phos-
phate deposits) and increasing costs for the production of mineral fertilizers
(nitrogen in particular), strongly indicates a need for a more effective utilization
of plant nutrients in crop production. This holds true in particular for increasing
the contribution of biological dinitrogen fixation (Chap. II.1-3), an improve-
ment in the utilization of mineral nutrients by selection and breeding of crop
plants for higher efficiency in general (Chap. IV.1) and for phosphate in particu-
lar (Chap. 1.2) and an increase in the "re-cycling" of nutrients especially in
densely populated urban areas (see below).

5.1.1 Nitrogen
The environmental aspect of mineral nutrition, which at present attracts particu-
lar interest and is also the subject of most controversial discussions, is the
application of nitrogenous fertilizers either in organic (e.g. manure) or inorganic
forms (e.g. nitrate- or ammonium-N). The potential losses of nitrogen from
soils by leaching of nitrate, denitrification or volatilization of NH3 and hence
a corresponding increase in pollution of groundwater and atmosphere are cer-
tainly enhanced when nitrogen fertilizers are added to the soil. In agroecosystems
both total loss and type ofloss of bound nitrogen are, however, more dependent
upon climatic and soil conditions and factors such as the supply/demand ratio,
timing of fertilizer application, crop rotation etc. than whether the nitrogen
has been supplied in inorganic or organic form (for further details see a review
by FRISSEL 1977). In this context it is necessary to stress that these potential
losses of nitrogen from agroecosystems are not necessarily decreased, for
example when the contribution of biological dinitrogen fixation is increased
at the expense of the use of nitrate fertilizers. It must be remembered that
the application of nitrate as a fertilizer can be better adjusted to the actual
demand of plants, whereas mineralization of organically bound nitrogen is regu-
lated primarily by environmental factors such as temperature and soil moisture.
As nitrate is an ideal storage form of nitrogen for higher plants, high concen-
trations of nitrate in the soils not only increase potential losses by denitrification
and leaching but can also lead to high nitrate levels in the plants. This occurs
regardless of whether nitrate fertilizers (KNAUER and SIMON 1968) or organically
bound nitrogen (LUND et al. 1975) was the original form supplied to the soil.
In plants used for human consumption (vegetables such as spinach for example),
1.1 General Introduction to the Mineral Nutrition of Plants 39

this nitrate accumulation is undesirable for plant quality. Unskilled handling

or processing of plant material high in nitrate can cause nitrite formation, which
is potentially toxic to infants (SCHUPHAN 1976), and in addition can increase
the possibility of nitrosamine formation, a potentially carcinogenic substance
(HILDEBRANDT 1977). Thus control of nitrate formation in soils and uptake
by plants are important in both relationships of mineral nutrition and environ-
ment as well as mineral nutrition and plant quality. Therefore in soils it is
important to control both the amount of nitrate (WEHRMANN and SCHARPF
1979) and to restrict nitrification as, for example, by the use of naturally occur-
ring substances (SAHRAWAT et al. 1977) or synthetic compounds such as dicyan-
diamide (AMBERGER and GUTSER 1978).

5.1.2 Heavy Metals

Heavy metals are integrated components of the biosphere and thus occur natu-
rally in soils and plants. Five heavy metals (Fe, Mn, Zn, Cu, Mo) are essential
for all higher plants (see Sect. 1) and in addition others like V and Co are
essential for animal and man. The demand of these five heavy metals for plants
is rather low (micronutrients). However, the use of chemically pure fertilizers,
together with the production of higher yields, increases the necessity of supplying
these micronutrients more systematically in fertilizer practice in intensive agricul-
ture and horticulture. Except under soil conditions of low redox potential where
iron toxicity may occur in lowland rice (HOWELER 1974) and manganese toxicity
may appear in sensitive genotypes (see Chap. IV.1), deficiency is much more
widespread as a result of limited availability due to high soil pH (Fe, Zn,
Mn) or, as in the case of Cu, high organic matter content or extreme soil
leaching.
Increasing the supply of at least some or even all of these five heavy metals
in fertilizers is not only necessary to increase plant yield and to prevent deficien-
cies in plants, but is also important from the view point of plant quality. This
holds true for Zn in particular. There is increasing evidence that hidden or
acute Zn deficiency in both monogastric animals (WETZEL et al. 1979) as well
as man (HAMBIDGE and WALRAVENS 1976) is quite widespread, especially when
cereals are the main diet and where the availability of Zn for plants is low
due to the specific ecological conditions (e.g. high soil pH in arid/semi-arid
regions). Zinc deficiency in animals and man is furthermore accentuated by
inadequate availability from cereals. The presence of phytin, i.e. the main storage
product for phosphorus in cereal grain and other seeds, seems to be causally
responsible for this low Zn availability (REINHOLD et al. 1973). This fact has
led human nutritionists to the questionable classification of phytin into the
group of" phytotoxicants" (OVERLEAS 1973).

5.2 Heavy Metal Toxicity

With the exception of the occurrence of Fe and Mn toxicity in crop plants

as mentioned above, heavy metal toxicity was not generally regarded as an
40 H. MARSCHNER:

agricultural or horticultural problem. It was predominantly only a research

object for geobotanists and plant ecologists. The distinct "heavy metal flora"
in mining areas, adapted for example especially to high soil levels of Zn and
Cu, and the mechanisms of tolerance - either by exclusion, or inactivation
within roots or leaves by compartmentation, complex formation etc. - was
always a fascinating research subject on environmental aspects of mineral nutri-
tion (KRAUSE 1958, ERNST 1974). The reports of tolerance of crop plants grown
in these mining areas (or on serpentine soils) with high levels of Zn, Cu and
also with Ni or Cr were quite rare (for further information see ANTONOVICS
et al. 1971, ERNST 1974) and presented without considering the possible implica-
tions and risks which may arise when these plants are used for animal and
human nutrition. An exception to this was the study of heavy metal toxicity
for animals and man as a result of direct exposure particularly for miners
(e.g. lead, mercury). However, it was only a decade ago that the role of elevated
heavy metal contents of plants as part of the food chain for heavy metal toxicity
of animals and man suddenly became of overwhelming importance. The" classi-
cal" example for this was the occurrence of the human sickness Itai-Itai in
Japan, caused by cadmium (Cd) released from industrial areas and passed at
least in part from the soil into the food chain via plants (YAMAGATA and SmGE-
MATSU 1970). This" accident" stimulated intensive world-wide research on envi-
ronmental aspects of heavy metals in the mineral metabolism of plants.

5.3 Heavy Metals in the Food Chain

In the first edition of this Encyclopedia in Volume IV (Mineral Nutrition)

published in 1956, the heavy metal Cd was not even mentioned, and others
like Pb, Ni or Hg only briefly discussed. Since then the position has changed
dramatically and it is now extremely difficult to keep up with the hundreds
of publications which appear each year on Cd and other heavy metals in the
soil/plant/animal/man system. It can therefore only be the aim of this section
to outline this development and the consequences arising from it, both for
future research on mineral nutrition and strategies to control entry of heavy
metals in the food chain via the soil and their subsequent transfer into plants,
animals and finally to man. An example illustrating the accumulation of Cd
in the food chain is shown in Table 22.
Considering that only part of the Cd in the top soil is immediately available
to plants the progressive increase in Cd accumulation in the food chain is obvi-
ous and can reach particularly high levels (more than 300 ppm) in the kidneys
of carnivores in contaminated areas (MARTIN and COUGHTREY 1975). This posi-
tive correlation between the Cd content in soils/plants and animals generally
occurs (CHANEY et al. 1978a), but not always (Table 23).
The increase in Cd content of the plants and thus the diet from the MSS-
amended soils did not increase the Cd content of the kidneys, presumably as
a result of restricted Cd resorption in the presence of the simultaneously elevated
Zn levels in the diet (CHANEY et al. 1978b).
These two examples stress the necessity for enhanced interdisciplinary re-
search on heavy metals in the food chain. The aim of this section is to present
1.1 General Introduction to the Mineral Nutrition of Plants 41

Table 22. Levels of Cd (ppm in dry matter) in various com-

ponent species of woodland ecosystems. (After MARTIN and
COUGHlREY 1975)

Collecting areas

Relatively Contaminated
uncontaminated (3 Ian from a
Pb-Zn smelter)

Top soil 2.0 42

Moss 1.3 25
Hedera helix 2.1 18
Earthworms 25.2 123

Table 23. Tissue content (ppm in dry matter) of Cd, Ni and Zn

in Guinea pig fed with plants grown on municipal sewage sludge
(MSS)-amended soil. (After CHANEY et al. 1978b)

Cd Ni Zn -

Soil (extractable)
Control 0.02 0.4 1.6
MSS 0.30 1.1 65.0

Plants (Swiss chard)

Control 0.5 1.4 70
MSS 2.7 3.5 580
Kidneys
Control 14.9 0.15 (No significant
MSS 14.9 0.05 differences)

some examples from the soil/plant system with special references to Cd, which
is taken up quite readily by many crop species and which is one of the most
toxic heavy metals for animals and man. For more detailed information the
reader is referred to PAGE and BINGHAM (1973), VETTER et al. (1974), DOYLE
(1977) and CHANEY and HORNICK (1978).

5.4 Heavy Metals in the Soil/Plant System

5.4.1 Content of Soils

The average concentration of Cd in the Earth's crust is 0.15 ppm. Depending
upon the parent material, the Cd content of soils, however, can vary consider-
ably: 0.1-0.3 ppm in soils deriving from igneous rocks and 0.1-10 ppm and
0.3-11 ppm respectively in soils deriving from metamorphic and sedimentary
rocks (WAKITA and SCHMITT 1970). As Cd is usually associated with Zn, Pb-Zn,
and Pb-Zn-Cu ores, the Cd content of soils in these mining areas can be several
times higher than in average soils (PAGE and BINGHAM 1973, ERNST 1974).
42 H. MARSCHNER:

Table 24. Heavy metal content of soils and pasture plants at different distances from
a lead-zinc smelter. (VETTER et a1. 1974)

Element Distances in Ian

0.75 1.5 3 6 12

Amount kg ha -1 in 0-80 em soil depth

Zn 7500 1850 620 350 240
Pb 2700 730 360 160 130
Cd 71 14 4.0 1.9 0.8
ppm in pasture plants (dry wt.)
Zn 523 118 116 74 78
Pb 98 55 11 12 7
Cd 2.1 1.1 0.6 0.4 0.2

Of much more importance, however, are the man-made elevated environ-

mental levels of Cd, and in particular from emissions from smelting and refining
industries (Zn, Pb, Cu). Depending upon the prevailing direction of winds,
elevated levels of heavy metals in soils can be detected even up to distances
of 40-65 km from a smelter (CARTWRIGHT et a1. 1976). In the immediate neigh-
borhood of large smelters soils and plants (uptake from the soil and adhering
dust on the leaves) can reach heavy metal levels which make agricultural produc-
tion impossible, not primarily because of toxicity to the plants but rather of
the hazards which might arise from using these plants for animal and human
nutrition. This situation is accentuated because in these areas contamination
by several heavy metals usually occurs simultaneously (Table 24).
Another important site of contamination of soils and plants with heavy
metals, and particularly Pb and Cd, occurs along roadsides. The source of
Pb is the lead-tetraethyl in petrol, whereas the Cd content is mainly due to
attrition of tires (LAGERWERFF and SPECHT 1970). The Pb content of soils and
plants along highways can reach 100 ppm (KLOKE et a1. 1966). However, most
of this Pb content of roadside plants is not the result of uptake from the soil
but rather of surface contamination (HOLL and HAMPP 1975) as the availability
of Pb in soils for uptake by plants is usually rather low.
Cd is also present in phosphate fertilizers. Depending upon origin, the Cd
content ofrock phosphates can vary between 9 and 130 ppm (REuss et a1. 1978).
Amounts of Cd supplied to soils with fertilizer phosphates thus depend largely
on the type of rock phosphate, but also on the processing procedure (CHANEY
and HORNICK 1978, lUNG et a1. 1979). Amounts added to the soil vary from
less than 1 g and up to 100 g Cd ha -1, based on a fertilizer rate of 50 kg
P ha - 1. Although only a small proportion of this Cd is immediately available
for uptake (lUNG et a1. 1979), the positive correlation between added phosphate
fertilizer (and thus also Cd) and the Cd content of plants can clearly be demon-
strated (REUSS et a1. 1978). As a consequence of this, long-term calculations
on world reserve of rock phosphates and comparisons between different types
1.1 General Introduction to the Mineral Nutrition of Plants 43

of phosphate fertilizer processing have to take Cd into account and its implica-
tion for the environment.
The main concern of increasing the heavy metal content of soils and thus
also the food chain is related to the use of municipal liquid and solid sewage
sludge. Here, clearly contradictory environmental aspects and goals exist, as
it is also necessary to restrict enlargement of urban trash disposal, prevent
further contamination of surface and ground waters and finally to bring about
a more efficient recycling of organic matter and plant nutrients also in agroeco-
systems with predominantly urban populations. However, in industrialized areas
in particular, the contamination of municipal liquid and sewage sludge with
heavy metals has reached levels which prohibit even short-term usage in agricul-
ture and horticulture. Again, here Cd is of major concern and a rather recent
problem, as is indicated by the production and usage of this element (electroplat-
ing, pigments and chemicals) which increased in world-wide usage more than
25-fold from 1920 to 1960 (PAGE and BINGHAM 1973). The increased usage
of municipal liquid and solid sewage sludge, together with contamination by
other heavy metal sources (see above), has resulted in large areas having elevated
heavy metal contents in soils and plants. This has stimulated intensive research
on the fate of these heavy metals in soils and the uptake and distribution in
plants.

5.4.2 Soil Factors Affecting Heavy Metal Accumulation in Plants

Similarly to other heavy metal cations such as Mn and Zn, the availability
ofPb (ZIMDAHL and FOSTER 1976) and Cd (HERMS and BRUMMER 1979) decreases
with increasing pH. The availability to plants also decreases with increasing
cation exchange capacity of the soil. In contrast to Zn or Mn, the binding
of Cd and Pb is closely related to the organic matter content of the soil (ANDERS-
SON 1977b). This can lead to a decrease in Pb uptake by plants despite an
increase in Pb supply to the soil by Pb-contaminated compost (ANDERSSON
1977a).
This effect of organic matter on plant availability ofPb has been also demon-
strated for Cd in a model experiment (Fig. 8).
Interactions between heavy metal availability and soil organic matter are
complicated by changes in redox potential and by the formation of metal-organic
complexes of different electrical charges (LAGERWERFF and MILBERG 1978). In
general a decrease in redox potential increases the possibility of H 2 S formation
and thus precipitation and immobilization of heavy metals such as Zn or Cd
(HERMS and BRUMMER 1979). In the absence of drastic decreases in redox poten-
tial, the effect of organic matter in decreasing availability by formation of com-
plexes seems to decrease in the order Cu > Pb > Zn > Cd (LAGERWERFF et al.
1977).
These few examples demonstrate that predictions concerning the contamina-
tion of the food chain by heavy metals require careful consideration, not only
of the soil factors, but also of whether a heavy metal such as Cd is supplied
as an inorganic cation (SYMEONIDES and McRAE 1977) or as a contaminant
of sewage sludge along with high contents of organic matter.
44 H. MARSCHNER:

8 A B
24
4
. 20
:::E
ci
.!; 16
:<:Ol 4
.~
IS.
Co
12

>.
6 2
o/lo 4
A
0
00 5 10 15 00 5 10 15 20
Cd applied (ppm) Cd applied (ppm)

Fig. 8. Effect of organic matter (0%-4% as peat) addition to the substrate (quartz sand)
on growth of soybean plants (A) and cadmium content of the leaves (B) at increasing
cadmium concentrations in the substrate. (STRICKLAND et al. 1979)

5.4.3 Genotypic Differences in Heavy Metal Uptake

In crops, appreciable genotypic differences exist in uptake of heavy metals such
as Cd, Ni, Hg or Cr, although these are less dramatic than in the case of
the "heavy metal flora" for example from serpentine soils or mining areas
(KRAUSE 1958, ANTONOVICS et al. 1971, ERNST 1974). With a few exceptions
(see below), Cd is the one of these four heavy metals which in general is most
readily taken up and translocated into the shoots (DIEZ and ROSOPULO 1976).
Nevertheless, pronounced genotypical differences also exist for Cd in accumula-
tion between crop species. These differences have been summarized by CHANEY
and HORNICK (1978) and are shown in Fig. 9.
Compared to these extended and systematic studies of Cd uptake by different
crop species much fewer comparative data are available for other elements
such as Ni, Cr or Hg. The uptake of Ni seems to be astonishingly high in
the vegetative parts of a number of species (DIEZ and ROSOPULO 1976) from
9 ppm in oats up to 21 ppm in wheat. In vegetables, particularly high Ni accumu-
lation has been found in lettuce (FRITZ et al. 1977), cabbage (HARA and SONODA
1979) and spinach (FOROUGHI et al. 1979). Of the vegetable crops spinach (KICK
and BRAUN 1977) or bean seem to have a particular ability to accumulate Cr
in the shoots as compared to lettuce (HUFFMAN and ALLAWAY 1973).
In higher plants, accumulation of Hg is usually quite low even in heavily
contaminated soils (DIEZ and ROSOPULO 1976). On the other hand amongst
the lower plants certain fungi (Basidiomycetes) are distinct Hg accumulators
(DOMSCH et al. 1976).
Certain of these fungi also have a high affinity for Cd. Collected from
an uncontaminated area in Yugoslavia, the range of Cd (ppm in dry matter)
varied from 0.62 in Cantharellus uberius up to 5.72 in Macrolepiota and 10.30
in Lactarius piperatus (BYRNE and RAVNIK 1976). Between species of the genus
Agaricus (champignon) a particular high variability occurs in Cd accumulation.
As compared to the Zn content in the dry matter of the fungi (300-500 ppm),
the Cd content varied between 0.1 and 170 ppm with oorresponding Cd accumu-
1.1 General Introduction to the Mineral Nutrition of Plants 45

paddy rice
upland rice
:::::I'sudangrass
:::::I w. clover
2, alfalfa
::::::::::J. bermudagrass
gfield bean
~wheat _ Denotes grain. fruit. or edible root
B zucchini spuash
-=:J soybean
~ tall fescue
corn
carrot
cabbage
radish
swiss chard
redbeet
r. lettuce
tomato
curlycress
spinach
turnip

o 20 IIJ 60 80 100 120 140' 160 180

Cadmium in plant tissue. ppm dry matter

Fig. 9. Cadmium uptake by different crop species grown on a soil enriched with 10 ppm
cadmium by sewage sludge. (After BINGHAM et a1.1975, 1976, from CHANEY and HORNICK
1978)

lation ratios (fungi/soil) between 3.1 in A. abruptibulbus and 292.5 in A. macro-

sporus (MEISCH et al. 1977). Because of the generally high level of Cd and Hg
in edible fungi, the excessive consumption should be avoided even of fungi
grown on soil not contaminated by man (LORENZ et al. 1978).
Despite the benefical effects of vanadium (V) on plant growth (see Chap.
V.5) and its functions in human and animal metabolism, it is still an open
question whether, for example, the excessive accumulation of V in certain species
of Allium of up to more than 100 ppm in the dry matter (CANNON 1963) com-
pared to less than 1 ppm in the dry matter in most other plant species (CANNON
1963) is harmful in human and animal nutritipn. Most of the V is supplied
to soils by phosphate fertilizers which can contain up to 5000 ppm V (for more
details see LASKE 1979).
The classification of plant species into heavy metal accumulators and non-
accumulators as has been shown for Cd in Fig. 9 has to be reexamined critically
in view of the varietal differences within certain species. An impressive example
for this is shown with lettuce (Fig. to). In plants grown on the same substrate
the Cd content of the tops varied by the factor 20.
Cadmium contents in the shoots of soybean varieties grown on the same
soil even varied by a factor of 40 (BOGGESS et al. 1978). Also in cereals such
as barley and wheat the varietal differences in Cd accumulation of the seeds
are considerable and consistent even when the varieties are grown on different
soils where both the amounts and binding states of Cd can be expected to
be different (PETTERSSON 1977).
These varietal differences in Cd uptake fit well in the general picture of
genotypic differences in mineral nutrition of higher plants (see Chap. IV.1 and
46 H. MARSCHNER:

Fig. 10. Cadmium content in

~ the tops of nine lettuce cultivars
2~ grown in solution culture with
0.1 ppm cadmium. (JOHN and
... 3 ~
c VAN LAERHOVEN 1976)
1; 4 ~
:;
l.) 5

.2'"
Q;
6
...J 7 'L. Z Z'll

8 '//..1

9 '////////////,'///, '//,Z Z '/1

2 3 4 5 6 7 8 9
ppm Cd in tops

Table 25. Average values for heavy metal content (ppm in dry matter) in different organs
of vegetables grown on various soils. (FRITZ et al. 1977)

Organs Cu Zn Pb Cr Cd Ni

Leaves 11.3 118 13.3 3.9 1.0 5.9

Storage roots, tubers 11.0 43 12.1 1.3 0.1 3.9
Fruits 9.1 40 9.7 0.1 0.1 2.7

Vol. 2/B, Chap.III.9, this Vol.) and open up new aspects for selection and
breeding programmes for low uptake of certain heavy metals like Cd.

5.4.4 Distribution Within the Plants and Their Organs

For heavy metal contamination of the food chain not the total uptake is of
primary importance; but rather the content in those parts which are used directly
for animal and human consumption. In many crop species, for example, this
includes only the fruits, seeds or storage roots, whereas in others it may consist
of the whole shoot (e.g. lettuce). As a general rule the heavy metal content
per unit dry matter decreases in the order leaves, storage roots, tubers, fleshy
fruits and seeds. An example from a range of vegetable crops is shown in
Table 25.
The generally high Pb content in all organs is presumably mainly caused
by surface contamination. In the case of Ni, however, high contents are found
even in the fruits in comparison to the leaves and appear to reflect a particularly
high mobility of this heavy metal in phloem transport. Nickel is readily translo-
cated also into seeds (Table 26). This has also been demonstrated in wheat,
where with increasing Ni content of the soil a similar increase in Ni content
of leaves and seeds has been reported (MITCHELL et al. 1978).
Fortunately this distribution between leaves and fruits or leaves and seeds
is much more in favour of the leaves for Cd than for Ni. On the other hand
in most plant species uptake and transport of Cd into the shoots is usually
1.1 General Introduction to the Mineral Nutrition of Plants 47

Table 26. Distribution of nickel with time fol-

lowing 24 h uptake from 1.0 J.lM NiCl z by Tissue Distribution (%) following
94-day-old soybean plants. (After CATALDO 63Ni2+ pulse
et al. 1978) 24h 21 days

Roots 85 51
Stems 5 4
Leaves 6 4
Pods 2 2
Leeds 2 39

10

~ period of cadmium supply-

.~ 8 cadmium-free period

Fig. 11. Time course of initial

cadmium uptake by tomato
plants (Lycopersicon esculentum)
based on successive runs of par-
tial cadmium depletion from nu-
trient solutions containing "0 "0
UU
O.1IlM cadmium. Each period of
cadmium supply ( - ) is separated
by a period with a cadmium-free
solution (- - -). (PETIT et al.
1978)
10 20 30 40 50 60
time (h)

directly proportionally to the external concentration (PETTERSSON 1976). In a

detailed study on Cd uptake by tomato plants grown under controlled condi-
tions, PETIT et al. (1978) found evidence (Fig. 11) that supplying Cd to the
roots induces the formation of new binding sites for Cd within the roots and
thus further stimulates uptake. This mechanism needs further attention and
investigation, and comparisons should be made with other plant species and
genotypes differing in their ability to take up Cd.
Although the translocation of Cd from leaves into seeds as compared to
fleshy fruits is usually low and the Cd content of cereal grains only exceeds
1 ppm on highly contaminated soils (PIETZ et al. 1978), in the average diet
of U.S. citizens for example more than 50% of the total Cd intake is derived
from cereals (CHANEY and HORNICK 1978). As the distribution of Cd at least
in wheat grains is rather uniform, the possibility of decreasing the Cd content
by processing procedures is limited (OCKER 1977/1978).
In addition to lowland rice (Fig. 9) maize appears to be the other main
cereal crop in which is a distinct restriction of Cd translocation into the seeds
(PIETZ et al. 1978). However, as already shown for Cd in lettuce (Fig. 10), experi-
ments with only one variety do not permit generalizations on the beha\Tior
of Cd in a species. For example, when 20 inbred lines of maize were grown
48 H. MARSCHNER:

on a sludge-amended soil, the Cd content of the seeds (mg Cd kg - 1 dry matter)

varied between 0.08 in one line to 3.87 in another (HINSLEY et al. 1978).

5.4.5 Heavy Metal Tolerance

The increased use of municipal sewage sludge on soils and the subsequent in-
crease in heavy metal content of crops and soil has initiated extensive compara-
tive studies between the heavy metal toxicity and possible yield reduction. As
may be exp~cted, results vary widely depending on experimental conditions.
The choice of plant species and varieties, soil type, soil pH, form of heavy
metal (e.g. inorganic Cd2+ as compared to Cd in sewage sludge), composition
of the nutrient solution, duration of the experiment etc. all influence the results
obtained. For details of results of some comparative studies in nutrient solution
using increasing concentrations of a range of heavy metals, the reader is referred
for example to FOROUGHI et al. (1979) and HARA and SONODA (1979). Most
papers, however, are exclusively concerned with uptake and phytotoxicity of
Cd as this heavy metal, as compared to Pb, Ni, Hg or Cr, is particularly toxic
for most crop species. Of the vegetables, spinach is much more Cd-sensitive
than, for example, tomato (BINGHAM et al. 1975). Cadmium toxicity in plants
is often correlated with chlorosis and wilting, symptoms which might be related
to inhibited Fe uptake (FOROUGHI and VENTER 1978) or the inhibition of xylem
differentiation and thus a disturbance of the water balance of the plants (LAMO-
REAUX and CHANEY 1977, KIRKHAM 1978).
For causal interpretations of genotypic differences in heavy metal tolerance,
it is necessary to obtain more information concerning the mechanisms involved,
such as exclusion in uptake, restricted transport into the shoot or high tissue
tolerance (e.g. physiological "inactivation" by compartmentation). Research
on tolerance mechanisms for Cd, Ni, Cr or Pb of crop plants is certainly neces-
sary and should utilize the extensive knowledge of heavy metal tolerance in
the natural vegetation as for Zn and Cu (MATHYS 1975, Wu et al. 1975, WAIN-
WRIGHT and WOOLHOUSE 1977) or Ni (LEE et al. 1978). The high mobility of
Cr or Ni in plants might be causally related to the formation of complexes
with organic acids (LEE et al. 1978) and explain, for example, the particularly
high rates of Ni transport into fruits and seeds of crop plants (see Tables 25
and 26).
Without doubt a better understanding of the tolerance mechanism for heavy
metals such as Cd in crop plants is also of importance for basic research in
plant physiology and biochemistry. It should be borne in mind, however, that
phytotoxity of Cd, for example, is irrelevant from the environmental viewpoint
in relation to heavy metal accumulation in the food chain. This discrepancy
in crop species between Cd toxicity as measured by yield reduction and Cd
accumulation in edible plant parts is shown in Table 27.
In future the strategy of selection and breeding programmes of crop species
has to concentrate not on increasing Cd tolerance but in restricting Cd uptake
and transfer into edible plant parts. The same strategy is necessary for forage
species where Cd tolerance see~s to be closely related to high Cd accumulation
within the shoot (BINGHAM et al. 1976).
1.1 General Introduction to the Mineral Nutrition of Plants 49

Table 27. Soil and plant tissue Cd levels associated with a 25% yield decrement. (After
BINGHAM et al. 1975)

Species Soil Cd levels Tissue Cd levels at 25%

producing yield decrement
a 25% yield decrement
Edible parts (leaves)
ppm Cd ppm Cd

Spinacea oleracea 4 75
Glycine max. 5 7 (7)
Lepidium sativum 8 80
Lactuca sativa 13 70
Zea mays 18 2 (35)
Daucus carota 20 19 (32)
Brassica rapa 28 15 (121)
Phaseolus vulgaris 40 2 (15)
Triticum aestivum 50 11 (33)
Raphanus sativus 96 21 (75)
Lycopersicon esculentum 160 7 (125)
Brassica oleracea 170 11
Oryza sativa 640 2 (3)

5.5 Concluding Remarks

Concern about the accumulation of certain heavy metals in the food chain,
with its consequent potential danger for human and animal nutrition is fully
justifiable. Interdisciplinary research programmes are necessary, as well as gov-
ernmental restrictive measures for the better control of Cd contamination at
the source site, including the use of municipal sewage sludge. It is necessary,
however, on the other hand to bear in mind that heavy metals are natural
components of all ecosystems and cannot be simply considered as environmental
pollutants. Increasing the amounts and availability of essential heavy metals
such as Zn, Fe, Mo or Mn in soils and plants can be important for both
increased plant production and plant quality for animal and human nutrition
(see Sect. 1). Research and discussion on heavy metals have to differentiate
between these various aspects and take this into consideration in comparative
evaluations.
Acknowledgement. The author thanks Mr. E.A. KIRKBY for valuable comments on the
manuscript.

References

Allen RD (1969) Mechanism of the seismonastic reaction in Mimosa pudica. Plant

PhysioI44:1101~1107
Amberger A, Gutser R (1978) Umsatz und Wirkung von Harnstoff-Dicyandiamid sowie
Ammonsulfat-Dicyandiamid-Produkten zu Weidelgras und Reis. Z Pflanzenernaehr
Bodenkd 141: 553-566
50 H. MARSCHNER:

Andersson A (1977a) The distribution of heavy metals in soils and soil material as
influenced by the ionic radius. Swed J Agric Res 7: 79-93
Andersson A (1977b) Some aspects on the significance of heavy metals in sewage sludge
and related products used as fertilizers. Swed J Agric Res 7: 1-5
Antonivics J, Bradshaw AD, Turner RG (1971) Heavy metal tolerance in plants. Adv
Ecol Res 7: 1-85
Armstrong MJ, Kirkby EA (1979) The influence of humidity on the mineral composition
of tomato plants with special reference to calcium distribution. Plant Soil 52: 427-435
Arnon DI (1950) Criteria of essentiality of inorganic micronutrients for plants. In:
McElroy WD, Glass B (eds) Trace elements in plant physiology. Chronica Botanica,
Waltham, Mass
Bangerth F (1976) A role of auxin and auxin transport inhibitors on the Ca content
of artificially induced parthenocarpic fruits. Physiol Plant 37: 191-194
Bangerth F (1979) Calcium-related physiological disorders of plants. Annu Rev Phyto-
pathol1 :97-122
Bangerth F, Mostafawi M (1969) EinfluB der Wasserversorgung und des Fruchtgewichtes
auf den Mineralstoffgehalt und die Stippigkeit von Apfelfriichten. Erwerbsobstbau
11:101-104
Bangerth F, Dilley DR, Dewey DH (1972) Effect of postharvest calcium treatment on
internal breakdown and respiration of apple fruit. J Am Soc Hortic Sci 97: 679-682
Barber SA (1979) Growth experiments for nutrients in relation to demand at the root
surface. In: Harley JL, Scott Russell R (eds) The soil-root interface. Academic Press,
London New York
Bateman DF, Lumsden RD (1965) Relation between calcium content and nature of
the pectic substances in bean hypocotyls of different ages to susceptibility to an
isolate of Rhizoctonia solani. Phytopathology 55: 734-738
Bergmann W, Neubert P (1976) Pflanzendiagnose und Pflanzenanalyse. Fischer, Jena
Bhat KKS, Nye PH (1974) Diffusion of phosphorus to plant roots in soil. III. Depletion
around onion roots without root hairs. Plant Soil 41 :383-394
Bingham FT, Page AL, Mahler RJ, Ganje TJ (1975) Growth and cadmium accumulation
of plants grown on a soil treated with a cadmium-enriched sewage sludge. J Environ
Qual 4: 207-211
Bingham FT, Page AL, Mahler RJ, Ganje TJ (1976) Yield and cadmium accumulation
of forage species in relation to cadmium content of sludge-amended soil. J Environ
Qual 5: 57-60
Boggess SF, Willavize S, Koeppe DW (1978) Differential response of soybean varieties
to soil cadmium. Agron J 70: 756-760
Bould D, Parfitt RI (1973) Leaf analysis as a guide to the nutrition of fruit crops.
X. Magnesium and phosphorus sand culture experiments with apple. J Sci Food
Agric 24:175--185
Bradfield EG (1976) Calcium complexes in the xylem sap of apple shoots. Plant Soil
44 : 495--499
Bradfield EG, Guttridge CG (1979) Dependence of calcium transport and leaf tipburn
in strawberry on relative humidity and nutrient solution concentration. Ann Bot
43:363-372
Brewster JL, Tinker PB (1970) Nutrient cation flows in soil around plant roots. Soil
Sci Soc Am Proc 34: 421-426
Brown JC (1978) Mechanism of iron uptake by plants. Plant Cell Environ 1 :249-257
Brownell PF (1979) Sodium as an essential micronutrient element for plants and its
possible role in metabolism. Adv Bot Res 7: 117-224
Brownell PF, Jackman ME (1966) Changes during recovery from sodium deficiency
in Atriplex. Plant Physiol 41: 617-622
Brownlee C, Kendrick RE (1979) Ion fluxes and phytochrome protons in mung bean
hypocotyl segments. I. Fluxes of potassium. Plant Physiol64:206-210
Broyer TC, Carlton AB, Johnson CM, Stout PR (1954) Chlorine - a micronutrient
element for higher plants. Plant PhysioI29:526-532
Bruinsma J (1977) Rolle der Cytokinine bei Bliiten- und Fruchtentwicklung. Z Pflanzener-
nahr Bodenkd 140: 15--23
1.1 General Introduction to the Mineral Nutrition of Plants 51

Brumagen DM, Hiatt AH (1966) The relationship of oxalic acid to the translocation
and utilization of calcium in Nicotiana tabacum. Plant Soil 24: 239-249
Buban T, Vaega A, Traub J, Kuegt E, Bruinsma J (1978) Effects of ammonium and
nitrate nutrition on the level of xylem sap of apple rootstocks. Z Pflanzenphysiol
89:289-295
Byrne AE, Ravnik V (1976) Trace element concentration in higher fungi. Sci Total Envi-
ron 6:65-78
Campell LG, Miller MH, Loneragan JF (1975) Translocation of boron to plant fruits.
Aust J Plant PhysioI2:481-487
Campbell NA, Stika KM, Morrison GH (1979) Calcium and potassium in the motor
organ of sensitive plants: Localization by ion microscopy. Science 204: 185-187
Cannon HL (1963) The biochemistry of vanadium. Soil Sci 96: 196--204
Carmi A, Koller D (1979) Regulation of photosynthetic activity in the primary leaves
of bean (Phaseolus vulgaris L.) by materials moving in the water-conducting system.
Plant PhysioI64:285-288
Cartwright B, Merry RH, Tiller KG (1976) Heavy metal contamination of soils around
a lead smelter at Port Pirie, South Australia. Aust J Soil Res 15:69-81
Cataldo DA, Garland TR, Wildung RE, Drucker H (1978) Nickel in plants. II. Distribu-
tion and chemical form in soybean plants. Plant PhysioI62:566--570
Chaney RL, Hornick SB (1978) Accumulation and effects of cadmium on crops. In:
Proc 10t Int Cadmium Conf Met Bull London
Chaney RL, Stoewsand GS, Bache CA, Lisk DJ (1978 a) Cadmium deposition and hepatic
microsomal induction in mice fed lettuce grown on municipal sludge-amended soil.
J Agric Food Chern 26:992-994
Chaney RL, Stoewsand GS, Furv AK, Backe CA, Lisk DJ (1978b) Elemental content
of tissue of guinea pig fed swiss chard grown on municipal sewage sludge-amended
soil. J Agric Food Chern 26:994-997
Chen CH, Lewin J (1969) Silicon as a nutrient element for Equisetum arvense. Can
J Bot 47:125-131
Collier GF, Huntington VD, Cox EF (1979) A possible role of chlorogeneic acid in
calcium-related disorders of vegetable crops. Commun Soil Sci Plant Anal 10: 481-490
Cooper RB, Blaser RE, Brown RH (1967) Potassium nutrition effects on net photosynthe-
sis and morphology of alfalfa. Soil Sci Soc Am Proc 31 :231-235
Corden ME (1965) Influence of calcium nutrition of fusarium wilt of tomato and polyga-
lacturonase activity. Phytopathology 55:222-224
Crittenden HW, Svec CV (1974) Effect of potassium on the incidence of Diaporthe
sojae in soybean. Agron J 66: 696--698
Diez Th, Rosopulo A (1976) Schwermetallgehalte in Boden und Pflanzen nach extrem
hohen Kliirschlammgaben. Landwirtsch Forsch Sonderheft 22/1 :236--248
Dixon B, Sagar GR, Shorrocks VM (1973) Effect of calcium and boron on the incidence
of tree and storage pit in apples of the cultivar Egremont Russel. Hort Sci 48: 403-411
Doll S, Rodier F, Willenbrink J (1979) Accumulation of sucrose in vacuoles isolated
from red beet tissue. Planta 144:407-411
Doman DC, Geiger DR (1979) Effect of exogenously supplied foliar potassium on
phloem-loading in Beta vulgaris L. Plant PhysioI64:528-533
Domsch KH, Grabbe K, Fleckenstein J (1976) Schwermetallgehalte im Kultursubstrat
und Erntegut des Champignons, Agaricus bisporus (Lange) Singer, beim Einsatz von
Miillkliirschlammkompost. Z Pflanzenernaehr Bodenkd 139:487-501
Doyle JJ (1977) Effects of low levels of dietary cadmium in animals - a review. J Environ
Qual 6:111-116
Egmond van F, Houba VJG (1970) Production of carboxylates (C-A) by young sugar-beet
plants grown in nutrient solution. Neth J Agric Sci 18: 182-187
EI-Sheikh AM, Ulrich A, Broyer TC (1967) Sodium and rubidium as possible nutrients
for sugar beet plants. Plant Physiol 42: 1202-1208
Ende van der J, KoornneefP, Sonneveld C (1975) Osmotic pressure of the soil solution.
Determination and effects on some glass-house crops. Neth J Agric Sci 23: 181-190
Epstein E (1972) Mineral nutrition of plants: Principles and perspectives. Wiley and
Sons, New York
52 H. MARSCHNER:

Epstein E, Hagen CE (1952) A kinetic study of the absorption of alkali cations by

barley roots. Plant PhysioI27:457-474
Epstein E, Norlyn JD, Rush EW, Kingsburg RW, Kelley DB, Cunningham GA, Wrona
AF (1980) Saline culture of crops: A genetic approach. Science 210:399-404
Ernst W (1974) Schwermetallvegetation der Erde. Fischer, Stuttgart
Faust M, Klein JD (1974) Levels and sites of metabolically active calcium in apple
fruit. J Am Soc Hort Sci 99: 93-94
Faust M, Shear CB (1969) Biochemical changes during the development of cork spot
of apples. Qual Plant Mater Veg 19: 22~ 265
Ferguson IB, Bollard EG (1976) The movement of calcium in woody stems. Ann Bot
(London) 40:1057-1065
Fidanovski F (1869) Der EinfluB von Silicium auf Pflanzen. Diss, TU Berlin, D 83 Nr
256
Foroughi M, Venter F (1978) Die Wirkung unterschiedlicher Cadmiumgaben auf die
Eisenversorgung einiger Gemiisearten. Landwirtsch Forsch 31: 300-308
Foroughi M, Teicher K, Venter F (1979) Die Wirkung steigender Gaben von Blei, Cadmi-
um, Nickel oder Zink auf Spinat in Niihrlosung. Landwirtsch Forsch Sonderheft
35:599-606
Frissel MJ (1977) Cycling of mineral nutrients in agricultural ecosystems. Agro-Ecosyst
4:1-354
Fritz PD, Foroughi M, Venter F (1977) Schwermetallgehalte in einigen Gemiisearten.
Landwirtsch Forsch Sonderheft 33jII:33~343
Fuente de la RK, Leopold AC (1972) Two components of auxin transport. Plant Physiol
50:491-495
Garg OK, Sharma AN, Kona GRSS (1979) Effect of boron on the pollen vitality and
yield of rice plant (Oryza sativa L. var. Jaya) Plant Soil 52: 591-594
Garrod JF (1974) The role of gibberellins in early growth and development of sugar
beet. J Exp Bot 25:94~954
Gauch HG (1972) Inorganic plant nutrition. Dowden Hutchinson & Ross, Strauchburg,
Pa
Geijn van de SC, Petit CM, Roelfsen H (1979) Measurement of the cation exchange
capacity of the transport system in intact stems. Methodology and preliminary results.
Commun Soil Sci Plant Anal10:22~236
Goring H, Mardanov AA (1976) EinfluB von Stickstoffmangel auf das KjCa-Verhiiltnis
und den Zytokiningehalt junger Kiirbispflanzen. Biochem Physiol Pflanz 170: 261-264
Goring H, Thien BH (1979) Influence of nutrient deficiency on proline accumulation
in the cytoplasm of Zea mays L. seedlings. Biochem Physiol Pflanz 174:9-16
Goldbach E, Goldbach H, Wagner H, Michael G (1975) Influence of N-deficiency on
the abscisic acid content of sunflower plant. Physiol Plant 34: 138-140
Goldbach H, Goldbach E, Michael G (1977) Transport of abscisic acid from leaves
to grains in wheat and barley plant. Naturwissenschaften 64:488
Goor van BJ, Wiersma D (1974) Redistribution of potassium, calcium, magnesium and
manganese in the plant. Physiol Plant 31: 163-168
Graham RD (1975) Male sterility in wheat plants deficient in copper. Nature (London)
254:514--515
Graham RD (1976) Anomalous water relations in copper-deficient wheat plants. Aust
J Plant Physiol 3: 229-236
Grasmanis VO, Edwards GE (1974) Promotion of flower initiation,in apple trees by
short exposure to the ammonium ion. Aust J Plant Physiol1 : 99-105
Hallock DL, Garren KH (1968) Pod breakdown, yield and grade of Virginia-type peanuts
as affected by Ca, Mg and K sulfates. Agron J 60:253-257
Hambidge KM, Wallravens PA (1976) Zinc deficiency in infants and preadolescent chil-
dren. In: Prasad AA, Overleas D (eds) Trace elements in human health and disease,
vol!. Zinc and copper. Academic Press, London New York
Hara T, Sonoda Y (1979) Comparison of the toxicity of heavy metals to cabbage growth.
Plant Soil 51: 127-133
1.1 General Introduction to the Mineral Nutrition of Plants 53

Hawker JS, Marschner H, Downton WJS (1974) Effects of sodium and potassium on
starch synthesis in leaves. Aust J Plant Physiol1 :491-501
Heldt HW, Chon CH, Maronde D, Herold A, Stankovic ZS, Walker DA, Kraminer
A, Kirk MR, Heber U (1977) Role of orthophosphate and other factors in the regula-
tion of starch formation in leaves and isolated chloroplasts. Plant Physipl
59:1146-1155
Helms K, David DJ (1973) Far red and white light-promoted utilization of calcium
by seedlings of Phaseolus vulgaris L. Plant Physiol 51 :37-42
Herms U, Briimmer G (1979) EinfluB der Redoxbedingungen auf die Loslichkeit von
Schwermetallen in Boden und Sedimenten. Mitt Dtsch Bodenkdl Ges 29:533-544
Herridge DF, Pate JS (1977) Utilization of net photosynthesis for nitrogen fIxation and
protein production in an annual legume. Plant PhysioI60:759-764
Hewitt EJ, Smith TA (1975) Plant mineral nutrition. Engl Univ Press, London
Hildebrandt EA (1977) Das Vorkommen von Nitrosaminen in Nahrungsmitteln und
die Folgerungen flir die Pflanzenernahrung. Z Pflanzenernaehr Bodenkd 140:397-408
Hill J, Robson AD, Loneragan JF (1978) The effect of copper and nitrogen supply
on the retranslocation of copper in four cultivars of wheat. Austr J Agric Res
29: 925--939
Hill J, Robson AD, Loneragan JF (1979) The effects of Cu supply and shading on
Cu retranslocation from old wheat leaves. Ann Bot 43:449-457
Hinsley TD, Alexander DE, Ziegler EC, Barrett GL (1978) Zinc and cadmium accumula-
tion by com inbreds grown on sludge-amended soil, Agron J 70:425--428
Hocking PJ, Pate JS (1977) Mobilization of minerals to developing seeds of legumes.
Ann Bot 41: 1259-1278
Hocking PJ, Pate JS (1978) Accumulation and distribution of mineral elements in "annual
lupins Lupinus albus and Lupinus angustifolius L. Aust J Agric Res 29: 267-280
Hocking PJ, Pate JS, Wee GC, McComb AJ (1977) Manganese nutrition of Lupinus
ssp. especially in relation to developing seeds. Ann Bot 41: 677-688
Hocking PJ, Pate JS, Atkins CA, Sharkey PJ (1978) Diurnal patterns of transport and
accumulation of minerals in fruiting plants of Lupinus angustifolius L. Ann Bot
42: 1277-1290
Holl W, Hampp R (1975) Lead and plants. Residue Rev 54:79-111
Horst WJ, Marschner H (1978a) Effect of silicon on manganese tolerance of bean plants
(Phaseolus vulgaris L.). Plant Soil 50:287-303
Horst WJ, Marschner H (1978b) Effect of excessive manganese supply on uptake and
translocation of calcium in bean plants (Phaseolus vulgaris L.). Z Pflanzenphysiol
87:137-148
Howeler RH (1974) Iron-induced oranging disease of rice in relation to physical and
chemical changes in a flooded oxisol. Soil. Sci Soc Am Proc 37: 898-903
Hsiao TC (1976) Stomatal ion transport. In: Liittge U, Pitman MG (eds) Transport
in plants II, Part B. Encyclopedia of plant physiology new ser, vol II. Springer,
Berlin Heidelberg New York
Hsu FC (1979) Abscisic acid accumulation in developing seeds of Phaseolus vulgaris
L. Plant Physiol 63: 552-556 .
Huffman EWD, Allaway WH (1973) Growth of plants in solution culture containing
low levels of chromium. Plant Physiol 52: 72-75
Humble GD, Raschke K (1971) Stomatal opening quantitatively related to potassium
transport. Evidence from electron probe analysis. Plant Physiol 48: 447-453
Hutchings VM (1978) Sucrose and proton cotransport in Ricinus cotyledons. II. H+
efflux and associated K + uptake. Planta 138:237-241
Hutton JT, Norrish K (1974) Silicon content of wheat husks in relation to water trans-
pired. Aust J Agric Res 25:203-212
Isermann I (1978) EinfluB von Chelatoren auf die Calcium-Verlagerung im SproB h6herer
Pflanzen. Z Pflanzenernaehr Bodenkd 141 :285--298
Itai C, Meidner H (1978) Functional epidermal cells are necessary for abscisic acid effects
on guard cells. J Exp Bot 29:765--770
54 H. MARSCHNER:

Jager de A (1979) Localized stimulation of root growth and phosphate uptake in Zea
mays L. resulting from restricted phosphate supply. In: Harley JL, Scott Russell
R (eds) The soil-root interface. Academic Press, London New York
Janistyn B (1973) Indol-3-Essigsaure induzierte Calciumionenabgabe bei Maiskoleoptilzy-
lindem. Z Naturforsch 28c:777-778
Jennings DH (1976) The effects of sodium chloride on higher plants. BioI Rev 51 :453--
486
John MK, Laerhoven van CJ (1976) Differential effects of cadmium on lettuce varieties.
Environ Pollut 10: 163--173
Johnson CM, Stout PR, Broyer TC, Carlton AB (1957) Comparative chlorine require-
ments of different plants species. Plant Soil 8: 337-353
Jones LHP, Handreck KA (1965) Studies of silica in the oat plant. III. Uptake of silica
from soils by the plant. Plant Soil 23: 79-96
Jung J, Isermann K, Henges G (1979) EinfluB von cadmiumhaltigen Diingerphosphaten
auf die Cadmiumameicherung von Kulturboden und Nutzpflanzen. Landwirtsch
Forsch 32:262-274
Kelley PM, Izawa S (1978) The role of chloride ion in photo system-II. I. Effects of
chloride ion on photosystem-II electron transport and on hydroxylamine inhibition.
Biochim Biophys Acta 502: 198-210
Kick H, Braun B (1977) Wirkung von chromhaltigen Gerbereischliimmen aufWachstum
und Chromaufnahme bei verschiedenen Nutzpflanzen. Landw Forsch 30:160-173
Kirkby EA, Mengel K (1967) Ionic balance in different tissues of the tomato plant
in relation to nitrate, urea or ammonium nutrition. Plant PhysioI42:6-14
Kirkham MB (1978) Water relations of cadmium-treated plants. J Environ Qual
7:334-336
Kloke A, Riebartsch K, Leh HO (1966) Verumeinigungen von Kulturpflanzen mit Blei
aus Kraftfahrzeugabgasen. Landwirtsch Forsch Sonderheft 20: 119-123
Knauer N, Simon C (1968) Uber den EinfluB der Stickstoffdiingung auf den Ertrag
sowie Nitrat-Mineralstoff- und Oxalsauregehalt von Spinat. Z Acker Pflanzenbau
128:197-220
Knight AH, Crooke WM (1973) Cation exchange capacity and chemical composition
of the floral parts of Antirrhinum and Lilium. Ann Bot 37:155-157
Koontz HV, Foote RE (1966) Transpiration and calcium deposition by unifoliate leaves
of Phaseolus vulgaris differing in maturity. Physiol Plant 14:313--321
Krause W (1958) Andere Bodenspezialisten. In: Michael G (ed) Handbuch der Pflanzen-
physiologie, vol IV Springer, Berlin Gottingen Heidelberg
Krauss A (1971) EinfluB der Emahrung des Salates mit Massennahrstoffen auf den
Befall mit Botrytis cinera Pers. Z Pflanzenemaehr Bodenkd 128: 12-23
Krauss A (1978) Tuberization and abscisic acid content in Solanum tuberosum as affected
by nitrogen nutrition. Potato Res 21: 183--193
Krauss A, Marschner H (1971) EinfluB der Stickstoffemahrung der Kartoffeln aufInduk-
tion und Wachstumsrate der Knollen. Z Pflanzenemaehr Bodenkd 128:153--168
Krauss A, Marschner H (1973) Langstreckentransport von Calcium in Stolonen von
Kartoffelpflanzen. Z Pflanzenemaehr Bodenkd 136: 229-240
Krauss A, Marschner H (1974) EinfluB der TagJNacht-Periodik auf Knollengewicht und
Ca-Verlagerung in Stolonen von Kartoffelpflanzen. Z Pflanzenemaehr Bodenkd
137:116-123
Krauss A, Marschner H (1975) EinfluB des Calciumangebots auf Wachstumsrate und
Calciumgehalt von Kartoffelknollen. Z Pflanzenemaehr Bodenkd 139': 317-326
Krauss A, Marschner H (1976) EinfluB von Stickstoffemahrung und Wuchsstoffapplika-
tion auf die Knolleninduktion bei Kartoffelpflanzen. Z Pflanzenemaehr Bodenkd
139:143--155
Krug H, Wiebe H-J, Jungk A (1972) Calciummangel an Blumenkohl unter konstanten
Klimabedingungen. Z Pflanzenemaehr Bodenkd 133: 213--226
Lagerwerff JV, Milberg RP (1978) Sign of charge of species of Cu, Cd and Zn extracted
from sewage sludge and effect of plants. Plant and Soil 49: 117-125
Lagerwerff N, Specht AW (1970) Contamination of roadside soil and vegetation with
cadmium, nickel, lead and zinc. Environ Sci TechnoI4:583--586
1.1 General Introduction to the Mineral Nutrition of Plants 55

Lagerwerff JV, Biersdorf GT, Milberg RP, Brower DL (1977) Effects of incubation
and liming on yield and heavy metal uptake by rye from sewage-sludged soil. J
Environ Qua16:427-431
Lamoreaux RJ, Chaney WC (1977) Growth and water movement in silver maple seedlings
affected by cadmium. J Environ Qual 6: 201-205
Laske P (1979) Gehalt giirtnerischer Boden und Erden an loslichem Vanadium. Kali-
Briefe 14:747-752
Liiuchli A (1967) Nachweis von Calcium- und Strontium-Ablagerungen im Fruchtstiel
von Pisum sativum mit der Rontgen-Mikrosonde. Planta 73:221-227
Liiuchli A, Pfliiger R (1978) Potassium transport through plant cell membranes and
metabolic role of potassium in plants. In: Proc 11 th Congr Int Potash Inst, Bern
Lee J, Reeves RD, Brooks RR, Jafre T (1978) The relation between nickel and citric
acid in some nickel-accumulating plants. Phytochemistry 17: 1033-1035
Lenz F (1970) EinfluB der Friichte auf das Wachstum, den Wasserverbrauch und die
Niihrstoffaufnahme von Auberginen. Gartenbauwissenschaft 35: 281-292
Lenz F, Biinemann G (1969) EinfluB von Erniihrung, Fruchtbehang und Ausliiufern
auf die Trockenstubstanzbildung und die Wasser- und Niihrstoffaufnahme bei Erd-
beeren. Erwerbsobstbau 11: 185-188
Lenz F, Doring HW (1975) Fruit effects on growth and water consumption in citrus.
Gartenbauwissenschaft 40: 257-260
Lenz F, Williams CN (1973) Effect of fruit removal on net assimilation and gaseous
diffusive resistance of soybean leaves. Angew Bot 47: 57-63
Liegel W (1970) Calciumoxalat-Abscheidung in Fruchtstielen einiger Apfelvaritiiten.
Angew Bot 44: 223-232
Lingle JC, Lorenz OA (1969) Potassium nutrition of tomatoes. J Am Soc Hortic Sci
94:679-683
Linser H, Raafat A, Zeid FA (1974) Reinprotein und Chlorophyll bei Daucus carota
im Verlauf der Vegetationsperiode des ersten Jahres unter dem EinfluB von Wachs-
tumsregulatoren (CCC, GA, BA). Z Pflanzenernaehr Bodenkd 137: 36-48
Loneragan JF (1978) Anomalies in the relationship of nutrient concentrations to plant
yield. 8th Int Colloq Plant Anal Fert Probl Auckland, NZ
Loneragan JF, Snowball K (1969) Calcium requirements of plants. Aust J Agric Res
20:465-478
Loneragan JF, Snowball K, Simmons WJ (1968) Response of plants to calcium concentra-
tion in solution culture. Aust J Agric Res 19:845-857
Loneragan JF, Snowball K, Robson AD (1976) Remobilization of nutrients and its
significance in plant nutrition. In: Transport and transfer processes in plants, Chap
39. Academic Press, London New York
Lorenz H, Kossen MTh, Kiiferstein FK (1978) Blei-, Cadmium- und Quecksilbergehalte
in Speisepilzen. Bundesgesundheitsblatt 21 :202-204
Lott JNA, Buttrose MS (1978) Globoids in protein bodies of legume seed cotyledons.
Aust J Plant Physiol 5: 98-111
Lougheed EC, Murr DP, Miller SR (1979) Effects of calcium and daminozide on ethylene
production and softening of apple fruits. Experimenta 35: 43-45
Liidders P, Biinemann G (1970) Die Wirkung des Zeitpunktes von Harnstoffspritzungen
auf Apfelbiiume. Z Pflanzenernaehr Bodenkd 125:144-155
Liidders P, Biinemann G, Ortmann U (1976) Die Wirkungjahreszeitlich unterschiedlicher
Kaliumverfiigbarkeit auf Apfelbiiume. IX. EinfluB auf den Mineralstoffgehalt aus-
dauernder Baumteile. Gartenbauwissenschaft 41 :260--269 '
Lund ZF, Doss BD, Lowry FE (1975) Dairy cattle manure - Its effect on yield and
quality of coastal bermuda grass. J Environ Qual 4: 358-362
Marschner H (1971) Why can sodium replace potassium in plants? Proc 8th Colloq
Int Potash Inst Bern
Marschner H, Ossenberg-Neuhaus H (1977) Wirkung von 2,3,5-Trijodbenzoesaure
(TIBA) auf den Calciumtransport und die Kationenaustauschkapazitiit in Sonnenblu-
men. Z Pflanzenphysiol 85: 29-44 .
Marschner H, Possingham JV (1975) Effect of K + and Na + on growth of leaf discs
of sugar beet and spinach. Z Pflanzenphysiol 85: 6-16
56 H. MARSCHNER:

Marschner H, Richter Ch (1974) Calciumtransport in Wurzeln von Mais- und Bohnen-

keimpflanzen. Plant Soil 40: 193-210
Marschner H, R6mheld V, Azarabadi S (1978) Iron stress response of efficient and
inefficient plant species. In: Proc 8th Colloq Plant Anal Fert Probl. Auckland, NZ
Martin G (1965) Necessite du chlore dans nutrition de Spirodela polyrhiza cultive en
conditions heterotrophes. CR Acad Sci 260:5928-5930
Martin MH, Coughtrey PJ (1978) Preliminary observations on the level of cadmium
in a contaminated environment. Chemosphere 3: 155-160
Martin P (1971) Wanderwege des Stickstoffs in Buschbohnenpflanzen bei Aufwartstrans-
port nach der Aufnahme durch die Wurzeln. Z PflanzenphysioI64:206-222
Mathys W (1975) Enzymes of heavy metal resistant and non-resistant populations of
Silene cucubalus and their interaction with some heavy metals in vitro and in vivo.
Physiol Plant 33:161-165
Mattoo AK, Liebermann M (1977) Localization of the ethylene-synthesizing system in
apple tissue. Plant Physiol 60: 794-799
Meisch HK, Schmitt JA, Reinle W (1977) Schwermetalle in h6heren Pilzen. Cadmium,
Zink und Kupfer. Z Naturforsch 32c:172-181
Menary RC, Staden van J (1976) Effect of phosphorus nutrition and cytokinins on
flowering in tlhe tomato, Lycopersicon esculentum L. Aust J Plant Physiol 3: 201-205
Mengel K (1979) Ernahrung und Stoffwechsel der Pflanzen, 5th edn. Fischer, Jena
Mengel K, Haeder H-E (1977) Effect of potassium supply on the rate of phloem sap
exudation and the composition of phloem sap of Ricinus communis. Plant Physiol
59:282-284
Mengel K, Kirby EA (1978) Principles of plant nutrition. lnt Potash lnst, Bern
Mengel K, Grimme H, Nemeth K (1969) Potentielle und effektive Verfiigbarkeit von
Pflanzennahrstoffen im Boden. Landwirtsch Forsch 23:(16. Sonderheft) 79-91
Mengel K, Haghparast MR, Koch K (1974) The effect of potassium on the fixation
of molecular nitrogen by root nodules of Vicia faba. Plant Physiol 54: 535-538
Michael G, Beringer H (1980) The role of hormones in yield formation. In: Proc 15th
lnt Potash Colloq, Wageningen '
Michael G, Koushiahi-Tork K, Wilberg E (1969a) EinfluB unterschiedlicher Luftfeuchtig-
keit auf Chlorophyll- und EiweiBabbau in Blattern von Tabakpflanzen. Flora
A 160:186-195
Michael G, Mounla MAKh, Goldbach H (1974) Nitrogen fertilizer application, hormone
activity and crop production. Plant Anal Fert Probl Proc 7th lnt Colloq, Hannover
Michael G, Schultz R, Wagner H (1969b) Modellversuch zur Bedeutung der Granne
fur die Versorgung des Gerstenkornes mit Cytokininen. Flora A 160:306-316
Micheal B, Zink F, Lantzsch HJ (1980) Effect of phosphate application on phytin-P
and other phosphate fractions in developing wheat grains. Z Pflanzenernaehr Bodenkd
369-376
Millikan CR, Hanger BC (1969) Movement of a foliar-applied 45Ca in Brussels sprouts.
Austr J BioI Sci 22: 545-558
Mitchell GA, Bingham ET, Page AL (1978) Yield and metal composition of lettuce
and wheat grown on soils amended with sewage sludge enriched with cadmium, cop-
per, nickel and zinc. J Environ Qual 7: 165-171
Mix GP, Marschner H (1976a) Ca1ciumgehalte in Fruchten von Paprika, Bohne, Quitte
, und Hagebutte im Verlauf des Fruchtwachstums. Z Pflanzenernaehr Bodenkd
139: 537- 549
Mix GP, Marschner H (1976b) EinfluB exogener und endogener Faktonin auf den Cal-
ciumgehalt von Paprika- und Bohnenfruchten. Z Pflanzenernaehr Bodenkd
139: 551-563
Mix GP, Marschner H (1976c) Calcium-Umlagerung in Bohnenfruchten wahrend des
Samenwachstums. Z Pflanzenphysiol 80: 354-366
Miyake Y, Takahashi E (1978) Silicon deficiency of tomato plants. Soil Sci Plant Nutr
24:175-189
Mohamed AH, Gnanam A (1979) A possible mechanism of ammonium ion regulation
of photosynthetic carbon flow in higher plants. Plant Physiol 64: 263-268
Monselise SP, Varga A, Knegt E, Bruinsma J (1978) Course of the zeatin content in
1.1 General Introduction to the Mineral Nutrition of Plants 57

tomato fruits and seeds developing on intact or partially defoliated plants. Z Pflanzen-
physioI90:451-460
Mostafa MAE, Ulrich A (1976) Absorption, distribution and form of Ca in relation
to Ca deficiency (tip bum) of sugar beet. Crop Sci 16:27-30
Mothes K (1932) Erniihrung, Struktur und Transpiration. Ein Beitrag zur kausalen Ana-
lyse der Xeromorphosen. BioI Zentralbl 52: 193-223
Naumann WD (1971) Calciumverteilung in Apfeln und Entwicklung von Stippigkeit
unter dem EinfluB von Wachstumsregulatoren. Gartenbauwissenschaft 36:63-68
Nitsch JP (1950) Growth and morphogenesis of strawberries as related to auxin. Annu
J Bot 37:211-215
Ocker H -D (1977/78) Geh!;llte und Verteilung von Schwermetallen in Getreide und Getrei-
deprodukten. Ber Landwirtsch 55: 796-808
Oertli JJ, Richardson WF (1970) The mechanism of boron immobility in plants. Physiol
Plant 23:108-116
Oldenkamp L, Smilde KW (1966) Copper deficiency in douglas fir (Pseudotsuga menziesii
Mirb. Franco). Plant Soil 25: 150-152
Overleas D (1973) Phytates. In: Toxicants occurring naturally in foods. Comm Food
Protec Nutr Board, Natl Res Counc. Natl Acad Sci, Washington
Page AL, Bingham FT (1973) Cadmium residues in the environment. Residue Rev
48:1-44
Palzkill DA, Tibbitts TW, Williams PH (1976) Enhancement of calcium transport to
inner leaves of cabbage for prevention of tipburn. I Am Soc Hortic Sci 101 :645-
648
Pate JS (1976) Exchange of solutes between phloem and xylem and circulation in the
whole plant. In: Zimmermann MH, Milburn JA (eds) Phloem transport. Encyclopedia
of plant physiology, new ser vol I. Springer, Berlin Heidelberg New York
Pate JS, Hocking PJ (1978) Phloem and xylem transport in the supply of minerals to
a developing legume (Lupinus albus L.) fruit. Ann Bot 42:911-921
Pate JS, Sharkey PJ, Lewis OAM (1976a) Phloem bleeding from legume fruits - a tech-
nique for study of fruit nutrition. Planta 120: 229-243
Pate JS, Sharkey PJ, Lewis OAM (1974 b) Xylem to phloem transfer of solutes in fruiting
shoots oflegumes, studies by a phloem-bleeding technique. Planta 122: 11-26
Pate JS, Kuo J, Hocking PJ (1978) Functioning of conducting elements of phloem and
xylem in the stalk of the developing fruit of Lupinus albus L. Aust J Plant Physiol
5:321-326
Peoples TR, Koch DW (1979) Role of potassium in carbon dioxide assimilation in Medi-
cago sativa L. Plant Physiol 63: 878-881
Perrenoud S (1977) Potassium and plant-health. Potash Inst Bern
Petit CM, Ringoet A, Myttenaere C (1978) Stimulation of cadmium uptake in relation
to the cadmium content of plants. Plant Physiol62: 554-557
Pettersson A (1976) Heavy metal ion uptake by plants from solutions with metal ion,
plant species and growth period variations. Plant Soil 45:445-459
Pettersson A (1977) Differences in cadmium uptake between plant species and cultivars.
Swed J Agric Res 7:21-24
Pfluger R, Wiedemann R (1977) Der EinfluB monovalenter Kationen auf die Nitratreduk-
tion von Spinacia oleracea L. Z Pflanzenphysiol 85: 125-133
Pietz RI, Vetter RJ, Masarik D, McFee WW (1978) Zinc and cadmium contents of
agricultural soils and com in Northwestern Indiana. J Environ Qual 7:381-38.)
Poovaiah BW (1979) Role of calcium in ripening and senescence. Commun Soil Sci
Plant Anal 10:83-88
Poovaiah BW, Leopold AC (1973 a) Deferral of leaf senescence with calcium. Plant Phy-
siol 52: 236-239
Poovaiah BW, Leopold AC (1973 b) Inhibition of abscission by calcium. Plant Physiol
51:848-851
Poovaiah BW, Leopold AC (1976) Effect of inorganic solutes on the bindings of auxin.
Plant Physiol 58: 783-785
Priestley CA (1976) Some effects of ringing and deblossoming branches of young apple
trees in leaf composition. J Exp Bot 27: 1325-1332
58 H. MARSCHNER:

Quinlan JD, Preston AP (1973) Effect of shoot tipping on fruiting and apple tree growth.
East Malling Res Stn Rep
Rahimi A, Bussler W (1973a) Die Diagnose des Kupfermangels mittels sichtbarer Sym-
ptome an h6heren Pflanzen. Z Pflanzenernaehr Bodenkd 135: 267-283
Rahimi A, Bussler W (1973b) Physiologische Voraussetzungen fiir die Bildung der Kup-
fermangelsymptome. Z Pflanzenernaehr Bodenkd 136: 25-32
Rahimi A, Bussler W (1973c) Die Wirkung von Kupfermangel auf die Gewebestruktur
hOherer Pflanzen. Z Pflanzenernaehr Bodenkd 135:183-195
Rahman MS, Wilson JH (1977) Effect of phosphorus applied as superphosphate on
rate of deVelopment and spikelet number. Austr J Agric Res 28: 183-186
Rajagopal V, Rao 1M (1974) Changes in the endogenous level of auxins and gibberellin-
like substances in the shoot apices of nitrogen-deficient tomato plants (Lycopersicon
esculentum Mill). Aust J Bot 22:429--435
Ramon Garcia L, Hanway JJ (1976) Foliar fertilization of soybean during the seed-filling
period. Agron J 68: 653-657
Rapper CD, Patterson DT, Parsons LR, Kramer PJ (1977) Relative growth and nutrient
accumulation rates for tobacco. Plant Soil 46:473-486
Raschke K (1979) Movements of stomata. In: Haupt W, Feinleib ME (eds) Physiology
of movements. Encyclopedia of plant physiology new ser vol VII. Springer, Berlin
Heidelberg New York
Raven JA (1977) H + and Ca 2 + in phloem and symplast. Relation of relative immobility
of the ions to the cytoplasmic nature of th~ transport path. New Phytol 79: 465-480
Reinhold JG, Nasr K, Lahimgarzadeh A, Hedayati H (1973) Effect of purified phytate
and phytate-rich bread upon metabolism of zinc, calcium, phosphorus and nitrogen
in man. Lancet 1 : 283-291
Reuss JO, Dooley HL, Grittis W (1978) Uptake of cadmium from phosphate fertilizers
by pear, radishes and lettuce. J Environ Qual 7:12&-133
Riley D, Barber SA (1971) Effect of ammonium and nitrate fertilization on phosphorus
uptake as related to root-induced pH changes at the root-soil interface. Soil Sci
Soc Am Proc 35: 301-306
Ringoet A, Sauer G, Gielink AJ (1968) Phloem transport of calcium in oat leaves. Planta
80: 15-20
R6mheld V, Marschner H (1979) Fine regulation of iron uptake by the Fe-efficient
plant species Helianthus annuus. In: Harley JL, Scott Russell R (eds) The soil-root
interface. Academic Press, London New York
Sahrawat KL, Mukarjee SK, Gulat KC (1977) Nitrification inhibitors. II. Studies with
furano compounds. Plant Soil 47: 687-691
Salama EI-D AMS, Wareing PF (1979) Effects of mineral nutrition on endogenous cytoki-
nins in plants of sunflowers (Helianthus annuus L.). J Exp Bot 30: 971-981
Sattelmacher B, Marschner H (1978) Nitrogen nutrition and cytokinin activity in Solanum
tuberosum. Physiol Plant 42: 185-198
Sattelmacher B, Marschner H (1979) Tuberization in potato plants as affected by applica-
tion of nitrogen to the roots and leaves. Potato Res 22: 39--47
Satter RL, Galston AW (1971) Phytochrome-controlled nyctinasty in Albizzia librium.
III. Interactions between endogenous rhythm and phytochrome in control of potas-
sium flux and leaflet movement. Plant Physiol 48: 740-746
Scaife MA, Clarkson DT (1978) Calcium-related disorders in plants - a possible explana-
tion for the effect of weather. Plant Soil 50: 723-725
Schilling G, Schalldach I (1966) Untersuchungen iiber Transport, Einbau und Verwertung
spat gediingten Mineralstickstoffs bei Pisum sativum L. Thaer Arch 10:895-907
Schuphan W (1976) Mensch und Nahrungspflanze. Jung, Den Haag
Seiler-Kelbitsch H, Michael G, Wilberg E (1974) Beziehungen zwischen Korngewicht
und Cytokinin-Aktivitat bei Sommergerste, untersucht am Beispiel des Abschneidens
von Bestockungstrieben. Angew Bot 48: 299--307
Seth AK, Wareing PF (1967) Hormone-directed transport of metabolites and its possible
role in plant senescence. J Exp Bot 18: 65-77
1.1 General Introduction to the Mineral Nutrition of Plants 59

Sharples RO, Johnson DS (1977) The influence of calcium on senescence changes in

apple. Ann Appl BioI 85:45Q-453
Shay FJ, Hale MG (1973) Effect of low levels of calcium on exudation of sugar and
sugar derivates from intact peanut roots under axenic conditions. Plant Physiol
51:1061-1063
Shear CB (1975) Calcium-related disorders of fruits and vegetables. Hort Sci 10:361-
365
Shear CB (1979) International symposium on calcium nutrition of economic crops. Com-
mun Soil Sci Plant Anal 10:1-501
Shear CB, Faust M (1970) Calcium transport in apple trees. Plant PhysioI45:67Q-674
Simojoki P (1972) Resultsof boron fertilizer experiments in barley. Ann Agric Fenniae
11 :333-341
Sinclair TR, Dewit CT (1976) Analysis of the carbon and nitrogen limitations to soybean
yield. Agron J 68:319-324
Sinha BK, Shingh NT (1976) Chloride accumulation near corn roots under different
transpiration, soil moisture and soil salinity regimes. Agron J 88: 346-348
Staden van J, Davey JE (1979) The synthesis, transport and metabolism of endogenous
cytokinins. Plant Cell Environ 2: 93-106
Strickland RC, Chaney WR, Lamoreaux RJ (1979) Organic matter influences phytotoxi-
city of cadmium to soybeans. Plant Soil 52: 393-402
Symeonides C, McRae SG (1977) The assessment of plant available cadmium in soils.
J Environ Qual 6: 12Q-123
Terry N (1977) Photosynthesis, growth and the role of chloride. Plant PhysioI60:69-75
Trobisch S, Schilling G (1969) Untersuchungen iiber Zusammenhiinge zwischen Massen-
entwicklung und N-Umsatz wiihrend der generativen Phase bei Sinapis alba L. Thaer
Arch 13: 867-878
Trobisch S, Schilling G (1970) Beitrag zur Kliirung der physiologischen Grundlage der
Samenbildung bei einjiihrigen Pflanzen und zur Wirkung spiiter zusiitzlicher N-Gaben
auf diesen ProzeB am Beispiel von Sinapis alba L. Thaer Arch 14:253-265
Tromp J (1979) The intake curve for calcium into apple fruits under various environmen-
tal conditions. Commun Soil Sci Plant AnaI10:32~335
Tromp J, Oele J (1972) Shoot growth and mineral composition of leaves and fruits
of apple as affected by relative air humidity. Physiol Plant 27:253-258
Varga A, Bruinsma J (1974) The growth and ripening of tomato fruits at different levels
of endogenous cytokinins. J Hort Sci 49: 13~ 142
Vaughan AKF (1977) The relation between the concentration of boron in the reproductive
and vegetative organs of maize plants and their development. Rhod J Agric Res
15: 163-170
Venkat-Raju K, Marschner H (1972) Regulation of iron uptake from relatively insoluble
iron compounds by sunflower plants. Z Pflanzenernaehr Bodenkd 133:227-241
Vetter H, Teichmann W (1968) Feldversuche mit gestaffelten Kupfer- und Stickstoff-
Diingergaben in Weser-Ems. Z Pflanzenernaehr Bodenkd 121 :97-111
Vetter H, Miihlop R, Fiirchtenicht K (1974) Immissionsstoffbelastung in der Nach-
barschaft einer Blei- und Zinkhiitte. Ber Landwirtsch 52: 327-350
Visser J, Locher JTh, Brouwer R (1971) Effects of aeration and mineral supply on
growth and mineral content of shoots and roots of apple trees (var. Golden Delicious
on MIX). Neth J Agric Sci 19:12~137
Vlamis J, Williams D (1967) Manganese and silicon interaction in the Gramineae., Plant
Soil 27:131-140
Volk RJ (1958) Silicon content of the rice plant as a factor influencing its resistance
to infections by the blast fungus, Piricularia oryzae. Phytopathology 48: 179-184
Vorm van der PDJ, Diest van A (1979) Aspects of the Fe and Mn nutrition of rice
plants. II. Iron and manganese uptake by rice plants, grown on aerobic water cultures.
Plant Soil 52: 19-29
Wagner HC (1974) Wuchsstoffgesteuerte Assimilateverlagerung bei Gerste. Angew Bot
48:331-338
60 H. MARSCHNER: 1.1 General Introduction to the Mineral Nutrition of Plants

Wagner H, Michael G (1971) Der EinfluB unterschiedlicher Stickstoffversorgung auf

die Cytokininbildung in Wurzeln von Sonnenblumenpflanzen. Biochem Physiol Pflan-
zen 162: 147-158
Wainwright SI, Woolhouse HW (1977) Some physiological aspects of copper and zinc
tolerance in Agrostis tenuis Sibth: Cell elongation and membrane damage. 1 Exp
Bot 28: 1029-1036
Wakita H, Schmitt RA (1970) Cadmium. In: Wedepohl KH (ed) Handbook of geochemi-
stry Vol 11-4/48. Springer, Berlin Heidelberg New York
Wehrmann 1, Scharpf HC (1979) Der Mineralstickstoffgehalt des Bodens als MaBstab
fiir den Stickstoffdiingerbedarf (Nmin-Methode). Plant Soil 52: 109-126
Wetzel R, Menke KH, Oelschlager W (1979) Gehalte an Spuren- und Mengenelementen
in verschiedenen Mais- und Hafersorten sowie in einigen Heuherkiinften. Landwirtsch
Forsch 32:101-109
Wiebe HI, Schatzler HP, Kiihn W (1977) On the movement and distribution of calcium
in white cabbage in dependence on the water status. Plant Soil 48 :409-416
Wieneke 1 (1974) Untersuchungen zur Translokation von 45Ca im Apfelbaum. Garten-
bauwissenschaft 39: 57-67
Wieneke 1 (1979) Calcium transport and its microautoradiographic localization in the
tissue. Commun Soil Sci Plant Anal 10:237-250
Wieneke 1, Fiihr F (1973 a) Untersuchungen zur Translokation von 45Ca im Apfelbaum.
I. Transport und Verteilung in Abhangigkeit vom Aufnahmezeitpunkt. Gartenbauwis-
senschaft 38:91-108
Wieneke 1, Fiihr F (1973 b) Mikroautoradiographischer Nachweis von Ca45 Kristallabla-
gerungen im Apfelstiel- und Fruchtgewebe. Angew Bot 47: 107-112
Wieneke 1, Biddulph 0, Woodbridge CG (1971) Influence of growth-regulating substan-
ces on absorption and translocation of calcium in pea and bean. 1 Am Soc Hortic
Sci 96: 721-724
Wiersum LK (1966) Calcium content of fruits and storage tissue in relation to the mode
of water supply. Acta Bot Neerl15:406-418
Wilkinson BG (1968) Mineral composition of apples. IX. Uptake of calcium by the
fruit. 1 Sci Food Agric 19:646-647
Willenbrink 1, Schuster WB (1978) Localized inhibition of translocation of 14C-assimila-
tes in the phloem by valinomycin and other metabolic inhibitors. Planta 319:261-265
Williams DE, Vlamis 1 (1957) The effect of silicon on yield and manganese-54 uptake
and distribution in the leaves of barley plants grown in culture solutions. Plant Physiol
32:404-409
Wu L, Thurman DA, Bradshaw AD (1975) The uptake of copper and its effect upon
respiratory processes of roots of copper-tolerant and non-tolerant clones of Agrostis
stolonifera. New Phytol 75:225-229
Yamagata N, Shigematsu 1 (1970) Cadmium pollution in perspective. Bull Inst Publ
Health 19:1-27
Yoshida S, Nasavero SA, Ramirez EA (1969) Effect of silica and nitrogen supply on
some leaf characters of the rice plant. Plant Soil 31 : 48- 56
Yuan HF, Chang YS (1978) Effect of available silicon in paddy soil on the growth
of rice plants. Bot Bull Acad Sinica 19: 125-138
Zimdahl RL, Foster 1M (1976) The influence of applied phosphorus, manure or lime
on uptake of lead from soil. 1 Environ Qual 5:31-34
1.2 The Significance of Rhizosphere Microflora
and Mycorrhizas in Plant Nutrition
A.D. ROVIRA, G.D. BOWEN and R.C. FOSTER

1 Introduction

Plant roots are always associated with micro-organisms, ranging from organisms
external to the root to root-infecting micro-organisms. The nature and signifi-
cance of these associations have interested soil microbiologists and, to a much
lesser extent, plant physiologists, for decades. The term "rhizosphere" was pro-
posed by HILTNER (1904) to define that soil influenced by living roots. The
rhizosphere is variable both in extent and composition, and from the root surface
to the bulk soil there is a gradient of many chemical, physical and biological
properties. To accommodate this gradient, microbiologists have proposed terms
such as "outer rhizosphere", "inner rhizosphere" and "rhizoplane" (the root
surface itself). For this paper we treat the rhizosphere as that zone of soil
extending from the root-soil interface to the point in the soil where the micro-
flora is unaffected by the root. The non-infective rhizosphere micro-organisms
can have a large effect on plant nutrition, as do the infective associations such
as mycorrhizal fungi (beneficial) and root diseases (detrimental).

2 Energy Supplies in the Rhizosphere

The populations of micro-organisms which build up in and around roots arise

because of the greater supply of energy-rich nutrients in the rhizosphere coming
from the roots than in the bulk soil; see HALE et al. (1979) for the most recent
review on root exudates.
ROVIRA et al. (1979) defined the origin and nature of these organic materials
derived from the root as in Fig. 1.

2.1 Exudates

Compounds of low molecular weight which leak from all cells into either the
intercellular spaces and then to the soil via the cell junctions or directly through
the epidermal cell walls into the soil. The release ,of these compounds is not
metabolically mediated.
62 A.D. ROVIRA et al.:

_ ROOT
ClASS OF Fig. 1. Diagram of a root
MATERIAL
showing the origins of
organic materials in the
rhizosphere (see text). (By
EPIDERMAL A D
courtesy of Academic Press)
COfITICAl CELLS
LYSED ANO INVADED
5
BY BACTER1A

~+------------1--3d

20mm

MICRO ORGA ISMS }

WITH MlCl106lAl AND 4
F'lANT MLlCILAGES
SPREADING INTO THE S04L

~ _ _ _ _ _ _ _ _ __ J __ l&2

~~----------------3b
TnI,, " - -_ SLOUGHED ROOT CAP CELLS

~~~--~~~~--------3a
ROOT CAP

2.2 Secretions

Compounds of low or high molecular weight which are released as a result

of metabolic processes.

2.3 Plant Mucilages

Four sources of plant mucilages which contribute to the organic materials in

the rhizosphere are:
a) Mucilage originating in the root cap and secreted by Golgi vesicles.
b) Hydrolysates of the polysaccharide of the primary cell wall between the
epidermal cells and sloughed root cap cells.
c) Mucilage secreted by the epidermal cells which still only have primary walls.
This includes mucilages secreted by root hairs. The development 'of secondary
walls in these cells further from the tip means that no further secretions
or mucilages are released, only exudates.
At this part of the root the primary plant mucilage is confined to the intercel-
lular grooves where bacteria are unable to enter.
d) Mucilage produced by bacterial degradation of the outer multilamellate pri-
mary cell walls of old, dead epidermal cells.
1.2 The Significance of Rhizosphere Microflora and Mycorrhizas 63

2.4 Mucigel

Following JENNY and GROSSENBACHER (1963) this term means the gelatinous
material at the surface of roots grown in normal non-sterile soils. It includes
natural and modified plant mucilages, bacterial cells and their metabolic prod-
ucts (such as capsules and slimes) as well as colloidal mineral and organic
matter from the soil. Whereas root mucilages entirely of plant origin and micro-
bially produced mucilages can only be studied and characterized in axenic cul-
tures, mucigel is a product of the entire root-soil-microbial complex with its
own distinctive morphological and biochemical properties. The mucigel could
be important in maintaining contact between the root and the soil.

2.5 Lysates

These are the compounds released from autolysis of older epidermal cells when
the plasmalemma fails. With further time the walls of these epidermal cells
are digested by micro-organisms and the cells become heavily colonized releasing
the products of microbial activity into the rhizosphere.
In studies on soluble exudates released by roots growing in nutrient solution,
ROVIRA (1969) estimated from 0.1 % to 0.5% of the carbon fixed by photosynthe-
sis was released into the solution. However, results from solution culture can
be misleading, and in simplified or "model" systems some attempt should be
made to simulate soil characteristics. Thus, BARBER and GUNN (1974) found
that the release of soluble carbon from roots growing in Ballotini (glass beads)
increased to 9% of root weight as the resistance to root penetration was in-
creased. In soil, the amounts can be higher still for BARBER and MARTIN (1976)
found that roots of wheat seedlings released 5% to 10% of the carbon assimi-
lated by the plant when grown without microorganisms and 12% to 18% in
the presence of a microflora over three weeks. These results are consistent with
those of MARTIN (1975, 1977), VANCURA (1964) - 7% to 10% of the biomass
under water stress; SHAMOOT et al. (1960) - 9% to 42% ofroot; BREISCH (1974)
- 10% to 20% of photosynthate; WAREMBOURG (1975), WAREMBOURG and
BILLES (1979) - 12% to 16% of photosynthate; BILLES and CORTEZ (1975)
- 4% to 8% of biomass; BALANDREAU and HAMAD-FARES (1975) - 10% to
20% of biomass. The release of such a large amount of carbon into the rhizo-
sphere will affect not only the microflora but also soil processes and the avail-
ability of nutrients.
Interactions between exudation and the physiology of the plant as affected
by environmental conditions and nutrition have been demonstrated. Temporary
wilting of plants increased the exudation of amino acids from roots into sand
or solution (KATZNELSON et al. 1954, 1955). Exudation of amino acids and
sugars increased with increasing light intensity and temperature (ROVIRA 1959).
The link between the exudation and composition of cell contents and cell perme-
ability is demonstrated in the results of BOWEN (1969) in which the greater
amounts of amino acids and amides were released from plants on low phosphate
64 A.D. ROVIRA et al.:

nutrition, this greater exudation being due to the higher content of free amino
acid and amide. RATNAYAKE et al. (1978) reported that more amino acids and
reducing sugars were released from roots grown under low P nutrition due
to a change in cell permeability related to changes in phospholipid content
resulting from the P deficiency.

3 Microbiology of the Rhizosphere

3.1 Populations of Micro-Organisms

Bacteria, fungi and actinomycetes are more abundant in the rhizosphere than
in soil away from roots - this information is based upon counts of colonies
of the organisms growing in nutrient agar from serial dilutions of soil and
roots (KATZNELSON et al. 1948, ROVIRA 1965a).
Estimates of populations in the rhizospheres of different plants range from
5x10 8 to 3x10 9 g-l soil (ROVIRA and DAVEY 1974). Such figures give the
impression that the root surface is covered by a mass of micro-organisms. How-
ever, direct microscope observations and counts of microbial populations of
roots reveal otherwise. BOWEN and THEODOROU (1973) showed usually less than
10% of the root surface of Pinus radiata up to 21 days old was occupied by
micro-organisms. In a later study with eight plant species from grassland ROVIRA
et al. (1974) found from 4% to 10% cover of roots by bacteria and a density
of fungal hyphae of 12 to 14 mm mm- 2 of root. SCHIPPERS and VAN VUURDE
(1978) used fluorochromes and fluorescence microscopy to measure the cover
of micro-organisms on wheat roots and found the cover to increase from 1 %
in that portion of root less than 4 days old and 2 % to 4% in the 4- to 8-day-old
segments.

3.2 Colonization of Roots by Micro-Organisms

The distribution of micro-organisms on the surfaces of roots is non-random;

NEWMAN and H.J. BOWEN (1974) showed that bacteria are aggregated in micro-
colonies - a result confirmed with subterranean clover by transmission electron
microscopy of ultra-thin sections of the root-soil interface (FOSTER and ROVIRA
1978). This non-random distribution occurs for two reasons, firstly, the non-
random distribution of micro-organisms in soil due to their association with
fragments of organic matter and, secondly the initial colonization of roots at
the junctions of epidermal cells due to these being better sites for microbial
growth (BOWEN and ROVIRA 1976). This is probably due to greater exudation
from the junctions of cells which act as "drainage outlets" for the materials
which accumulate in the cell-free spaces of the epidermal and cortical cells
of the root (see Fig. 1).
I.2 The Significance of Rhizosphere Microflora and Mycorrhizas 65

Studies based upon the isolation and characterization of rhizosphere and

soil bacteria have shown a selective stimulation of Gram-negative non-sporing
rods with characteristic nutritional requirements (LOCHHEAD 1940, LOCHHEAD
and CHASE 1943).
The dynamics of this selective stimulation of Gram-negative bacteria was
indicated by CHAN and KATZNELSON (1961) when they found that while Pseudo-
monas sp. and Arthrobacter sp. colonized roots and culture solution equally
well in pure culture, the Arthrobacter level was reduced to one hundredth of
this level in the presence of Pseudomonas. BOWEN and ROVIRA (1976) found
that Pseudomonas sp. and Bacillus sp. had generation times of 5.2 and 39 h
respectively on roots (77 and > 100 h respectively in soil). Such a difference
explains the reported dominance of the former in natural rhizospheres. Even
within different Pseudomonas species there were up to 500-fold differences in
populations on roots (BOWEN 1979).
The interactions between bacteria and mycorrhizal fungi on roots of Pinus
radiata were demonstrated by BOWEN and THEODOROU (1973); they found that
the length of root colonized by Rhizopogon luteolus was not affected by Bacillus
sp. but reduced to one third by Pseudomonas sp. Another Bacillus sp. stimulated
the growth of mycorrhizal fungi significantly.

4 Mathematical Modelling of the Rhizosphere

The experimental approach which yielded the above results contrasts with the
mathematical modelling of NEWMAN and WATSON (1977), who assumed that
the growth and distribution of micro-organisms around roots was controlled
by the concentration of soluble carbon compounds which diffused into soil
and depended on such factors as soil moisture. Using the computer model,
they predicted the effects of changes in soil moisture, initial microbial population
and root exudation upon the concentrations of bacteria at different distances
from roots. According to the model, a change in soil moisture from - 0.1
to -1.0 bar matric potential would lead to a 31-fold increase in the level of
bacteria at the surface of the root; at 0.3 mm and 1.8 mm from the root the
concentration of micro-organisms would fall by 75% and 50% respectively
at the lower moisture content. Newman and Watson predicted that a 10-fold
increase in carbon exuded would lead to a 120-fold increase in popUlation at
the root surface and a 3-fold increase in the population of bacteria in the
rhizosphere. Such a model could give a better identification offactors controlling
microbial growth at and near the root, and in the context of plant nutrition,
could indicate the extent of competition between roots and micro-organisms
for nutrients diffusing towards roots. A simulation computer model needs vali-
dation by experimentation and more hard data is needed to confirm the
Newman-Watson model.
66 A.D. ROVIRA et al.:

5 Microscopy of the Rhizosphere

5.1 Light Microscopy

Direct microscopy of the rhizosphere was used by STARKEY (1939) when he

examined glass slides which had been buried in soil across the paths of roots;
these studies revealed the increased population and the diversity of micro-organ-
isms associated with roots. Using the aniline blue stain of JONES and MOLLISON
(1948) for bacteria enabled ROVIRA (1956, 1965a) to observe that for roots
growing in nutrient solution and glass beads the initial colonization occurred
along junctions of epidermal cells followed by the development of micro-colo-
nies. These observations were quantified by MALAJCZUK and BOWEN (unpub-
lished) to show that during the early colonization of roots of Eucalyptus calo-
phylla there were 50 times more bacteria along cell junctions than between
the junctions (BOWEN and ROVIRA 1976).
A further tool in light microscopy is the use of fluorescence microscopy;
TROLLDENIER (1965) used fluorescence with acridine orange to distinguish living
from dead bacteria and revealed a heavy coating of bacteria on the roots and
root hairs of clover grown in sand or agar. VAN VUURDE and ELENBAAS (1978)
found a good contrast between the fluorescence of micro-organisms and roots
and soil with successive staining with coriphosphine, congo red and acridine
orange.
The use of fluorescent antisera to detect specific microorganisms on roots
was used by TRINICK (1970) to follow competition between strains of Rhizobium
and by SCHANK et al. (1979) to identify Azospirillum brasilense associated with
the roots of grass.
Although light microscopy of roots greatly helped to develop our concepts
of ecology of micro-organisms on roots, the technique is limited to the surfaces
of washed roots - a disadvantage overcome by the application of electron mi-
croscopy to the rhizosphere, a technique which demonstrates the complexity
of this environment through which all nutrients and water must pass to reach
the root.

5.2 Scanning Electron Microscopy (S.E.M.)

The application of S.E.M. to the root-soil interface (GRAY 1967, DART 1971,
CAMPBELL and ROVIRA 1973, ROVIRA and CAMPBELL 1974, ELLIOTT et al. 1979)
has greatly improved our concept of the nature of the rhizosphere. S.E.M.
has the advantage over Transmission Electron Microscopy that relatively large
areas can be viewed and that the technique for preparing specimens is simpler.
However, one disadvantage of S.E.M. is that the dehydrated mucilages on roots
are opaque (Fig. 2) and hence the actual root surface can be viewed only when
this mucilage lifts off (Fig. 3).
1.2 The Significance of Rhizosphere Microflora and Mycorrhizas 67

Fig. 2. S.E.M. of a root of wheat (Triticum aestivum L.) ·which has been washed gently
to remove most of the rhizosphere soil. M mucige1, B bacterium, A soil aggregate
68 A.D. ROVIRA et al.:

Fig, 3. S.E.M . of a diseased wheat root near the edge of a lesion caused by the root
pathogen Gaeumannomyces gram in is (" Take-all "). The mucigel is disintegrating to expose
the bacteria at the rhizoplane. B bacterium, M mucigel
1.2 The Significance of Rhizosphere Microflora and Mycorrhizas 69

5.3 Transmission Electron Microscopy (T .E.M.)

5.3.1 General Description

Although SEM can be used to examine large areas of root, it can only reveal
ultrastructural details of materials near surfaces and cannot distinguish different
rhizosphere microorganisms unless they have distinctive shapes, sizes or charac-
teristic surface markings. In TEM of ultra-thin sections of root surfaces, how-
ever, different kinds of organism may be distinguished (FOSTER and ROVIRA
1976, 1978). By examining serial sections the distribution of micro-organisms
in the soil fabric from outside the rhizosphere into the root tissue can be deter-
mined. SEM and TEM therefore provide complementary views of rhizosphere
organization and both are needed (OLD and NICHOLSON 1975, BALENDRAU and
KNOWLES 1978). It is difficult to cut ultra-thin sections of roots and adhering
clay and sand particles; there are further problems of fixing and embedding
such heterogeneous material.
The ultrastructure of the root surface has been widely investigated by TEM
(SCOTT et al. 1958, JENNY and GROSSENBACHER 1963, DART and MERCER 1964,
LEPPARD 1974, FOSTER and ROVIRA 1976, 1978), but few of these studies have
been made on roots of field-grown plants. Usually plants have been grown
in reconstituted soils in pots and often the soils have been sterilized and known
organisms have been introduced (GREAVES and DARBYSHIRE 1972, DART and
MERCER 1964). In view of the complexity of natural rhizospheres, these artificial
root-soil interfaces may oversimplify rhizosphere ultrastructure; in this paper
we refer only to rhizospheres of roots from field soils.
It has been known for a long time that the epidermal cells of roots have
a complex multilamellate structure (BELFORD and PRESTON 1961, SETTERFIELD
and BAYLEY 1957, FOSTER 1962, PRESTON 1974). The outer layer of the cell
wall is largely mucilaginous but contains a few cellulose micro fibrils disposed
in open networks arranged in widely spaced lamellae. The inner layer is compact
and consists of dense lamellae of cellulose microfibrils. The precise nature of
the root-soil interface as observed by SEM or TEM depends on which of
these layers is exposed at the root surface either by the lytic action of soil
bacteria or by the methods used to extract the roots from soil and to prepare
them for electron microscopy. This has meant that the exact nature of the
root-soil interface has been a matter of some controversy. In the classical study
of JENNY and GROSSENBACHER (1963) the outer wall layer was thought to become
increasingly tenuous and it was necessary to use electron-dense markers to dem-
onstrate the surface. LEPPARD (1974) and LEPPARD and RAMAMOORTHY (1975)
demonstrated that under some conditions the surface of the root is decorated
with tufts of fibres at right angles to the root axis. However, SCOTT et al. (1958),
GREAVES and DARBYSHIRE (1972), DART and MERCER (1964), showed that muci-
lage is bounded by a thin but well-defined electron-dense membrane, which
SCOTT et al. (1958) termed a "cuticle".
Experiments with roots of several species grown in' natural soils have shown
that in the absence of mechanical abrasion from the surrounding soil, the muci-
lage is bounded externally by a delicate trilamellate electron-dense "cuticle"
70 A.D. ROVIRA et al.:

(Fig. 4). This layer may be quite thick in the Fabaceae; even in the presence
of bacteria, the root, at this stage has a smooth surface and the outline of
the individual root cells is quite distinct as seen in S.E.M.
Under natural conditions, however, where the root has to force its way
into the soil fabric, and where soil bacteria lyse its surface, the fine "cuticle"
is ruptured and the mucilages of the outer cell wall layer are released into
the soil fabric (Fig. 5). At this stage the root surface appears granular and
indistinct in both S.E.M. and T.E.M. and cell outlines are often obscure.
It has long been known that roots are covered with a mucilaginous material
(SCHWARTZ 1883, DAWES and BOWLER 1959). Roots of plants from some fami-
lies (Ericaceae, Fabaceae) secrete copious amounts of mucilage, but other species
produce little or none (Rosa, Dianthus, Chrysanthemum) (LEISER 1968). Within
the same soil, different species in the same family (e.g. Poaceae) may secrete
quite different amounts of mucilage (FOSTER 1978).
Most plant-derived mucilage is associated with the root cap and epidermal
cells in the root hair zone.

5.3.2 Origin and Fine Structure of Root Mucilage

Plant-derived mucilage is secreted by the dictyosomes (MOLLENHAUER and
MORRE 1966, LEECH et al. 1963), and pulse-labelling techniques have shown
that mucilage is transported through the cytoplasm in enlarged Golgi (dictyo-
some) vesicles to the plasmalemma where the contents of the vesicles are dis-
charged into the cell wall and thence onto the root surface (NORTHCOTE and
PICKETT-HEAPS 1966).
The precise ultrastructure of mucilage as seen by T.E.M. depends on the
conditions under which roots were grown and on methods used in the prepara-
tion for electron microscopy. Conventional glutaraldehyde/Os0 4 double fixa-
tion reveals a granular material containing two populations of particles differing
in size and electron density. When plants are water-stressed a particularly sticky
mucilage may form at root surfaces (NAMBIAR 1976, MARTIN 1977) and when
field-grown roots are treated with electron-dense colloids, changes in mucilage
structure may be detected as localized patches or layers of electron dense materi-
al. The chemical composition of mucilage may vary according to the environ-
mental conditions in the soil, and this may in tum affect the ion exchange
properties of the root surface (see below).

Fig. 4. T.E.M. of an ultrathin section of a clover (Trifolium subterraneum L.) root showing
the granular, mucilaginous, outer primary layer of the epidermal cell wall enclosed in
an electron dense cuticle. Conventional fixation. Cl clay, R Rhizobium, P primary cell
wall, C cuticle, S secondary cell wall

Fig. 5. T.E.M. of Paspalum root stained with lanthanum hydroxide. The mucilage of
the primary wall surrounds and embeds soil minerals and becomes infected by fungi
and bacteria. B Bacterium, W cell wall material, M mucigel
1.2 The Significance of Rhizosphere Microflora and Mycorrhizas 71
72 A.D. ROVIRA et al.:

5.3.3 Microbial Invasion of the Mucilage and the Formation of Mucigel

When plants are grown in natural soil the mucilage is quickly invaded by soil
microorganisms which sometimes penetrate the cuticle more or less intact (DART
and MERCER 1964), but in most cases the cuticle is soon removed (Fig. 5).
Further activity of the soil microflora removes the mucilage at the root
surface and eventually it is restricted to the grooves between the epidermal
cells (BOWEN and FOSTER 1978).
Conditions in the rhizosphere encourage microorganisms which secrete car-
bohydrate slimes and capsule materials (WEBLEY et al. 1965, FOSTER and ROVIRA
1978). Some of these materials stain with Luffs ruthenium redjOs04 complex
(CAGLE et al. 1972, FOSTER 1978) and many have a characteristic fibrous texture
so that carbohydrates from roots and microorganisms can be distinguished.
However, many bacteria and fungi produce gels which are granular, and which
may not react with RujOs, so that it is no longer possible to determine whether
the gel at the root surface is mainly derived from the root or from the microor-
ganisms; this complex carbohydrate mix is called mucigel.

5.3.4 Functions of Root Mucilage and Mucigel

The amount of organic material released from roots is high (SAMTSEVICH 1965,
MARTIN 1975) and unless the release of so much mucilage was of benefit to
the host such a loss of energy-rich material would be rejected. All nutrients
entering the root must first pass through the mucilage and mucigel. Electron
probe and heavy metal precipitation techniques have shown that mucigel
adsorbs many ions including CI-, POl-, Ca 2 + and Fe2+/3+, and can hold
them against mild leaching. Mucigel may serve to trap the transient flux of
nutrients released from dead microorganisms: this may be a function of the
COO- groups in acidic polysaccharides which JENNY and GROSSENBACHER
(1963) and BALANDREAU and KNOWLES (1978) suggest are important in plant
nutrition. LEPPARD (1974) and LEPPARD and RAMAMOORTHY (1975) suggest that
fibrils of acidic polysaccharides may be involved in the extraction of ions from
minerals and their movement into the root. The ion exchange properties of
roots may be in the root surface carbohydrates derived from roots and microbes.
Mucilage may support nitrogen-fixing bacteria (DOBEREINER 1974, KNOWLES
1977) as well as microorganisms which produce substances which are antagonis-
tic to root pathogens. Transfer of water in the vapour phase is very slow,
and one role of root mucigel may be to maintain contact between the minimal
voids zone near the root and the root surface and thus maintain a continuous
phase for water movement to the root as they shrink and swell with changes
in diurnal water stress.

5.3.5 The Outer Rhizosphere

The outflow of energy-rich mucilages, cell exudates and (later) celllysates results
in a proliferation of microorganisms at the root-soil interface. Microbial
numbers may rise in favourable micro sites to 200 x 10 9 cm - 3 (FOSTER and
ROVIRA 1978). There is also a great diversity of forms which can be distinguished
on the basis of cytoplasmic structure. Many forms in the grooves between the
I.2 The Significance of Rhizosphere Microflora and Mycorrhizas 73

epidermal cells may be restricted to this specialized habitat (FOSTER and ROVIRA
1978, OLD and NICHOLSON 1975).
Some stratification of bacteria occurs within the rhizosphere and many of
the unusual forms which are not obtained by traditional methods of plate count-
ing occur in the first 5 microns from the root surface. Further out from the
root the rhizosphere may be dominated by one or two morphological types.
The density of the bacterial population falls off within 20 11m from the surface,
a finding which supports the NEWMAN-WATSON (1977) model.
The main distinction between bacteria in the rhizosphere and those in the
bulk soil is that the rhizosphere forms are filled with storage materials such
as polyhydroxybutyrate (PHB), polysaccharide granules and polyphosphate
granules.

5.3.6 Invasion of the Root by Microorganisms

Bacteria attack the empty cells of the epidermis and cortex behind the root
hair zone (FOSTER and ROVIRA 1973, 1976, 1978, OLD and NICHOLSON 1975)
and the cortex of a normal healthy root may be inhabited by a wide range
of microorganisms without any sign of disease in the root. The cells fill with
bacteria, many of which are rich in PHB and polyphosphate. Some bacteria
attack the cell wall and eventually only the inner cell wall layers may rernain,
so that the cellulose fibrils are then exposed to the surface. Microorganisms
within the root show a narrow range of morphology and some small bacteria
seem to be found exclusively within the root. This specialized environment has
been called the endorhizosphere (BALANDREAU and KNOWLES 1978) and has
also been proposed as a site of non-symbiotic N 2 fixation (DOBEREINER and
DAY 1975).
The mixed microbial population of the endorhizosphere quickly causes the
collapse and decay of the outer cortical cells of the root; and in some annual
crops, such as wheat, the endodermis may be the functional surface of older
parts of the root system at anthesis.

6 The Role of Rhizosphere Microorganisms in Plant Nutrition

In view of the colonization of roots and rhizosphere by microorganisms the

question asked is "Does this microbial population influence the nutrition of
the plant?" Unless special precautions for axenic culture (viz. plants ,grown
under sterile conditions) are observed, it is not possible to study nutrient uptake
by plants without the microflora. The axenic culture of plants involves the
surface sterilization of seed followed by germination under sterile conditions
using agar on which contaminated seeds are detected. The germ-free seedlings
are transferred to sterile plant nutrient solution, sand or soil (BARBER 1967,
1978, ROVIRA and BOWEN 1969, REUSZER 1949, 1962).
Rhizosphere microorganisms can affect plant nutrition through their influ-
ence on (a) availability of nutrients, (b) growth and morphology of roots, (c)
nutrient uptake processes and (d) physiology and development of plants.
74 A.D. ROVIRA et al.:

6.1 Availability of Nutrieuts

6.1.1 Nutrient Release and Immobilization

KATZNELSON and ROUATT (1957) found that the respiration of rhizosphere soil
was two to four times that of non-rhizosphere soil; this differential was similar
for both unamended soil and soil with added amino acids. Whether this greater
metabolic activity in rhizosphere soils releases or immobilizes plant nutrients
will depend upon the nutrient status of the soil. An example of this is the
finding by BARBER (1966) that under low phosphate conditions the micro flora
competes with the root for phosphate and reduces the uptake of phosphate
into the plant.

6.1.2 Nitrification and Denitrification

There is now considerable evidence that roots of some plants, especially grasses
and trees, inhibit nitrification. This inhibition will influence the whole nitrogen
balance of the system (THERON 1966, WOLDENDORP 1963, RICE 1974).

6.1.3 Nitrogen Fixation

The fixation of nitrogen under temperate swards was found by PARKER (1957)
to average 60 kg ha -1 p.a. over a 3-year period of undisturbed sward. Consider-
able interest in the nitrogen fixation in the rhizospheres of tropical grasses
has been aroused by the findings of D6bereiner and associates (DOBEREINER
et al. 1972, DART and DAY 1975). The fixation of nitrogen in undisturbed grass
swards in tropical soils is significant in maintaining production of pastures
free from legumes and is discussed in Chap. II, 3, this Volume.

6.1.4 Phosphate Availability

The ability of rhizosphere bacteria to dissolve water-insoluble forms of phos-
phate has been demonstrated by GERRETSON (1948), SPERBER (1958) and DUFF
et al. (1963), but it is unlikely this mechanism (which is dependent upon the
production of acid) will operate to any great extent in the rhizosphere. The
production of acid is dependent upon an abundant supply of carbohydrate
and population densities found in nutrient media which are much higher than
those which occur on roots. In his review of the role of microorganisms in
the phosphorus cycle in soil, HAYMAN (1975) points out that those symbiotic
organisms involved in phosphate uptake, e.g. mycorrhizas (di~cussed later. in
this chapter) have an enormous advantage over free-living organisms and hence
will have a much greater impact on phosphate uptake.

6.1.5 Minor Nutrients

6.1.5.1 Manganese
TIMONIN (1946) found that the susceptibility of different oat varieties to manga-
nese deficiency was related to the abundance of manganese-oxidizing bacteria
1.2 The Significance of Rhizosphere Microflora and Mycorrhizas 75

supported in the rhizosphere. This influence of the micro flora could be elimi-
nated by partial soil sterilization.

6.1.5.2 Molybdenum
In a study of the causes for different levels of dental caries in residents in
two districts in New Zealand, it was found that the vegetables from these districts
differed in their content of molybdenum. These differences were shown by
LOUTIT (1968) and LoUTIT et al. (1972) to be due to differences in the rhizosphere
microfloras: those organisms from the soil producing low molybdenum plants
accumulated more molybdenum than organisms from the soil producing vegeta-
bles with higher molybdenum levels.

6.2 Growth and Morphology of Roots

6.2.1 Root Length and Root Hairs

Rhizosphere microorganisms can reduce root length (BOWEN and ROVIRA 1961 a)
and root hair length and density (BOWEN and ROVIRA 1961 b, WELTE and TROLL-
DENIER 1965, DARBYSHIRE and GREAVES 1970). Such effects will influence the
uptake of slowly diffusible nutrients such as phosphate from soil by roots.
The reductions in root length and root hair growth caused by microorganisms
could lead to a 50% to 60% reduction in the uptake of ions of low mobility
in soil.

6.2.2 Proteoid Roots

PURNELL (1960) reported localized intense lateral root production on many
Proteaceae occurring in low nutrient soils and referred to them as "proteoid"
roots. Clumps of proteoid roots form dense mats in the surface soil and, al-
though they are particularly well developed in pockets of humus-rich soil, they
are not restricted to these parts of the profile. Similar roots are now known
to occur in other families, e.g. on the legume Viminaria juncea (LAMONT 1972),
species of lupin (TRINICK 1977) and on Casuarina equisetifolia (H.G. DIEM per-
sonal communication); future studies may show them to be much more wide-
spread. MALAJCZUK and BOWEN (1974) and LAMONT and McCOMB (1974)
showed proteoid roots to be caused by rhizosphere microorganisms.
In phosphate-poor soils, the production of proteoid roots has resul~ed in
a doubling of growth of 4-month-old seedlings of Banksia grandis probably
due to the intense exploration of soil for poorly mobile ions such as phosphate
and a higher phosphate influx to proteoid roots (MALAJCZUK and BOWEN 1974).
The physical nature of proteoid roots and their position near the soil surface
make them ideally suited for uptake of nutrients leached from leaves and litter
and they probably assist in the tightly coupled nutrient cycles so important
in ecosystems of low fertility soils. It is highly probable that further study
of plants in low nutrient soils, will produce other instances of microbial modifi-
cation of roots which lead to increased nutrient uptake.
76 A.D. ROVIRA et al.:

6.3 Nutrient Uptake Processes

The microorganisms associated with plant roots have been shown to affect
the actual processes of nutrient uptake and translocation as described in reviews
by BARBER 1978, BOWEN and ROVIRA 1969. The influence of the rhizosphere
population on phosphate uptake and translocation is affected by the plant age;
with seedlings the microflora increases uptake, translocation and incorporation
(BOWEN and ROVIRA 1966, ROVIRA and BOWEN 1969) while with older plants
the microorganisms compete with the roots for phosphate and reduce its uptake
into the plant (BARBER 1966, BARBER et al. 1976). With 8-day-old barley seed-
lings the presence of rhizoplane micro flora increased the translocation of phos-
phate to tops from 15 to 20 p mol mg- 1 in high phosphate plants and 8 to
22 p mol mg- 1 in low phosphate plants during a 30-min uptake period. By
contrast in 3-week-old barley plants, plants with microorganisms on their roots
took up 50% more phosphate over 24 h than did sterile plants, but the propor-
tion of total phosphate translocated to tops was 48% and 34% for sterile and
non-sterile plants respectively.

6.4 Physiology and Development

The rate of plant development, including flowering time, is affected by certain

components of the rhizosphere micro flora and this effect has been demonstrated
by the application of cultures of known bacteria onto seed (MISHUSTIN and
NAUMOVA 1962, ROVIRA 1965b, BROWN et al. 1968, BROWN 1974, 1976). These
effects are now attributed to the production of plant growth-regulating sub-
stances such as indolyl acetic acid and gibberellic acid (BROWN 1972). The
inoculation of crops was practised extensively in the U.S.S.R. but the unpredict-
ability of crop responses to such seed inoculation has thrown doubt upon the
practical value of this technique for broad-scale farming (MISHUSTIN and
NAUMOVA 1962, RIDGE and ROVIRA 1968).

7 Mycorrhizas

The roots of most higher plants are usually mycorrhizal, the three main types
of mycorrhizal association involved with inorganic nutrition being: (1) ecto-
trophic mycorrhizas, restricted mainly to a few tree families (e.g. Pinaceae,
Myrtaceae, Fagaceae); (2) ericaceous mycorrhizas of Ericaceae; Epacridaceae,
and Empetraceae and (3) vesicular-arbuscular (v.a.) mycorrhizas infecting most
other higher plants except for the majority of species in a few families, primarily
in the order Centrospermae and in the families Brassicaceae, Fumariaceae, Cy-
peraceae, Urticaceae, Commelinaceae (GERDEMANN 1975).
So far only phycomycetes of the family Endogonaceae have been shown
to cause v.a. mycorrhizas; the ascomycete Pezizella ericae has been identified
as one of the endophytes of Vaccinium macrocarpon (READ 1974) forming erica-
1.2 The Significance of Rhizosphere Microflora and Mycorrhizas 77

fs hn epi x

200 }J
Fig. 6. Comparison of A uninfected root and B ectotrophic mycorrhiza. Is fungal sheath ;
hn Hartig net; epi epidermis; x xylem. (Adapted from CHILVERS and PRYOR 1965. By
courtesy of the Australian Journal of Botany)

ceous mycorrhizas. However, a wide range of basidiomycetes, ascomycetes and

at least one zygomycete (Endogone) are the cause of ectotrophic mycorrhizas.
Ectotrophic mycorrhizal fungi usually form a sheath of hyphae (Fig. 6)
which encloses short, branching lateral roots. From the sheath, root-like bundles
of hyphae (mycelial strands) pass outwards into the soil fabric often permeating
it and organic matter (see BOWEN 1973). Hyphae also grow inwards into the
host tissue to form the Hartig-net in the root cortex, thus giving an intimate
connection between soil distant from the root and the cortical cells of the root.
The fine structure of ectomycorrhizas is discussed by FOSTER and MARKS (1973).
In the endotrophic associations, hyphae actually enter the cells of the outer
cortex of the roots. Here they become much divided to form tree-like "arbus-
cules" each branch of which is surrounded by the plasmalemma of the host.
In v.a. mycorrhizas, the arbuscule increases the surface area of contact between
the fungus and root cell by two to three times and some 23-fold increase in
cytoplasm occurs in the plant cell (Cox and TINKER 1976), indicating an effect
of the micro-organisms on the plant's metabolism. Vesicular-arbuscular mycor-
rhizas also often have large ellipsoid vesicles between or within the cell. No
well-defined fungus sheath occurs with v.a. mycorrhizas, but all can occur with
ericaceous mycorrhizas. In endomycorrhizas, as with ectomycorrhizas, extensive
fungal growth into soil and litter can occur (Fig. 7).
78 A.D. ROVIRA et al.:

)
y
C
Fig. 7. Structure of v.a. mycorrhiza. c cortex; r.h . root hair; fm . fungal mycelium;
s spore; v vesicle; a arbuscule. (SANDERS et al. 1975. By courtesy of Academic Press)

7.1 Plant Responses to Infection

Table 1 illustrates the increases in plant growth and differences in nutrient status
caused by mycorrhizas of the three main types; there is now a voluminous
literature reporting such results. Increases in absorption of a number of ions
(Table 1) can result from the relief of a primary deficiency. Thus, HATCH (1937)
found ectomycorrhizal Pinus strobus to have twice the nitrogen ' uptake of non-
mycorrhizal plants. A recalculation of his data shows nitrogen absorption g - l
mycorrhizal roots was 1.6 times that of non-mycorrhizal roots. However, the
phosphate absorption of mycorrhizal roots g-l was 2.9 times that of noo-
mycorrhizal root systems and he may have been dealing primarily with a phos-
phate deficiency.
The concentration of non-limiting elements, but not the total uptake, may
decrease in mycorrhizal plants (Table 1, lines 4, 5) by dilution caused by the
increased growth of mycorrhizal plants. Mycorrhizal increases of uptake of
1.2 The Significance of Rhizosphere Microflora and Mycorrhizas 79

Table 1. Plant response to mycorrhizal infection

Type of mycor- Dry wt. N P K Ca

rhiza and host

1. Ectomycorrhizas (Pinus elliottii var. elliotti a)

Uninoculated 0.26 d 2.4 e 0.22e 2.27e
Rhizopogon sp. 1.96 25.2 2.2 23.8
Isolate R8122 2.81 38.9 3.9 35.6
2. v.a. Mycorrhizas (Rough lemon b)
Uninoculated 0.67 f 0.10% 1.51% 3.15%
Inoculated 11.29 0.17% 0.67% 3.05%
3. Ericaceous mycorrhiza (Vaccinium macrocarpon C )
Uninoculated 184 e 0.81 e
Inoculated 235 1.82

a LAMB and RICHARDS (1971) d g plane 1

b KLEINSCHMIDT and GERDEMANN (1972) e mg plant- 1
C READ and STRIBLEY (1973) f g pot-1

a nutrient in limited supply have been shown with phosphate (ectomycorrhizas

and v.a. mycorrhizas; see BOWEN 1973, SANDERS et al. 1975, and ABBOTT and
ROBSON 1977), with nitrogen (ericaceous mycorrhizas; STRIBLEY and READ 1976,
see also BOWEN and SMITH 1981), and zinc (v.a. mycorrhizas; GILMORE 1971);
and increases in absorption of potassium, sulphate and 90Sr (a calcium tracer)
have also been reported (BOWEN 1980). Enhanced ion absorption from soil
occurs more commonly with poorly mobile ions (see below).
Increase:; in growth due to mycorrhizas are often greatest at sUb-optimal
levels of the limiting nutrient. At very low levels of nutrients there can be
altogether too little for growth of the particular plant species and at high levels
the mycorrhiza may be superfluous to the plant's needs; e.g. ABBOTT and
ROBSON (1977) and MENGE et al. (1978) (v.a. mycorrhizas and phosphate nutri-
tion of clover and citrus respectively); STRIBLEY and READ (1976) (ericaceous
mycorrhizas and nitrogen nutrition of Vaccinium macrocarpon). Indeed, at high
nutrient levels mycorrhiza development can be reduced or inhibited (as in MOSSE
1973). In at least one instance depression of mycorrhizal plants at supra-optimal
levels of phosphate application has been ascribed to toxicity caused by enhanced
absorption by the mycorrhizas (MOSSE 1973).
The large increases in plant growth from mycorrhizal infection suggest esti-
mates of production potentials of soils may be seriously underrated, for almost
no agronomic-nutrition studies of plants have ensured adequate mycor'rhizal
infection with highly efficient fungus. The search for plants and genotypes capa-
ble of tolerating nutritionally poor conditions may be assisted by the use of
appropriate mycorrhizal fungi.

7.2 Mechanisms of the Response

Inflows of phosphate to v.a. mycorrhizal plants have been measured at up
to 14 times that into non-mycorrhizal plants in soil (SANDERS and TINKER 1973,
80 A.D. ROVIRA et al.:

SMITH et al. 1979). The specific absorption rates of ammonium are greater with
mycorrhizal than with non-mycorrhizal V. macrocarpon (ericaceous mycorrhi-
zas) (STRIBLEY et al. 1975). However, these data themselves do not indicate
the mechanisms involved. The three general areas which could affect uptake
are: the availability of the nutrients in soil, the effects of the fungi on nutrient
absorption by the root itself, and the roles of the fungi in nutrient absorption
by the symbiosis.

7.2.1 Nutrient Availability

Several studies indicate that v.a. mycorrhizas use the same sources of inorganic
phosphate as uninfected plants (e.g. HAYMAN and MOSSE 1972, BARROW et al.
1977), and this is probably also the case for ectomycorrhizas (BOWEN 1973).
Both the plants, mycorrhizas and uninfected roots have surface phosphatases
(BOWEN 1973, BARTLETT and LEWIS 1973) and, given the opportunity, can absorb
simple organic phosphates.
The position with nitrogen nutrition (reviewed by BOWEN and SMITH 1981)
is less clear. Although some mycorrhizal fungi possess nitrate reductase and
can use nitrate, most prefer ammonium sources; little, if any, nitrate absorption
by mycorrhizas has been demonstrated so far. The fungi and higher plants
can each use a wide range of relatively simple organic nitrogen compounds,
provided they can compete effectively with soil micro flora for them; but complex
organic nitrogen sources are generally not used. Claims for atmospheric nitrogen
fixation by mycorrhizal fungi can be interpreted more satisfactorily as fixation
by free-living prokaryotic organisms on the surface of non-sterile mycorrhizal
roots.

7.2.2 Absorption Characteristics of the Root

Most studies of ion absorption from solution do not distinguish between absorp-
tion by the fungus component and effects of the fungus on the absorption
characters of the higher plant. The probable production of cytokinins or other
growth-regulating compounds in the mycorrhizal association are manifested
in enlargement of the cells or the nucleus, increase of host cytoplasm (Cox
and TINKER 1976) and, at least in ectomycorrhizas, the plant cells have increased
longevity. However, the existing data do indicate increases of phosphate absorp-
tion by ectomycorrhizas and by v.a. mycorrhizas from solutions are due primar-
ily to absorption by theJungus component (HARLEY 1969, BOWEN et al. 1975).

7.2.3 Absorption by the Fungus Component

As indicated above, enhancement of ion uptake (in monocultures) by mycorrhi-
zas occurs more commonly with poorly mobile ions in soil. There is a close
relation between rooting intensity and absorption of such ions (BARLEY 1970,
NYE and TINKER 1977). Uninfected roots .quickly deplete a narrow zone of
soil around them for ions such as phosphate, a zone more or less delineated
by the root hairs (see NYE and TINKER 1977) after which replenishment is slow.
The growth of fungal hyphae into soil and litter can dramatically increase the
1.2 The Significance of Rhizosphere Microflora and Mycorrhizas 81

"rooting density" and the volume of soil being effectively used by the plant.
Uptake of nutrients by the mycorrhizal hyphae and their translocation to the
root has been experimentally demonstrated for phosphate (MELIN and NILSSON
1950, SKINNER and BOWEN 1974a, HATTINGH et al. 1973 and PARSON and REID
1973), for ammonium (MELIN and NILSSON 1952, STRIBLEY and READ 1974),
calcium (MELIN and NILSSON 1955) and for zinc and sulphate (COOPER and
TINKER 1978). There could also be an enhanced absorption of potentially toxic
heavy metals such as lead and arsenic (see BOWEN 1978) but this requires further
study. SANDERS and TINKER (1973) calculated the phosphate fluxes in hyphae
of v.a. mycorrhizas to account for the increased inflow to mycorrhizal roots.
SANDERS et al. (1977) found that inflow of phosphate to onions growing in
soil was closely related to the development of external hyphae with four v.a.
mycorrhizal fungi; but COOPER and TINKER (1978) recorded little close relation
between influx of phosphate, zinc and sulphate and length of mycorrhizal
hyphae, and considered uptake by the fungus to be controlled by the host.
Obviously a great deal more study is required of such possible factors, but
it is necessary also to know whether uptake occurs along all of the hyphae.
Individual hyphae of v.a. mycorrhizal fungi have been observed for up to 8 cm
from roots (RHODES and GERDEMANN 1975) and mycelial strands of ectomycor-
rhizal fungi can intensively grow through soil and litter for up to 12 cm from
the root (SKINNER and BOWEN 1974b). SANDERS and TINKER (1973) estimated
there were up to 80 cm hyphae cm -1 of v.a. mycorrhizal root. However, there
is a serious lack of experimental data on soil and plant factors affecting fungal
growth through soil, and on the epidemiology of the infection: both these pa-
rameters would be expected to affect response greatly, with soil, fungus strain,
and plant species being important variables (SKINNER and BOWEN 1974 b, BEVEGE
and BOWEN 1975, SANDERS et al. 1977).
Influx of ions at the surface of the mycorrhiza may be no higher than
that at the extending zone of an uninfected root, but the influx is much more
sustained (BOWEN 1973). An enhanced absorption by mycorrhizas has been
recorded from solutions (where transfer from the medium to the root surface
is not limiting) for phosphate (HARLEY 1969, BOWEN 1973, CRESS et al. 1979),
sulphate (MORRISON 1963) and zinc (BOWEN et al. 1974). Absorption of these
ions is metabolically mediated and for phosphate at least, kinetic analyses have
suggested at least two carrier systems (BOWEN and THEODOROU 1967, for both
ectomycorrhizas; and CRESS et al. 1979, for v.a. mycorrhizas). The ectomycor-
rhiza studies showed mycorrhizas caused by different fungi differed both in
the number of carrier sites and their affinity for phosphate. The studies on
v.a. mycorrhizas indicated that at low solution concentrations of phosphate
(1-20 /lm), the carrier systems of mycorrhizas (plus roots) had a greater affinity
for phosphate than those of roots alone, but that the mycorrhizas had the
same number of carrier sites. Unfortunately, the Hofstee plot interpretation
approach is not a unique interpretation of uptake analyses and furthermore
the kinetics may have been affected by the uninoculated roots having a higher
phosphate concentration than the mycorrhizal roots did. Suggestions that my-
corrhizas may be able to absorb phosphate at lower concentrations than the
uninfected root can (CRESS et al. 1979) need further evaluation but may be
of only limited relevance in phosphate uptake from soil.
82 A.D. ROVIRA et al.:

In all types of mycorrhizas, phosphate is converted to inorganic polyphos-

phates (Cox et al. 1975, LING-LEE et al. 1975, CALLOW et al. 1978), in which
form it is translocated in v.a. mycorrhizas (Cox et al. 1975). There is consider-
able storage of phosphate in ectomycorrhizas, probably as polyphosphate, dur-
ing periods of plentiful phosphate supply (HARLEY 1969). However, v.a. mycor-
rhizas do not appear to have as high a storage capacity (BOWEN et al. 1975)
and after a short lag period (COOPER and TINKER 1978) phosphate is transferred
rapidly to the higher plant via the arbuscules. Electron-probe analysis indicates
the arbuscules to be high in phosphorus (SCHOKNECHT and HATTINGH 1976).
It has been suggested that phosphate release depends on degeneration of the
arbuscule (KINDEN and BROWN 1975) but it is likely that the major part of
the phosphate is transferred from the living arbuscule to the plant as ortho-
phosphate before arbuscule degeneration (Cox and TINKER 1976, BOWEN et al.
1975, COOPER and TINKER 1978). GIANINAZZI et al. (1979) have speculated on
the roles of acid and alkaline phosphatases within arbuscules of v.a. mycorrhizal
fungi, and suggested vacuolar alkaline phosphatase may be involved in the
transport of phosphate within v.a. fungi. They have also described an alkaline
phosphatase in v.a. fungi specific to mycorrhizal infection (GIANINAZZI-PEARSON
and GIANINAZZI 1978).
More is needed to be known about ammonium-assimilating enzymes of
roots and their mycorrhizal fungi before authoritative statements can be made
about the relative rates of uptake in the symbionts (BOWEN and SMITH 1981).
Incorporation of ammonium into amino acids results in the production of excess
protons (one H+ per ammonium, RAVEN and SMITH 1976). All plants and micro-
organisms, including mycorrhizas, regulate their intracellular pH by H+ extru-
sion to the bathing medium or soil, thus reducing the pH in culture medium
or rhizosphere (NORKRANS 1950, SMILEY 1974). In ectotrophic beech mycorrhi-
zas, and possibly other mycorrhizas, NHt is assimilated to glutamine in the
fungal sheath, and is then transferred to the root. Proton extrusion must take
place from the fungal cells, which may assume both assimilatory and pH regula-
tory roles from the root (discussed by RAVEN et al. 1978). S.E. SMITH (personal
communication) has suggested that if nitrate uptake and assimilation by mycor-
rhizal fungi were to occur and pH regulation were by OH- extrusion rather
than by cation/carboxylate storage, mycorrhizal development might result in
considerable savings of photosynthate used for carboxylate production to regu-
late pH. It is significant in this regard that potassium content of many v.a.
mycorrhizal plants is often considerably lower than that of non-mycorrhizal
plants (S.E. SMITH personal communication).

7.3 Energy Requirements of Mycorrhizas

The energy sources for the fungus are derived mainly from the higher plant.
In ectomycorrhizas, sucrose in the higher plant is hydrolyzed to glucose and
fructose, and the concentration gradient is maintained by their conversion into
trehalose, mannitol and glycogen (LEWIS and HARLEY 1965, BEVEGE et al. 1975).
Such fungal sugars do not occur with v.a. mycorrhizas (BEVEGE et al. 1975,
HAYMAN 1974), and concentration of sugars may be kept low by their use
1.2 The Significance of Rhizosphere Microflora and Mycorrhizas 83

in growth, conversion to organic acids or storage as glycogen. The assimilate

cost of v.a. mycorrhizas has received little study. Sometimes, but not always,
ectomycorrhizas can be an assimilate sink; it has been suggested that the growing
root and ectomycorrhizas both act as assimilate sinks, the mycorrhizal sink
becoming major when root elongation slows and its sink strength declines
(BOWEN 1978). Usually the benefits of enhanced nutrition by mycorrhizas far
outweighs the assimilate costs but there are suggestions that in some highly
fertile soils mycorrhizas may be superfluous but still use up assimilate. This
was proposed as the reason for occasional growth reductions by v.a. mycorrhizas
in highly fertile soils (CRUSH 1976).

7.4 Overview of Mycorrhizas

Most studies of mycorrhizas have focussed on phosphate uptake from nutrient-
poor soils. Although their key role is in nutrient (and possibly water) uptake
from soils there is little doubt that some aspects of the physiology of the mycor-
rhizal plant are different from that of the uninfected plant, e.g. the infection
causes a marked stimulation of polyphenols, and terpenes in roots (FOSTER
and MARKS 1967, KRUPA and FRIES 1971). However, lack of appropriate con-
trols has precluded a good comparison of the physiology of mycorrhizal and
non-mycorrhizal plants in most studies. There is some evidence that differences
occur, such as a decreased nematode reproduction or a decreased length of
mycorrhiza root in some studies, (SCHENCK et al. 1975); SCHONBECK (1979)
reported an increased susceptibility of endomycorrhizal plants to foliage diseases
but decreased susceptibility to root diseases.
Even in nutrient absorption the overview of mycorrhizas may have been
too restricted. BOWEN (1980) has suggested they are more probably seen as
an alternate root strategy for ion absorption in a great many situations. In
many situations other than nutrient-poor soils, it is now well accepted that
mycorrhizal responses and plant dependence on mycorrhizas are maximal with
species developing low root densities, often with poor root hair formation
(BAYLIS 1975, CRUSH 1974). However, the growth of fungi into soil and litter
also enables mycorrhizal plants to compete effectively with soil micro-organisms
for available nutrients (BOWEN 1968, 1980, GADGIL and GADGIL 1975). Further-
more, the dual uptake system, each member with its own properties, allows
more flexibility in handling potentially deleterious situations. As long as the
mycorrhizal fungus is not as susceptible to a deleterious soil factor as the root,
the mycorrhizal mode of nutrition can be used by the plant to combat the
deleterious factor. Such factors can include root disease, high soil salinity,disad-
vantageous soil pH, and high soil temperatures (BOWEN 1980), e.g. this may
explain differences in mycorrhizal and non mycorrhizal plant growth with regard
to soil pH and soil temperature demonstrated by MOAWAD (1979). The reasons
for differing sensitivities to environmental factors between different mycorrhizal
fungi and between the fungi and their plants is an area awaiting more study
by physiologists.
Recently, SUTTON and SHEPPARD (1976) showed, by inoculating pots of steril-
ized dune sand with Glomus, that endomycorrhizal Phaseolus vulgaris had some
84 A.D. ROVIRA et al.:

five times (by weight) more sand aggregates kg- 1 than did non-mycorrhizal
plants of similar size, the sand grains being held together by the fungal hyphae.
TISDALL and OADES (1979) ascribed the greater efficiency of ryegrass compared
with clover roots in stabilizing soil aggregates in a loam, to the ryegrass support-
ing a larger population of v.a. mycorrhizal hyphae in soil. The hyphae were
covered with a layer of amorphous material, probably polysaccharide, to which
clay particles were firmly attached.
The increase in ion uptake from endomycorrhizas can have large conse-
quences for nitrogen fixation by legumes and by non-legumes such as Casuarina
spp .. v.a. mycorrhizas have been demonstrated to increase uptake of phosphate,
sulfate and zinc, and probably would increase uptake of cobalt, molybdenum,
copper, and iron, all of which are involved in nitrogen fixation. MOSSE et al.
(1976) found clover, Centrosema and Stylosanthes did not nodulate in grossly
phosphate-deficient soil if they were not also mycorrhizal. SMITH et al. (1979)
demonstrated an almost quadrupling of nitrogenase activity in mycorrhizal
clover compared with non-mycorrhizal clover of the same nutrient content.
The biological variability between mycorrhizal fungi, both in their infection
of the plant under various soil conditions and growth into soils (BEVEGE and
BOWEN 1975, SANDERS et al. 1977) with different soil-plant-fungus combina-
tions, leads to very great differences between isolates of mycorrhizal fungi in
their stimulation of the plant (Table 1, lines 2, 3). Modification of the one
soil similarly leads to different responses to different fungi, e.g. MOSSE (1972)
found that stimulation of plant growth with one v.a. mycorrhizal fungus was
three times greater than with another in a soil modified to pH 5.8 although
both behaved similarly at pH 4.8. Similarly, management practice will change
the dominant type of fungus present. MOSSE (1977) recorded that the addition
of phosphate to some phosphate-deficient soils advantaged the introduced v.a.
mycorrhizal fungi in competition with the indigenous mycorrhizal fungi.
In brief, the soil-mycorrhizal-plant system leads to a much more dynamic
picture of nutrition of plants and their physiology, and increased scope for
management of nutrition systems, than the simple two-membered concept of
plant and soil.

8 General Conclusions

"Traditional" plant physiology has usually considered only the higher plant.
However, the wide range of microorganisms associating with ro<;>ts and interact-
ing intimately with them adds a new dimension to our concept of "normal"
plants. These microorganisms, especially those that function symbiotically with
the higher plant, e.g. mycorrhizal fungi, can have a large impact on uptake
of plant nutrients from soils: they can be regarded as an extension of the root,
and another agent of the plant plasticity, complementary to genetic plasticity,
selecting species and genotypes coping with low fertility sites.
The last 10 years has seen considerable conceptual and experimental develop-
ments in rhizosphere biology. These have been helped considerably by the appli-
1.2 The Significance of Rhizosphere Microflora and Mycorrhizas 85

cation of electron microscopy of the root-soil interface (including the applica-

tion of histochemical methods), by the application of traditional population
dynamics and simulation modelling and, to a limited extent, by physiological
approaches. This microbial dimension to studies of plant biology can be a
rewarding field for plant physiologists. For example, rhizosphere micro-organ-
isms can double the losses of assimilates from "healthy" roots and to a level
where it is immediately apparent that they must be considered in any attempt
to construct an assimilate budget for the plant. Questions can be raised on
the mechanisms of assimilate loss from roots and the relation between loss
and composition of the root cells, which is in tum affected by the rest of
the plant's physiology. With mycorrhizal systems it is particularly appropriate
to consider the energy costs of the symbiosis, the mechanisms of transfer of
ions and metabolites between fungus and host, not to mention the physiology
of infection and the fine balance between symbiosis and parasitism.
Interesting questions. are whether micro-organisms associated with roots
change the plants' physiology and the biochemical bases of such effects. What
are the nature of the compounds produced by microorganisms which enhance
assimilate loss, increase phosphate translocation, induce earlier flowering and
cause extensive root modification important in nutrient uptake?
In view of these effects of rhizosphere micro-organisms on plants is it possible
to manipulate the microflora in the field? The extensive Russian work on inocu-
lation of seed with Azotobacter spp. and Bacillus spp. with claims of large
yield increases has been questioned; MISHUSTIN and NAUMOVA (1962) found
that inoculation increased crop yields significantly in <;me third of field trials.
Similar results by RIDGE and ROVIRA (1968) and BROWN (1976) demonstrate
that some manipulation of the microflora is possible. However, studies on the
distribution of the introduced bacteria show little spread from the seed down
the roots in unsterilized soil, so the effects result from a localized concentration
of the bacteria around the seed. Most colonization of roots results from contact
with micro-organisms in the soil; these organisms will certainly be the primary
colonizers, making it difficult for introduced bacteria to spread and find sites
which provide energy materials for multiplication. An alternative strategy for
establishing favourable rhizosphere microflora may be to grow crops which
build up the desired organisms which will then be placed strategically through
the soil to colonize roots of subsequent crops.
The successful introduction of efficient species of ectomycorrhizal fungi by
inoculation of pine seed has been demonstrated in both field and glasshouse
trials with corresponding growth responses in the trees (THEODOROU and BOWEN
1970). The complex interactions between introduced micro-organisms in the
rhizosphere are demonstrated in the experiments by BOWEN and THEODOROU
(1979), who found that the growth of ectotrophic mycorrhizas on pine roots
was increased, decreased or not affected by different species of bacteria. More
studies of this type are needed to understand the basic principles of population
dynamics in the rhizosphere before manipulation of the rhizosphere becomes
a reality in the field.
Finally, we wish to emphasize that most studies on plant physiology and
the uptake of nutrients are conducted on plants which support microorganisms
86 A.D. ROVIRA et al. :

in and around roots. It will be necessary, in future, in critical studies on the

uptake of nutrients by roots to consider, and if necessary, allow for the involve-
ment of microorganisms in the uptake of nutrients.

Acknowledgements. We thank Dr. R. CAMPBELL for the preparation of the SEM speci-
mens, Miss Y. McEwAN and Mrs. P. UDOMPONGSANON for preparation of the TEM
specimens and Mr. T.W. COCK and Mr. J.A. COPPI for preparing the plates for publica-
tion.

References

Abbott LK, Robson AD (1977) Growth stimulation of subterranean clover with vesicular
arbuscular mycorrhizas. Aust J Agric Res 28: 639-649
Balandreau J, Hamad-Fares I (1975) Importance de la fixation d'azote dans la rhizosphere
du riz. Soc Bot Fr, Colloq Rhizosphere:l09-119
Balandreau J, Knowles R (1978) The rhizosphere. In: Dommergues YR, Krupa SV
(eds) Interactions between non-pathogenic soil microorganisms and plants. Elsevier,
Amsterdam New York
Barber DA (1966) Effect of microorganisms on nutrient absorption by plants. Nature
212:638-640
Barber DA (1967) The effects of microorganisms on the absorption of inorganic nutrients
by intact plants. I. Apparatus and culture techniques. J Exp Bot 18: 163-169
Barber DA (1978) Nutrient uptake. In: Dommergues YR, Krupa SV (eds) Interactions
between non-pathogenic soil microorganisms and plants. Elsevier, Amsterdam New
York
Barber DA, Gunn KB (1974) The effect of mechanical forces on the exudation of organic
substances by the roots of cereal plants grown under sterile conditions. New Phytol
73:39-45
Barber DA, Martin JK (1976) The release of organic substances by cereal roots into
soil. New Phytol 76:69-80
Barber DA, Bowen GD, Rovira AD (1976) Effects of Inicroorganisms on the absorption
and distribution of phosphate in barley. Aust J Plant Physiol 3: 801-808
Barley KP (1970) The configuration of the root system in relation to nutrient uptake.
Adv Agron 22: 159-201
Barrow NJ, Malajczuk N, Shaw TC (1977) A direct test of the ability of vesicular arbuscu-
lar mycorrhiza to help plants take up fixed soil phosphate. New Phytol 78:269-276
Bartlett EM, Lewis DH (1973) Surface phosphatase activity of mycorrhizal roots of
beech. Soil BioI Biochem 5: 249-257
Baylis GTS (1975) The magnoloid root and mycotrophy in root systems derived from
it. In: Sanders FE, Mosse B, Tinker PB (eds) Endomycorrhizas. Academic Press,
London New York
Belford DS, Preston RD (1961) The structure and growth of root hairs. J Exp Bot
12: 157-168
Bevege DI, Bowen GD (1975) Endogone strain and host plant differences in development
ofvesicular-arbuscular mycorrhizas. In: Sanders FE, Mosse B, Tinker PB (eds) Endo-
mycorrhizas. AcadeInic Press, London New York
Bevege DI, Bowen GD, Skinner MF (1975) Comparative carbohydrate physiology of
ecto- and endomycorrhizas. In: Sanders FE, Mosse B, Tinker PB (eds) Endomycorrhi-
zas. Academic Press, London New York
Billes G, Cortez J (1975) Etude de la production de polysaccharides bacteriens au niveau
de la rhizosphere de Festuca arundinacea. Soc Bot Fr Colloq Rhizosphere:41-45
Bowen GD (1968) Phosphate uptake by mycorrhizas and uninfected roots of Pinus radiata
in relation to root distribution. Proc 9th Int Cong Soil Sci 2: 219-228
1.2 The Significance of Rhizosphere Microflora and Mycorrhizas 87

Bowen GD (1969) Nutrient status effects on loss of amides and amino acids from pine
roots. Plant Soil 30: 139-142
Bowen GD (1973) Mineral nutrition in Ectomycorrhizae. In: Marks GC, Kozlowski
TT (eds) Ectomycorrhizae. Academic Press, London New York
Bowen GD (1978) Dysfunction and shortfalls in symbiotic response. In: Horsfall JD,
Cowling EB (eds) Plant disease - an advanced treatise, vol III. Academic Press,
London New York
Bowen GD (1979) Integrated and experimental approaches to study the growth of organ-
isms around root and seeds. In: Schippers B, Gams W (eds) Soil-borne pathogens.
Academic Press, London New York
Bowen GD (1980) Mycorrhizal roles in tropical plants and ecosystems. In: Mikola P
(ed) Tropical mycorrhiza research. Oxford Univ Press, Oxford
Bowen GD, Foster RC (1978) Dynamics of microbial colonisation of plant roots. In:
Broughton WJ, John CK (eds) Soil microbiology and plant nutrition. Univ Malaya
Press, Malaya
Bowen GD, Rovira AD (1961 a) Plant growth in irradiated soil. Nature (London)
211:665-666
Bowen GD, Rovira AD (1961 b) Effects of microorganisms on plant growth. I. Develop-
ment of roots and root hairs in sand and agar. Plant Soil 15:166-188
Bowen GD, Rovira AD (1966) Microbial factor in short-term uptake studies with plant
roots. Nature 211 :655-666
Bowen GD, Rovira AD (1969) The influence of microof/~~anisms on growth and metabo-
lism of plant roots. In: Whittington WJ (ed) Root growth. Butterworths, London
Bowen GD, Rovira AD (1976) Microbial colonization of plant roots. Annu Rev Phyto-
pathol14:121-144
Bowen GD, Smith SE (1981) The effects of mycorrhizas on nitrogen uptake by plants.
In: Clark FE, Rosswall T (eds) Nitrogen cycling in terrestrial ecosystems. Ecol Bull
(Stockholm)
Bowen GD, Theodorou C (1973) Studies on phosphate uptake by mycorrhizas. Proc
14th IUFRO Congr, Munich
Bowen GD, Theodorou C (1973) Growth of ectomycorrhizal fungi around seeds and
roots. In: Marks GC, Kozlowski TT (eds) Ectomycorrhizae: their ecology and physi-
ology. Academic Press, London New York
Bowen GD, Theodorou C (1979) Interactions between bacteria and ectomycorrhizal
fungi. Soil BioI Biochem 11: 119-126
Bowen GD, Skinner MF, Bevege DC (1974) Zinc uptake by ecto- and endo-mycorrhizas
and uninfected roots of conifers. Soil BioI Biochem 6: 141-144
Bowen GD, Bevege DI, Mosse B (1975) Phosphate physiology of vesicular-arbuscular
mycorrhizas. In: Sanders FE, Mosse B, Tinker PB (eds) Endomycorrhizas. Academic
Press, London New York
Breisch H (1974) Contribution a l'etude du role des exudats racinaires dans les processes
d'aggregation des sols. These Doct Spec Univ Nancy
Brown ME (1972) Plant growth substances produced by microorganisms of soil and
rhizosphere. J Appl Bact 35:443-451
Brown ME (1974) Seed and root bacterization. Annu Rev Phytopathol12: 181-197
Brown ME (1976) Microbial manipulation and plant performance. In: Skinner FA, Carr
JG (eds) Microbiology in agriculture fisheries and food. Soc Appl Bacteriol Symp
4. Academic Press, London New York
Brown ME, Jackson RM, Burlington AK (1968) Effects on tomato plants, Lycopersicon
esculentum, by seed or root treatment with gibberellic acid and indolyl-3-acetic acid.
J Exp Bot 19:544-552
Cagle GD, Pfister RM, Vela GR (1972) Improved staining of extracellular polymers
for electron microscopy. Appl Microbio124:477-487
Callow JA, Capaccio LCM, Parish G, Tinker PB (1978) Detection and estimation of
polyphosphate in vesicular-arbuscular mycorrhizas. New Phytol 80: 125-134
Campbell R, Rovira AD (1973) A study of the rhizosphere by scanning electron microsco-
py. Soil BioI Biochem 5:747-752
88 A.D. ROVIRA et al.:

Chan ECS, Katznelson H (1961) Growth interactions of Arthrobacter globiformis and

Pseudomonas sp. in relation to the rhizosphere effect. Can 1 Microbiol 7: 759-767
Chilvers GA, Pryor LD (1965) The structure of eucalypt mycorrhizas. Aust 1 Bot
13:245-259
Cooper KM, Tinker PB (1978) Transfer and translocation of nutrients in vesicular-
arbuscular mycorrhizas II. Uptake and translocation of phosphorus, zinc and sulphur.
New Phytol 81 :43-52
Cox G, Tinker PB (1976) Translocation and transfer of nutrients in vesicular arbuscular
mycorrhizas. I. The arbuscule and phosphorus transfer: a quantitative ultrastructure
study. New Phytol 77:371-378
Cox G, Sanders FE, Tinker PB, Wild lA (1975) Ultrastructural evidence relating to
host-endophyte transfer in a vesicular-arbuscular mycorrhiza. In: Sanders FE, Mosse
B, Tinker PB (eds) Endomycorrhizas. Academic Press, London New York
Cress WA, Throneberry AO, Londsey DL (1979) Kinetics of phosphorus absorption
by mycorrhizal and nonmycorrhizal tomato roots. Plant Physiol 64: 484-487
Crush lR (1974) Plant growth responses to vesicular-arbuscular mycorrhizas. VII.
Growth and nodulation of some herbage legumes. New Phytol 73:743-749
Crush lR (1976) Endomycorrhizas and legume growth in some soils of the Mackenzie
Basin, Canterbury, New Zealand. NZ 1 Agric Res 19:473-476
Darbyshire IF, Greaves MP (1970) An improved method for the study of the interrela-
tionships of soil microorganisms and plant roots. Soil BioI Biochem 2: 63-71
Dart PI (1971) Scanning electron microscopy 01 plant roots. 1 Exp Bot 22: 163-168
Dart PI, Day 1M (1975) Non-symbiotic nitrogen fixation in soil. In: Walker N (ed)
Soil microbiology. Butterworth, London
Dart PI, Mercer FV (1964) The legume rhizosphere. Arch MikrobioI47:344-378
Dawes Cl, Bowler E (1959) Light and electron microscope studies of the cell wall structure
of the root hairs of Raphanus sativus. Am 1 Bot 46:561-565
D6bereiner 1 (1974) Nitrogen-fixing bacteria in the rhizosphere. In: Quispel A (ed) The
biology of nitrogen fixation. Elsevier/North-Holland, Biomedical Press, Amsterdam
New York
D6bereiner 1, Day 1M (1975) Associative symbiosis in tropical grasses: characterization
of microorganisms and dinitrogen fixation sites. In: Proc Int Symp N 2 fixation. Inter-
disciplinary discussions. Pullman, Washington State Univ
D6bereiner 1, Day 1M, Dart PI (1972) Nitrogenase activity and oxygen sensitivity of
the Paspalum notatum- Azotobacter paspali association. 1 Gen Microbiol 71: 103-116
Duff RB, Webley DM, Scott RO (1963) Solubilization of minerals and related materials
by 2-Betagluconic acid-producing bacteria. Soil Sci 95: 105-114
Elliott LF, Gilmour CM, Cochran VL, Coby C, Bennett D (1979) Influence of tillage
and residues on wheat root microflora and root colonization by nitrogen-fixing bacte-
ria. In: Harley lL, Scott-Russell R (eds) The soil-root interface. Academic Press,
London New York
Foster RC (1962) Cell wall structure and growth. PhD thesis, Univ Leeds, England
Foster RC (1978) Ultramicromorphology of some South Australian Soils. In: Emerson
WW, Bond RD, Dexter AR (eds). Modifications of soil structure. Wiley and Sons,
New York
Foster RC, Marks GC (1967) Observations on the mycorrhizae of forest trees. II. The
rhizosphere of Pinus radiata D. Don. Aust 1 BioI Sci 20:915-926
Foster RC, Marks GC (1973) Structure, morphogenesis and ultrastructure of ectomycor-
rhizae. In: Marks GC, Kozlowski TT (eds) Ectomycorrhizae. Academic Press, London
New York
Foster RC, Rovira AD (1973) The rhizosphere of wheat roots studied by electron micro-
scopy. Bull Ecol Res Commun (Stockholm) 17:93-95
Foster RC, Rovira AD (1976) Ultrastructure of wheat rhizosphere. New Phytol
76:343-352
Foster RC, Rovira AD (1978) The ultrastructure of the rhizosphere of Trifolium subter-
raneum L. In: Loutit MW, Miles lAR (eds) Microbial ecology. Springer, Berlin Hei-
delberg New York
I.2 The Significance of Rhizosphere Microflora and Mycorrhizas 89

Gadgil RL, Gadgil PD (1975) Suppression of litter decomposition by mycorrhizal roots

of Pinus radiata. NZ J For Sci 5:33-41
Gerdemann JW (1975) Vesicular-arbuscular mycorrhizae. In: Torrey JG, Clarkson DT
(eds) The development and function of roots. Academic Press, London New York
Gerretson FC (1948) The influence of microorganisms on the phosphate intake by the
plant. Plant Soil 1: 207-230
Gianinazzi S, Gianinazzi-Pearson Y, Dexheimer J (1979) Enzymatic studies on the metab-
olism of vesicular-arbuscular mycorrhiza. III. Ultrastructural localization of acid and
alkaline phosphatase in onion roots infected by Glomus mosseae (Nicol. & Gerd).
New Phytol 82:127-132
Gianinazzi-Pearson A, Gianinazzi S (1978) Enzymatic studies on the metabolism ofvesi-
cular-arbuscular mycorrhiza. II. Soluble alkaline phosphatase specific to mycorrhizal
infection in onion roots. Physiol Plant PathoI12:45-53
Gilmore AE (1971) The influence of endotrophic mycorrhizae on the growth of peach
seedlings. J Am Soc Hort Sci 96: 35-38
Gray TRG (1967) Stereoscan electron microscopy of soil microorganisms. Science
155:1668-1670
Greaves MP, Darbyshire JF (1972) The ultrastructure of the mucilaginous layer of plant
roots. Soil BioI Biochem 4: 443-449
Hale MG, Moore LD, Griffin GJ (1979) Root exudates and exudation. In: Dommergues
YR, Krupa SV (eds) Interactions between non-pathogenic soil micro-organisms and
plants. Elsevier, North Holland Biomedical Press, Amsterdam New York
Harley JL (1969) The biology of mycorrhiza. Hill, London
Hatch AB (1937) The physical basis of mycotrophy in the genus Pinus. Black Rock
For Bull 6: 1-168
Hattingh MJ, Gray LE, Gerdemann JW (1973) Uptake and translocation of 32P-Iabelled
phosphate to onion roots by mycorrhizal fungi. Soil Sci 116: 383-387
Hayman DS (1974) Plant growth responses to vesicular-arbuscular mycorrhiza. VI Effect
oflight and temperature. New Phytol 73: 71-80
Hayman DS (1975) Phosphorus cycling by soil microorganisms and plant roots. In:
Walker N (ed) Soil microbiology. Butterworth, London
Hayman DS, Mosse B (1972) Plant growth responses to vesicular-arbuscular mycorrhizas.
IV. Increased intake of labile P from soil. New Phytol 71 :41-47
Hiltner L (1904) Uber neuere Erfahrungen und Probleme auf dem Gebiet der Boden-
mikrobiologie und unter besonderer Beriicksichtigung der Griindiingung und Brache.
Arb Dtsch Landwirtsch Ber 98: 59-78
Jenny H, Grossenbacher KA (1963) Root soil boundary zone as seen in the electron
microscope. Proc Soil Soc Am 27: 273-277
Jones PCT, Mollison JE (1948) A technique for the quantitative estimation of soil micro-
organisms. J Gen Microbiol 2: 54-69
Katznelson H, Rouatt JW (1957) Manometric studies with rhizosphere and non-rhizo-
sphere soil. Can J Microbiol J: 673-678
Katznelson H, Lochhead AG, Timonin MJ (1948) Soil microorganisms and the rhizo-
sphere. Bot Rev 14:543-587
Katznelson H, Rouatt JW, Payne TMB (1954) Liberation of amino acids by plant roots
in relation to desiccation. Nature 174: 1110-1111
Katznelson H, Rouatt JW, Payne TMB (1955) The liberation of amino acids and reducing
compounds by plant roots. Plant Soil 7: 35-48
Kinden DA, Brown MF (1975) Electron microscopy ofvesicular-arbuscular mycorrhizas
of yellow poplar. II. Intercellular hyphae and vesicles. Can J Microbiol21 : 1768-1780
Kleinschmidt GD, Gerdemann JW (1972) Stunting of citrus seedlings in fumigated
nursery soils related to the absence of ectomycorrhizas. Phytopathology 62: 1447-1453
Knowles R (1977) The significance of asymbiotic dinitrogen fixation in bacteria. In:
Hardy RWF, Gibson AH (eds) A treatise on dinitrogen fixation, sect IV: Agronomy
and ecology. Wiley and Sons, New York
Krupa S, Fries N (1971) Studies on the ectomycorrhizae of pine. I. Production of volatile
organic compounds. Can J Bot 49: 1425-1431
90 A.D. ROVIRA et al.:

Lamb RJ, Richards BN (1971) Effect of mycorrhizal fungi on the growth and nutrient
status of slash and radiata pine seedlings. Aust For 35: 1-7
Lamont BB (1972) "Proteoid" roots in the legume Viminariajuncea. Search 3:91-92
Lamont BB, McComb AJ (1974) Soil microorganisms and the formation of proteoid
roots. Aust J Bot 22:681-688
Leech JH, Mollenhauer HH, Whaley WG (1963) Ultrastructural changes in the root
apex. Symp Soc Exp BioI 17: 74-84
Leiser AT (1968) A mucilaginous root sheath in Ericaceae. Am J Bot 55:392-398
Leppard GG (1974) Rhizoplane fibrils in wheat: demonstration and derivation. Science
185:1066-1067
Leppard GG, Ramamoorthy S (1975) The aggregation of wheat rhizoplane fibrils and
the accumulation of soil-bound cations. Can J Bot 53: 1729-1735
Lewis DA, Harley JL (1965) Carbohydrate physiology of mycorrhizal roots of beech.
III. Movement of sugars between host and fungus. New Phytol 64:256-259
Ling-Lee M, Chilvers GA, Ashford AE (1975) Polyphosphate in three different kinds
of tree mycorrhiza. New Phytol 75: 551-554
Lochhead AG (1940) Qualitative studies of soil microorganisms. III. Influence of plant
growth on the character of the bacterial flora. Can J Res Sec C 18: 42- 58
Lochhead AG, Chase FE (1943) Qualitative studies of soil microorganisms. V. Nutritional
requirements of the predominant bacterial flora. Soil Sci 55: 185-195
Loutit MW (1968) Soil microorganisms and molybdenum concentration in plants. Trans
9th Int Congr Soil Sci 3: 491-498
Loutit MW, Hillas J, Spears GFS (1972) Studies on rhizosphere microorganisms and
molybdenum concentration in plants. II Comparison of isolates from the rhizosphere
of plants grown in two soils under the same conditions. Soil BioI Biochem 4: 267-270
Malajczuk N, Bowen GD (1974) Proteoid roots are microbially induced. Nature
251:316-317
Martin JK (1975) 14C-Iabelled material leached from the rhizosphere of plants supplied
continuously with 14C02. Soil BioI Biochem 7: 395-399
Martin JK (1977) Factors influencing the loss of organic carbon from wheat roots.
Soil BioI Biochem 9: 1-9
Melin E, Nilsson H (1950) Transfer of radioactive phosphorus to pine seedlings by
means of mycorrhizal hyphae. Physiol Plant 3: 88-92
Melin E, Nilsson H (1952) Transport of labelled nitrogen from an ammonium source
to pine seedlings through mycorrhizal mycelium. Svensk Bot Tidskr 46:281-285
Melin E, Nilsson H (1955) Ca45 used as an indicator of transport of cations to pine
seedlings by means of mycorrhizal mycelia. Svensk Bot Tidskr 49: 119-122
Menge JA, Labanauskas CK, Johnson ELV, Platt RG (1978) Partial substitution of
mycorrhizal fungi for phosphorus fertilization in the greenhouse culture of citrus.
Soil Sci Soc Am 42: 926-930
Mishustin EN, Naumova AN (1962) Bacterial fertilizers: their effectiveness and mode
of action. Mikrobiologiya 31: 543-555
Moawad M (1979) Ecophysiology ofv.a. mycorrhiza in the tropics. In: Harley JL, Russell
RS (eds) The soil-root interface. Academic Press, London New York
Mollenhauer HH, Morre DJ (1966) Golgi apparatus and plant secretion. Annu Rev
Plant Physiol 17: 27-46
Morrison TM (1963) Uptake of sulphur by excised beech mycorrhizas. New Phytol
62:44-49
Mosse B (1972) The influence of soil type and Endogone strain on the growth ofmycorrhi-
zal plants in phosphate deficient soils. Rev Ecol BioI Sol 9: 529-537
Mosse B (1973) Plant growth responses to vesicular-arbuscular mycorrhiza. IV. In soil
given additional phosphate. New Phytol 72: 127-136
Mosse B (1977) Plant growth responses to vesicular-arbuscular mycorrhiza. X. Responses
of stylosanthes and maize to inoculation in unsterile soils. New Phytol 78: 277-288
Mosse B, Powell CW, Hayman DW (1976) Plant growth responses to vesicular-arbuscular
mycorrhiza. IX. Interactions between v.a. mycorrhiza, rock phosphate and symbiotic
nitrogen fixation. New Phytol 76:331-342
1.2 The Significance of Rhizosphere Microflora and Mycorrhizas 91

Nambiar EKS (1976) The uptake of zinc-65 by oats in relation to soil water content
and root growth. Aust J Soil Res 14:67-74
Newman EI, Bowen HJ (1974) Patterns of distribution of bacteria on root surfaces.
Soil BioI Biochem 6: 205-209
Newman EI, Watson A (1977) Microbial abundance in the rhizosphere. A computer
model. Plant Soil 48: 17-56
Norkrans B (1950) Studies in growth and cellulolytic enzymes of Tricholoma with special
reference to mycorrhiza formation. Symb Bot Upsal11(1): 1-126
Northcote DH, Pickett-Heaps JD (1966) A function of the Golgi apparatus in polysaccha-
ride synthesis and transport in the root cap cells of wheat. Biochem J 98: 159-167
Nye PH, Tinker PB (1977) Solute movement in the soil-root system. Blackwell, Oxford
Old KM, Nicholson TH (1975) Electron microscopical studies of the microflora of roots
of sand dune grasses. New Phytol 74: 51-58
Parker CA (1957) Non-symbiotic nitrogen-fixing bacteria in soil. III. Total nitrogen
changes in a field soil. J Soil Sci 8: 48-59
Pearson V, Read DJ (1973) The biology ofmycorrhiza in the Ericaceae. II. The transport
of carbon and phosphorus by the endophyte and the mycorrhiza. New Phytol
72:1325-1331
Preston RD (1974) Physical biology of plant cell walls. Chapman and Hall, London
Purnell HM (1960) Studies of the family Proteaceae. I. Anatomy and morphology of
the roots of some Victorian species. Aust J Bot 8: 38-50
Ratnayake M, Leonard RT, Menge JA (1978) Root exudation in relation to the supply
of phosphorus and its possible relevance to mycorrhizal formation. New Phytol
81 :543-552
Raven JA, Smith FA (1976) Nitrogen assimilation and transport in vascular land plants
in relation to intracellular pH regulation. New Phytol 76:415-431
Raven JA, Smith SE, Smith FA (1978) Ammonium assimilation and the role ofmycorrhi-
zas in climax communities in Scotland. Trans Bot Soc Edinburgh 43: 27-35
Read DJ (1974) Pezizella ericae sp. nov. the perfect state of a typical mycorrhizal endo-
phyte of Ericaceae. Trans Br Mycol Soc 63: 381-383
Read DJ, Stribley DP (1973) Effect of mycorrhizal infection on nitrogen and phosphorus
nutrition of Ericaceous plants. Nature 244: 81-82
Reuszer HW (1949) A method for determining the carbon dioxide production of sterile
and non-sterile roots. Soil Sci Soc Am Proc 14: 175-179
Reuszer HW (1962) Axenic culture in the determination of root functions and the interre-
lationships of microorganisms. Soil Sci 93: 56-61
Rhodes.LH, Gerdemann JW (1975) Phosphate uptake zones of mycorrhizal and non-
mycorrhizal onions. New Phytol 75:555-561
Rice EL (1974) Allelopathy. Academic Press, London New York
Ridge EH, Rovira AD. (1968) Microbial inoculation of wheat. Trans 9th Int Congr
Soil Sci 3:473-481
Rovira AD (1956) A study of the developments of the root surface microflora during
the initial stages of plant growth. J Appl Bacteriol 19: 72-79
Rovira AD (1959) Root excretions in relation to the rhizosphere effect. IV. The influence
of plant species, age of plant, light, temperature and calcium nutrition on exudation.
Plant Soil 11 : 53-64
Rovira AD (1965a) Plant root exudates and their influence upon soil microorganisms.
In: Baker KE, Snyder WC (eds) Ecology of soil-borne plant pathogens. Berkeley
Univ
Rovira AD (1965b) Effects of Azotobacter, Bacillus and Clostridium on the growth of
wheat. In: Macura J, Vancura V (eds) Plant microbe relationships. Czech Acad Sci,
Prague
Rovira AD (1969) Plant root exudates. Bot Rev 35: 35-57
Rovira AD, Bowen GD (1969) Phosphate incorporation by sterile and non-sterile roots.
Aust J BioI Sci 19:1167-1169
Rovira AD, Campbell R (1974) Scanning electron microscopy of microorganisms on
the roots of wheat. Microb Ecol 1: 15-23
92 A.D. ROVIRA et al. :

Rovira AD, Davey CB (1974) Biology of the rhizosphere. In: Carson EW (ed) The
plant root and its environment. Univ Press, Charlottesville, Virg
Rovira AD, Newman EI, Bowen HJ, Campbell R (1974) Quantitative assessment of
the rhizoplane microflora by direct microscopy. Soil BioI Biochem 6:211-216
Rovira AD, Foster RC, Martin JK (1979) Note on terminology: Origin, nature and
nomenclature of the organic materials in the rhizosphere. In: Harley JL, Scott-Russell
R (eds) The soil-root interface. Academic Press, London New York
Samtsevich SA (1965) Active excretions of plant roots and their significance. Sov Plant
PhysioI12:731-740
Sanders FE, Tinker PB (1973) Phosphate flow into mycorrhizal roots. Pestic Sci
4:385-395
Sanders FE, Mosse B, Tinker PB (1975) Endomycorrhizas. Academic Press, London
New York
Sanders FE, Tinker PB, Black RLB, Palmerley SM (1977) The development of endomy-
corrhizal root systems. I. Spread of infection and growth-promoting effects with four
species of vesicular-arbuscular endophyte. New Phytol 78: 257-268
Schank SC, Smith RL, Weiser GC, Bouton JH, Quesenberry KH, Tyler ME, Milam
JR, Littell RC (1979) Fluorescent antibody technique to identify Azospirillum brasi-
lense associated with roots of grasses. Soil BioI Biochem 11 :287-296
Schenk NC, Kinloch RA, Dickson DW (1975) Interaction of endomycorrhizal fungi
and root-knot nematode on soybean. In: Sanders FE, Mosse B, Tinker PB (eds)
Endomycorrhizas. Academic Press, London New York
Schippers B, Vuurde van JWL (1978) Studies of microbial colonization of wheat roots
and the manipulation of the rhizosphere microflora. In: Loutit MW, Miles JAR
(eds) Microbial ecology. Springer, Berlin Heidelberg New York
SchOnbeck F (1979) Endomycorrhiza in relation to plant diseases. In: Schippers B, Gams
W (eds) Soil-borne plant pathogens. Academic Press, London New York
Schoknecht JD, Hattingh MT (1976) X-ray microanalysis of elements in cells of v.a.
mycorrhizal and non-mycorrhizal onions. Mycologia 68: 296- 304
Schwartz F (1883) Die Wurzelhaare der Pflanzen. Untersuch Bot Inst Tiibingen
1:135-188 .
Scott FM, Hamner KC, Baker E, Bowler F (1958) Electron microscope studies of the
epidermis of Allium cepa. Am J Bot 45:449--461
Setterfield G, Bayley ST (1957) Studies on the mechanism of deposition and extension
of primary cell walls. Can J Bot 35 :435-444
Shamoot S, McDonald L, Bartholemew MV (1960) Rhizodeposition of organic debris
in soil. Soil Sci Am Proc 32: 817-820
Skinner MF, Bowen GD (1974a) The uptake and translocation of phosphate by mycelial
strands of pine mycorrhizas. Soil BioI Biochem 6: 53-56
Skinner MF, Bowen GD (1974b) The penetration of soil by mycelial strands of pine
mycorrhizas. Soil BioI Biochem 6: 57-61
Smiley RW (1974) Rhizosphere pH as influenced by plants, soils, and nitrogen fertilizers.
Proc Soil Sci Soc Am 38: 795-799
Smith SE, Nicholas DJD, Smith FA (1979) Effect of early mycorrhizal infection on
nodulation and nitrogen fixation in Trifolium subterraneum L. Aust J Plant Physiol
6:305-316
Sperber 11 (1958) The incidence of apatite-solubilizing organisms in the rhizosphere and
soil. Aust J Agric Res 9: 778-781
Starkey RL (1939) Some influence of the development of higher plants upon the microor-
ganisms in the soil. VI. Microscopic examination of the rhizosphere. Soil Sci
45:207-209
Stribley DP, Read DJ (1974) The biology ofmycorrhiza in the Ericaceae. IV. The effect
of mycorrhizal infection on the uptake of 15N from labelled soil by Vaccinium macro-
carpon Ait. New Phytol 73:1149--1155
Stribley DP, Read DJ (1975) Some nutritional aspects of the biology of ericaceo us mycor-
rhizae. In: Sanders FE, Mosse B, Tinker PB (eds) Endomycorrhizae. Academic Press,
London New York
1.2 The Significance of Rhizosphere Microflora and Mycorrhizas 93

Stribley DP, Read DJ (1976) The biology ofmycorrhiza in the Ericaceae. VI. The effects
of mycorrhizal infection and concentration of ammonium nitrogen on growth of
cranberry (Vaccinium macrocarpon Ait.) in sand culture. New Phytol 77:63-72
Stribley DP, Reid DJ, Hunt R (1975) The biology of mycorrhiza in the Ericaceae V.
The effects of mycorrhizal infection, soil type and partial soil-sterilization (by gamma
irradiation) on the growth of cranberry (Vaccinium macrocarpon Ait.). New Phytol
75: 119-130
Sutton JC, Sheppard BR (1976) Aggregation of sand dune soil by endomycorrhizal fungi.
Can J Bot 54:326-333
Theodorou C, Bowen aD (1970) Mycorrhizal responses of radiata pine in glasshouse
and in field experiments with different fungi. Aust For 34: 183-191
Theron 11 (1966) The mineralization of nitrogen in soils under grass. S Afr J Agric
Sci 6: 155-164
Timonin MI (1946) Microflora of the rhizosphere in relation to the manganese-deficiency
disease in oats. Soil Sci Soc Am Proc 11 : 284-292
Tisdall JM, Oades JM (1979) Stabilization of soil aggregates by the root systems of
rye grass. Aust J Soil Res 17:429-441
Trinick MJ (1970) Rhizobium interactions with soil microorganisms. PhD thesis Univ
Washington WA
Trinick MJ (1977) Vesicular-arbuscular infection and soil phosphorus utilization in
Lupinus spp. New Phytol 78:297-304
Trolldenier a (1965) Fluoreszenmikroskopische Untersuchung von Mikroorganismenkul-
turen in der Rhizosphare. Z Bakteriol Parasitenkd Infekt Hyg III 119:256-259
Vancura V (1964) Root exudates of plants. I. Analysis of root exudates of barley and
wheat in their initial phases of growth. Plant Soil 21 :231-248
Vuurde van JWL, Elenbass PEM (1978) Use of fluorochromes for direct observation
of microorganisms associated with wheat roots. Can J Microbiol24: 1272-1275
Warembourg FR (1975) Le degagement de CO 2 dans la rhizosphere des plantes. Soc
Bot Fr Colloq Rhizosphere 122:77-87
Warembourg FR, Billes a (1979) Estimating carbon transfers in the rhizosphere. In:
Harley JL, Scott-Russell R (eds) The soil-root interface. Academic Press, London
New York
Webley DM, DuffRB, Bacon JSD, Farmer VC (1965) A study ofpolysaccharide-produc-
ing organisms occurring in the root region of certain pasture grasses. J Soil Sci
16:149-157
Welte E, Trolldenier a (1965) Effect of soil microflora and seed microflora on the growth
of plants. In: Macura J, Vancura V (eds) Plant microbe relationships. Czech Acad
Sci, Prague
Woldendorp JW (1963) The influence of living plants on denitrification. Meded Landb-
hoogesch Wageningen 63(13):1-100
1.3 Modern Solution Culture Techniques
c.J. ASHER and D.G. EDWARDS

It has been known since the very early days of plant physiology that land
plants as well as water plants could be grown successfully without soil in aqueous
culture media. Although the famous experiments of WOODWARD (1699) repre-
sent the earliest recorded use of solution culture! techniques to study the growth
ofland plants, the experiments of DE SAUSSURE (1804) are probably more signifi-
cant in that they involved the first use of nutrient solutions of known initial
composition, prepared by dissolving various salts in distilled water. Subse-
quently SACHS (1860) and KNOP (1865) showed that plants could be grown
to maturity in simple solution culture systems not very different from those
currently in use in many plant research laboratories throughout the world.
The application of solution culture techniques has contributed much to current
knowledge within the broad field of plant physiology, especially in the areas
of mineral nutrition and water relations. However, with any set of techniques
which have been in use for a long period of time there is the risk that excessive
familiarity will cause research workers to lose sight of some inherent limitations
of the techniques and use them in ways that are inappropriate.
In this chapter, emphasis will be placed on the uses and limitations of various
types of solution culture techniques rather than on detailed descriptions of
particular techniques. For the latter, readers are referred particularly to the
excellent and comprehensive review by HEWITT (1966).

1 Major Differences Between Solution Culture and Soil Culture

Solution culture techniques are often used to simulate, under at least partially
controlled conditions, situations which occur when plants are growing in the
field. However, if solution culture is to be used as a model experimental system
for soil-grown plants it is important that the differences between the two culture
systems are kept firmly in mind.

1.1 Mechanical Support

Soil-grown plants receive their major mechanical support from the root system
which ramifies through the soil matrix. This occurs also in sand culture and
Throughout the present chapter the general term solution culture has been taken to
include a wide range of related plant culture techniques used for experimental purposes.
These include conventional (non-flowing) water cultures, sand and gravel culture, mist
culture and various types of flowing solution culture systems. The term hydroponics
is used to denote crop production by a solution culture method
I.3 Modern Solution Culture Techniques 95

gravel culture, but in water culture systems an alternative means of mechanical

support is required. Frequently the plants are supported by wrapping a short
section of the stem with cotton or dacron wool, or a piece of soft plastic foam,
and then wedging the stem into a hole in the lid of the culture vessel, or into
a cork or rubber stopper inserted in a still larger hole in the lid. Where substan-
tial radial expansion of the stem is expected during the course of an experiment,
the cork should have a radial split so that it can be removed easily and replaced
with a second cork with a larger hole, to prevent undue pressure on the base
of the stem. This system works quite well with single-stemmed plants such
as cotton or tomato, but is less convenient for plants which tiller from the
base, e.g. most cereals and grasses.
For species which branch close to the ground or form a rosette of leaves
at the soil surface, adequate support may be obtained by passing the tap root
or the base of the stem through a slit in a piece of soft plastic film glued
across a hole in the lid with contact adhesive (AsHER et al. 1965). Radial growth
of the root or stem base is accommodated by stretching or tearing of the film.
However, the method is unsuitable for supporting large plants or those which
tiller from the base.
Basket supports provide a highly versatile plant support system suitable
for species with widely differing growth habits (ASHER 1978). The use of these
was advocated by GERICKE (1929) for intensive hydroponic crop production,
but gravel culture systems proved to be more practical. For experimental pur-
poses it is convenient to construct the baskets from chemically inert, non-phyto-
toxic plastics such as nylon, polyethylene, or polypropylene which are easily
cleaned, a point of considerable importance in micronutrient studies. Usually
the sides of the basket are solid, and the base is formed from plastic mesh
through which the roots grow to reach the nutrient solution. Use of mesh
in which the fibers are not bonded at points of intersection allows lateral dis-
placement of the fibers to accommodate radial expansion of the roots. Usually
the baskets are filled with a mulch of black polyethylene beads to exclude
light from the nutrient solution and to provide additional support for the base
of the stem. Spray from the bursting of bubbles at the surface of adequately
aerated nutrient solutions keeps the lower 1 to 1.5 cm of the bead layer wet
enough for seed germination and for the development of adventitious roots.
The system is eminently suited for species which tiller or develop rhizomes.
Baskets as small as 2 cm diameter (Fig. 1) and as large as 25 cm diameter have
been found convenient in plant nutrient studies with various species (ASHER
1978). Two examples of basket supports are shown in Fig. 1.

1.2 Spatial Variation in Root Environment Parameters

Soil is an extremely complex and heterogeneous medium often varying substan-

tially in physical, chemical or biological characteristics over comparatively small
horizontal or vertical distances. A consequence of this variation is that different
parts of a plant root system may be located in markedly different environments.
In addition, the ionic environment experienced by any particular portion of a
plant root system in the soil depends on a balance between rates of ion move-
96 C.l. ASHER and D.G. EDWARDS:

Fig. 1 a, b. Basket supports for solution culture experiments. a 10 cm inside diameter

basket being planted with sprouted wheat seeds for a flowing solution culture experiment
(cf. Sect. 2.3). Mulch of black polyethylene beads in previously planted pots visible
at left and in background. b 2 cm diameter basket used to support a maize seedling

ment to the root surface by mass flow and diffusion, and rates of absorption
by that portion of root (BARBER 1962). Hence differential uptake of water and
nutrient ions by different parts of the root system adds further to the heterogene-
ity of the environment to which the root system of a soil-grown plant is exposed.
Because of the complexity and heterogeneity of the soil it is extremely diffi-
cult to characterize the root environment of soil-grown plants precisely or to
exercise adequate control over root environment parameters, the effects of which
we may wish to study. However in well-stirred solution culture systems it is
possible to provide a very nearly homogeneous root environment and to measure
with acceptable accuracy such root environment parameters as pH, p02, Eb
(oxidation-reduction potential), nutrient ion concentrations, and temperature.
Within wide limits these parameters can be varied independently of each other,
something which frequently cannot be achieved in soil. Solution culture thus
represents a preferred system for studying effects of root environment para-
meters on the physiology and growth of plants.
However, it must always be kept in mind that solution culture systems are
much simpler than soil systems, and as a consequence, responses of soil-grown
plants to various root environment parameters may be modified by factors
not present in solution culture systems. Such factors include large numbers
of soluble mineral elements not known to be essential for plant growth and
I.3 Modern Solution Culture Techniques 97

therefore omitted from nutrient solutions, the wide variety of soluble organic
compounds commonly occurring in the soil solutions, and the varied popula-
tions of microorganisms associated with the roots of soil-grown plants (see
Sect. 1.4).

1.3 Temporal Variation in Root Environment Parameters

1.3.1 Nutrient Depletion

In soils, adsorption/desorption reactions, the slow dissolution of primary miner-
als, and the decomposition of soil organic matter serve to replenish the soil
solution as mineral nutrients are withdrawn by the plant roots. Although both
local depletion (BARBER 1962) and general depletion of the soil solution (BURD
and MARTIN 1924) may occur over a period of time, in the majority of soils
the solution in contact with the roots is buffered to a substantial degree against
changes in mineral composition. Continued growth of roots into "new" soil
further tends to stabilize the average ionic environment of the whole root system.
In simple water culture systems, the composition of the nutrient solution
is essentially unbuffered and large changes in solution composition can occur
within a relatively short space of time (Table 1). The rate of depletion depends
on a number of factors including the rate of nutrient uptake by the plants,
the initial concentrations of ions in the nutrient solution, and the volume of
nutrient solution per plant. In sand culture the depletion problem tends to
be more severe than in water culture because of the absence of stirring and
the relatively small volumes of nutrient solution retained in the sand after drain-
age.
The need to supply adequate total amounts of mineral nutrients in conve-
niently small volumes of nutrient solution has led to the use of high initial
concentrations of many ions. In the case of phosphorus, the initial concentra-
tions frequently exceed those occurring in soil solutions by 3 to 4 orders of
magnitude. Thus Hoagland's solution contains 1,000 /lM phosphorus (HOAG-
LAND and ARNON 1950) and the Long Ashton solution, 1,330/lM (HEWITT

Table 1. Effects of initial potassium concentration and volume of nutrient solution per
plant on changes with time in the concentration of potassium in nutrient solutions sup-
porting the growth of young cassava plants. (Modified after SPEAR et al. 1978)

Time from Volume of solution (l plant - 1)

planting
(days) 1 2.5 8 1 2.5 8 2.5 8

Potassium concentrations in solution (11M)"

0 20 20 20 300 300 300 1875 1875 1875
4 2 2 1 3 128 218 1065 1510 1590
10 <1 <1 <1 <1 1 95 130 950 1440
18 <1 <1 <1 <1 <1 1 <1 37 740
27 <1 <1 <1 <1 <1 <1 <1 9 280

a Limit of detection (by atomic absorption spectroscopy) was 1 11M K

98 C.J. ASHER and D.G. EDWARDS:

1966). However, the modal soil solution phosphorus concentration in 149 tem-
perate soils was found to be less than 1 11M (REISENAUER 1966), and in six
tropical soils GILLMAN and BELL (1978) found values below 0.2 11M.

1.3.2 pH Shifts
Unequal absorption of anions and cations from the root environment can lead
to substantial changes in pH. Because of the large amounts in which nitrogen
is required compared with the other mineral nutrients, the form in which nitro-
gen is supplied tends to exert a major influence on the direction of pH change.
Thus absorption of nitrogen in anionic form (N03) generally leads to an in-
crease in the pH of the root environment, whereas absorption of cationic nitro-
gen (NHt) leads to a decrease in pH. TRELEASE and TRELEASE (1935) showed
in water culture experiments with wheat that by varying the N0 3/NHt ratio
they could cause the pH to increase, decrease, or remain approximately constant.
However, species differences in nutrient absorption and assimilation may be
important, since it has been consistently observed with non-nodulated jackbeans
(Canavalia ensi[ormis) that the pH of the nutrient solution decreases markedly
with time even when all the nitrogen is supplied in the N0 3 form (ASHER
unpublished). With concentrated nutrient solutions some phosphate buffering
of pH can be expected, particularly in near-neutral solutions (pKa = 7.21). How-
ever, in general, pH shifts in nutrient solutions are likely to be greater than
in soils because of the lack of an exchange complex to adsorb or desorb hydrogen
IOns.

1.4 Root-Microorganism Interactions

Roots of soil-grown plants exude substantial quantities of organic compounds

into the soil. According to BARBER and MARTIN (1976) as much as 12%-18%
of the total carbon fixed from the atmosphere by cereal plants may be lost
in this manner. Accumulation of these organic substances in the soil adjacent
to roots provides a rich substrate for the growth of a wide range of microorgan-
isms. (For a review on factors affecting the colonization of root surfaces by
microorganisms see BOWEN and ROVIRA 1976). In many cases the precise effects
of these microorganisms on the growth and function of roots has not been
elucidated and much remains to be learned in this area.
In solution culture roots appear to produce much smaller quantities of
exudate than in soil (BOWEN and ROVIRA 1976) and the water-soluble compo-
nents of the exudate will be dispersed through the bulk of a well-stirred nutrient
solution. Hence the quantities of organic substrates available to support micro-
bial growth on root surfaces of plants grown in solution culture are likely
to be much less than for roots of soil-grown plants. However, despite this
difference, it has proved possible to study some important root-microorganism
interactions in solution culture and much has been learned from these experi-
ments (see Sect. 2.1.7).
1.3 Modern Solution Culture Techniques 99

2 Uses and Limitations of Existing Solution Culture Methods

2.1 Non-Renewed or Intermittently Renewed Water Cultures and Sand Cultures

These are the simplest and most widely used solution culture methods in plant
research laboratories today. Large variations exist among research workers in
the initial composition of the nutrient solutions employed, the volume of solu-
tion provided per plant, and the frequency of solution renewal, if any (see
HEWITT 1966). Frequently small volumes (~ 11 plane 1) of rather concentrated
nutrient solution (~20,000 11M total nutrients) are employed for reasons of
experimental convenience. In some cases the initial concentrations are so high
that complete dissolution of the salts is not possible or precipitation subse-
quently occurs (e.g. BOND 1951; NORRIS and DATE 1976). In other cases thresh-
olds for toxicity in sensitive species are exceeded (e.g. WILLIAMS and VLAMIS
1957) so that growth is restricted at least until nutrient depletion lowers the
concentration of the toxic ion to a safe level. Since the initial concentrations
commonly employed in nutrient solutions far exceed both the minimum concen-
trations necessary for healthy growth (ASHER 1978) and the concentrations com-
monly occurring in soil solutions, better results- might often be obtained in
solution culture experiments if the initial concentrations were reduced, and the
volume of solution or frequency of renewal increased correspondingly. The
composition of some well-known nutrient solutions with differing sources of
nitrogen are shown in Table 2.

2.1.1 Use in Teaching, Demonstration, and Diagnosis

Water culture and sand culture provide convenient means ofraising plant mate-
rial in a standard manner for a wide variety of teaching and demonstration
purposes. With water culture the study of root anatomy and morphology is
facilitated by easy access to the root system. However, in soil-grown plants
anatomical features may be modified by physical contact with the soil and
by exposure to rhizosphere organisms not necessarily present in the nutrient
solution. Hence care should be taken not to extrapolate too widely from observa-
tions made on the root systems of plants grown in water culture.
Symptoms of many nutritional disorders are readily induced by modifying
the initial composition of the nutrient solution. For the majority of temperate
species of economic importance, the visible symptoms of deficiency of at least
most of the essential elements, and symptoms of the most common mineral
toxicities, are well-known and adequately described in the literature. These de-
scriptions represent an extremely valuable aid to the diagnosis of nutritional
problems under both greenhouse and field conditions. However, for many tropi-
cal crop and pasture species, and for many native plants of both tropical and
temperate regions this is not the case. Hence very considerable scope still exists
for solution culture experiments aimed at the production and description of
nutritional disorders in these species.
Table 2. Some well-known nutrient solutions having either no nitrogen (for studies on nitrogen fixation) or the nitrogen supplied as nitrate, .....
0
0
ammonium, or nitrate plus ammonium. (Modified after ASHER 1978, all concentrations in IlM)

Nitrogen source None Nitrate Ammonium Nitrate plus ammonium

Reference BOND NORRIS HOAGLAND SIDERIS ADDOMS HOAGLAND MULDER JOHNSON

(1951)' and DATE and ARNON eta!. (1937)d and ARNON (1948) et a!.
(1976)" b (1950) (1943)C (1950) (1957)"

Nitrogen
NO; 15,000 14,000 3,460 14,000
NH: 2,000 11,110 1,000 2,270 2,000
Potassium 10,060 3,000 6,000 1,500 6,390 6,000 6,370 6,000
Calcium 5,320 2,000 5,000 1,000 2,920 4,000 1,453 4,000
Magnesium 2,030 1,000 2,000 1,000 1,385 2,000 1,014 1,000
Phosphorus 2,610 1,000 1,000 500 6,390 1,000 2,190 2,000
Sulfur 4,930 3,005 2,000 1,500 6,950 2,000 3,620 1,000
Chlorine 10,075 1,000 18 2,000 18 279 50
Iron 1,495 5.4 25 f 100 12 25 f 93 4f
Boron 46 12 46 46 16 46 4 25
Manganese 9 4.6 9 9 7 9 4.5 5
Zinc 0.8 0.4 0.8 1.5 0.8 0.9 2
Copper 0.3 0.16 0.3 1.6 0.3 1 0.5 (J
Molybdenum 0.1 0.03 0.1 0.1 0.4 0.1 :-.
Other elements 1,0.8 Na,9 >
Total elements 36,576 11,027 31,099 9,659 35,180 30,099 20,759 30,087

, Concentrations which would occur if all nutrient salts dissolved completely and no precipitation occurred
~
Pl
::l
b A relatively dilute nutrient solution now commonly used in place of the more concentrated solution of NORRIS (1964) which has been 0.
found to inhibit the growth and nodulation of some legume species. Note however that precipitation (?CaS0 4 ) still occurs in this solution t:J
(NORRIS and DATE 1976) P
CA relatively dilute nutrient solution used for experiments with pineapple tr1
d Solution "B" 0
" Often used in place of Hoagland-Arnon solution, frequently at 1/ 10th strength for young seedlings and 1/3 to 1/2 strength for older
plants. NH: : NO; of 1 : 7 maintains pH close to 5 til
~
f Renewed twice weekly
1.3 Modern Solution Culture Techniques 101

2.1.2 Production of Roots for Ion Transport Studies

Much of our existing knowledge concerning processes of ion uptake and long-
distance transport has come from studies of whole plants or excised roots raised
in simple water culture systems. Excised roots of low salt status are 9ften pro-
duced by substituting a dilute calcium sulphate solution (ca. 200 J..LM) for the
nutrient solution (see EpSTEIN 1972).

2.1.3 Nutrient Essentiality

The simplicity of water culture systems, and the availability of technology for
reducing to very low levels contaminating elements in water and nutrient salts
makes water culture a highly suitable technique for investigating the possible
roles of elements not at present regarded as essential for plant growth. Sand
culture is less suitable for this purpose, because of the need to purify the sand
as well as the water and nutrient salts.
Less than half the elements known to occur in plant tissues have been shown
to be essential for plant growth, and many of the remaining elements have
not been studied carefully for possible roles using modern plant culture and
purification techniques. Unfortunately this important area of plant nutrition
research has been relatively neglected during the past two decades. Thus only
two major discoveries have been made since 1959: the essential role of cobalt
for the growth of nodulated legumes (AHMED and EVANS 1960; HALLSWORTH
et al. 1960; REISENAUER 1960), and the essentiality of silicon for the growth
of the sedge Equisetum arvense (CHEN and LEWIN 1969) and for tomato (MIYAKE
and TAKAHASHI 1978). At present few laboratories appear to be directing signifi-
cant resources to the search for additional essential elements for plant growth.
However, following the demonstration by DIXON et al. (1975) that nickel is
required for the functioning of the enzyme urease, a number of groups have
recently commenced research on nickel as a possible essential element.
The present low level of research activity on nutrient essentiality for plants
contrasts markedly with the high level of activity among animal science groups.
Thus SCHWARZ (1974) drew attention to six discoveries of new trace elements
for animals during the 15-year period 1959 to 1974 (chromium, tin, vanadium,
fluorine, silicon, and nickel), and indicated that a further 14 elements were
under intensive investigation at that time.

2.1.4 Effects of Root Environment Parameters

In general, simple water culture and sand culture systems are not very suitable
for the study of effects of root environment parameters because of difficulty
in maintaining constant pH and nutrient ion concentrations (see Sects. 1.3.1,
1.3.2 and 2.1.4.1). The preferred method for such studies is by means of contin-
uously flowing solution cultures (see Sect. 2.3). However, since the majority
of laboratories are not yet equipped with flowing culture equipment, it is impor-
tant to examine critically the limitations of conventional solution culture systems
and to determine the purposes for which they may legitimately be used, and
those for which their use is inappropriate.
102 C.J. ASHER and D.G. EDWARDS:

2.1.4.1 Temperature Studies

Active ion uptake processes are markedly temperature-sensitive, but relatively
little attention has been paid to effects of root temperature on growth and
nutrient uptake by intact plants. Satisfactory control of root temperature can
be obtained in water culture experiments by standing the pots in large constant
temperature waterbaths. At the authors' laboratory four such waterbaths mea-
suring approximately 2 m xl m x 20 cm deep have been in use for some years.
The baths are constructed from stainless steel and insulated on the sides and
bottom with polyurethane foam. The heat load from above is reduced by cover-
ing the baths with sheets of l.5-cm-thick waterproof plywood carrying a pattern
of holes of a suitable size to fit snugly around the tops of the pots. Aeration
of the nutrient solutions provides sufficient stirring to prevent the development
of appreciable temperature gradients in the nutrient solution. Using this equip-
ment FORNO et al. (1979) showed that when five cassava cultivars were grown
in full strength Hoagland's solution (HOAGLAND and ARNON 1950) all developed
severe boron deficiency at a root temperature of 18° C, but absorbed sufficient
boron for healthy growth at 28° C. Further experiments on effects of root tem-
perature on mineral nutrition are needed.

2.1.4.2 pH Studies
ARNON and JOHNSON (1942) successfully used water culture techniques to study
effects of pH on tomato, lettuce and Bermuda grass. However, precise pH
control was difficult to achieve even with a large volume of solution (10-201
plant -1), continuous slow addition of acid or alkali to the pots according to
treatment, and twice-daily checks of pH on an individual pot basis. In some
treatments deviations from the desired pH values exceeded 0.2 pH units, despite
the precautions taken.
Where it is necessary only to maintain a favourable pH for plant growth,
addition of solid CaC0 3 to the nutrient solution may be effective. Thus in
experiments on the role of cobalt in symbiotic nitrogen fixation in lucerne,
WILSON and REISENAUER (1963) were able to maintain a pH of between 6.5
and 7.0 by the addition of CaC0 3 at a rate of 1 g 1- 1 . KNYPL (1976) compared
CaC0 3 and CaHP0 4 as pH buffering agents in culture media for Spirodela
oligorrhiza. Although use of CaC0 3 gave markedly better control of pH, growth
was superior with CaHP0 4 . However, as a saturated solution of CaHP0 4 corre-
sponds with a phosphate concentration of approximately 1,800 /lM, use of this
salt could lead to the maintenance of phosphate concentrations in the toxic
range for susceptible species (cf. ASHER 1978).

2.1.4.3 Effects of Nutrient Ion Concentrations

The usefulness of conventional solution culture methods for studying effects

of nutrient ion concentrations on nutrient uptake and plant growth is severely
restricted by the problem of nutrient depletion. Attention was drawn to this
problem long ago by STILES (1916) but it is clear from a study of the modern
1.3 Modern Solution Culture Techniques 103

plant nutrition literature that many research workers still do not take adequate
account of nutrient depletion when designing solution culture experiments or
interpreting their results.
In any particular situation the magnitude of the depletion problem will
depend on a number of factors, important amongst which are (1) the accuracy
with which concentration needs to be controlled to achieve the objectives of
the experiment, (2) the range of concentrations over which the experiment is
to be conducted, (3) the volume of solution provided per unit weight of roots,
and (4) the mean rate of uptake per unit root weight at the specified concentra-
tions. Since the root weight of young plants will often increase many-fold during
the course of an experiment, the maximum probable root weight should be
used when designing measures to restrict solution depletion to acceptable levels.
Probable levels of depletion can be characterized either in terms of the ratio
of anticipated daily uptake to the total nutrient supply (ASHER 1981) or in
terms of the frequency of renewal which would be required to limit depletion
to a specified percentage of the initial concentration of an ion. In the latter
case the maximum permissible time interval in hours between solution renewals
(T) can be calculated approximately from the following equation:

D V C
T=-·_·- (1)
100 R U
where D = maximum acceptable depletion (%), V = volume of solution per pot
(or per plant) (litres), C = initial concentration of an ion in the solution (IlM),
R=root weight per pot (or per plant) (g fresh wt.), and U = uptake rate per
unit root weight (Ilmol g - 1 fresh wt. h - 1) at concentration C. (Strictly speaking,
the value of U should be for the mean concentration between solution renewals,
i.e. C(200-D). However, for small values of D the slight underestimation of
200
T resulting from using the U value corresponding with C can be neglected).
Values of U taken from short-term uptake experiments with low salt roots
may seriously overestimate the average uptake rates that would occur in longer-
term experiments (JOHANSEN et al. 1968) and hence yield misleadingly low values
of T. Probably the best source of U values for experiments of several days
duration or longer are flowing culture experiments in which the plant root
systems have been exposed to a constant external concentration of the ion
for an extended period of time.
In Table 3 maximum acceptable times between solution renewals have been
calculated for solution to root (~) ratios of 0.1, 0.5 and 2.5 I g-l fresh root
wt., and maximum depletions (D) of 5% or 10%, using published uptak,e data
from flowing culture experiments on zinc (CARROLL and LONERAGAN 1969) and
phosphorus (LONERAGAN and ASHER 1967) as an example of a micronutrient
and a macronutrient element. For most purposes V values in excess of 0.5 I g-l
R
are probably not practical. In Carroll and Loneragan's study, root weights
after 46 days at 0.25 IlM zinc ranged from 85 g poC 1 (lucerne) to 154 g poC 1
104 c.J. ASHER and D.O. EDWARDS:

Table 3. Maximum acceptable times between solution renewals calculated for various
solution: root ratios and maximum depletion levels, using published information on
rates of zinc and phosphorus uptake

C Adequacy for
(IlM) plant growth

D=5% D=10%

0.1 0.5 2.5 0.1 0.5 2.5

Zinc a Solution renewal times, T (h)

0.01 Deficient 0.00013 0.4 2.0 9.8 0.8 3.9 20
0.25 Adequate 0.00076 1.6 8.2 41 3.3 16.4 82
6.25 Adequate-toxic 0.0134b 2.3 11.7 58 4.5 23 117

Phosphorus C
0.04 Deficient 0.013 0.02 0.08 0.4 0.03 0.15 0.8
5 Adequate 0.437 0.06 0.3 1.4 0.1 0.6 2.9
24 Adequate-toxic 0.512d 0.2 1.2 5.9 0.5 2.3 11.7

a CARROLL and LONERAGAN (1969) C LONERAGAN and ASHER (1967)

b Toxic for 3 out of 5 species studied d Toxic for 3 out of 8 species studied

(wheat). At V = 0.5 these root weights would require pot sizes of 42.5 and
R
77 I respectively. Again, for experiments of more than a few days' duration
it is probably impractical to renew nutrient solutions more often than once
or at most twice per day because of the amount of labour involved.
The data in Table 3 show that for zinc concentrations in the adequate to
toxic range it may be practical to use a simple, large volume water culture
system particularly if 10% depletion is acceptable between solution renewals.
However, the data indicate that adequate control of zinc concentration in the
deficiency range is not likely to be achieved at practical values of T and V
R
(Table 3). For phosphorus, the data show that, because of the much higher
rates of uptake at equivalent concentrations, adequate control of solution phos-
phorus concentration could not be obtained even at levels toxic to some species.
The situation with other mineral elements will vary depending on the magnitude
of the values to be inserted in Eq. (1). However, the limited evidence at present
available suggests that effective control of concentration in simple solution cul-
ture systems is likely t() be practical only with respect to studies of toxicity
of the more slowly absorbed mineral elements, e.g. aluminum, micronutrients,
and some environmental contaminants.
In sand culture systems the depletion problem tends to be intensified by
the following factors:
1. relatively small volumes of solution are retained in the pots after drainage;
2. the solution retained in the pots is unstirred, hence zones of local depletion
may occur adjacent to root surfaces;
1.3 Modem Solution Culture Techniques 105

3. in sub-irrigation systems, a substantial part of the total solution volume

is stored outside the pots between flushing cycles, and consequently is not
available for uptake for much of the time.
The impact of these factors can be lessened by increasing the frequency
of flushing, but cannot be eliminated entirely. Sand cultures are thus unsuitable
for quantitative studies on effects of external nutrient ion concentrations. As
pointed out by ASHER (1980), nutrient depletion in solution culture experiments
leads to overestimation of critical external concentrations for both deficiency
and toxicity.

2.1.5 Establishment of Critical Tissue Concentrations

Critical tissue concentrations for the diagnosis of nutrient deficiencies and toxici-
ties are frequently established from water culture or sand culture experiments.
Although many plant and environmental factors have been shown to affect
measured critical concentrations (BATES 1971), it has been widely assumed that
critical tissue concentrations are a comparatively stable plant characteristic un-
likely to be affected by temporal variation in the external supply of the element
concerned. However, recent work of SPEAR et aL (1978, 1979) raises serious
doubts about this assumption. Thus critical potassium concentrations in index
leaves and whole tops of cassava plants grown in small pots (11 solution
plant - 1) in which depletion of solution potassium was rapid, were much lower
than in plants grown in larger pots (2.5 or 8 1 plane 1) in which depletion
was slower, or in flowing solution culture in which depletion was prevented.
Similar results were obtained in a flowing culture experiment in which depletion
was simulated by transferring plants from various constant potassium concen-
trations to a nutrient solution containing no added potassium.
These results suggest that arbitrary decisions concerning details of the solu-
tion culture technique to be employed can have large effects on measured critical
concentrations. The results also call into question the validity of using critical
concentrations established in solution culture experiments for diagnosis of nutri-
tional problems in the field where temporal variations in nutrient supply may
be quite different from those in a particular solution culture experiment.

2.1.6 Control of Plant Nutrient Status

One of the consequences of nutrient depletion in conventional solution culture
systems is that large changes in the nutrient status of plants can occur during
the course of an experiment. Thus plants which ultimately show severe deficiency
of an element may have passed through an earlier period of luxury consumption
or even toxicity of the same element (ASHER and COWIE 1970). Hence effects
of nutrient status on the physiology and growth of plants are difficult to exam-
ine. Even in sand cultures watered at regular intervals with fresh nutrient solu-
tion the nutrient status of plants is likely to decline, since the total supply
will be fixed by the characteristics of the system (initial concentration, volume
retained, frequency of application) whereas growth and absolute demand for
nutrients will tend to increase exponentially with time. Although these problems
106 C.J. ASHER and D.G. EDWARDS:

can be avoided by means of flowing solution culture techniques (Sect. 2.3),

most plant science laboratories do not yet have the necessary equipment. Hence
an alternative approach to the experimental control of plant nutrient status
is required.
ASHER and COWIE (1970) devised a method by which the nitrogen status
of grain sorghum plants could be controlled in water culture from an early
seedling stage to grain maturity. Frequent small additions of nitrogen were
made to each nutrient solution. In the control treatment the size and frequency
of these were calculated from a previously established growth curve, and data
on average concentrations of nitrogen in the tissues of healthy plants, to provide
adequate nitrogen at all times. In the deficient treatments fixed percentages
of the amounts of nitrogen applied in the control treatment were applied on
each occasion. Since the success of the method depends on the careful prepara-
tion of a program of nutrient additions ASHER and COWIE (1970) suggested
that the technique be called Programmed Nutrient Addition.
ASHER and COWIE (1970) showed that nutrient addition programs could
be prepared for deliberate variations in nutrient status during the course of
an experiment so that effects of nutrient deficiency at particular stages of growth
could be studied. Subsequently LEE et al.(1981) used the technique to control
the nitrogen status of ginger in the field (Fig.2a). MALIK et al. (1978) used
a variation of the technique to control the nitrogen status of cotton plants
under greenhouse conditions. As no growth curve was available, a nitrogen
concentration of approximately 4,000 J.lM was maintained in the control pots
and uptake in this treatment computed every second day from solution analyses.
Additions of nitrogen to the nitrogen-deficient treatments were then made on
the basis of fixed percentages of the uptake in the control treatment. Where
experiments are of short enough duration to be confined to the period of expo-
nential growth, nutrient addition programs can be devised from theoretical
growth curves based on the initial plant weight and assumptions concerning
the probable relative growth rate of the control plants (Fig. 2b). A somewhat
similar approach was adopted in mist culture experiments of INGESTAD and
LUND (1979) who proposed the concept of a relative nutrient addition rate
analogous to relative growth rate.
The technique has the advantages that no special equipment is required
and that there is no restriction to the number of treatments which can be
employed. In view of the success obtained with it so far under both greenhouse
and field conditions the method appears to have wide applicability.

2.1.7 Study of Symbiotic Associations with Microorganisms

2.1.7.1 Nitrogen-Fixing Systems

Sand culture and water culture methods have been used extensively in the study
of symbiotic nitrogen fixation in legumes and other higher plants. The Leonard
jar technique (LEONARD 1944) is a specialized sand culture system developed
specifically for such studies and has become a standard technique in many
laboratories. An automatic sub-irrigated sand culture system devised by
I.3 Modern Solution Culture Techniques 107

50
1000
a b

_ 800
40
o o
L
0.
"-
en
E
"0
30
~ 600 "0
Q)
0.
0.
o 0.
0.
C o
Q)
en c
~ 400 ~ 20
o
c
.n

.8o
~

200 10

o
Ti me (days)

Fig. 2a, b. Examples of nutrient addition programs. a Nitrogen addition programs used
to control the nitrogen status of ginger in the field. (Modified after LEE et al. 1981).
b Boron addition programs used by FORNO et al. (1979) to control the boron status
of cassava in water culture. (Two additional programs causing moderate and severe
boron deficiency not shown)

ANDREW (1974) has proved effective in the study of soil acidity factors (pH,
Ca, AI) affecting nodulation and nitrogen fixation in legumes (ANDREW 1976;
CARVALHO et al. 1981). Water culture studies have also contributed much valu-
able information on effects of soil acidity factors (LONERAGAN and DOWLING
1958; MUNNS 1968).
Considerable scope exists for the further application of solution culture
methods to the study of factors affecting nodulation in legumes. Unfortunately,
however, some legume species do not nodulate readily in simple water culture
systems and there is an urgent need for information on the reasons why this
is so. One possibility which has not yet been examined in detail is that high
initial nutrient concentrations may cause morphological changes in the roots
which hinder nodulation. Thus BREWSTER et al. (1976) showed that in tlie non-
leguminous species rape, the growth of root hairs was completely inhibited
at phosphorus concentrations of 100 11M and higher. If a similar situation exists
in some legumes in which infection by Rhizobium is via root hairs, the common
use of initial phosphate concentrations of the order of 1,000 11M would effective-
ly remove the potential sites of infection. Another possibility worthy of investiga-
tion is that inadequate production or retention of root exudates at root surfaces
108 c.J. ASHER and D.G. EDWARDS:

results in conditions unfavourable for multiplication of free-living Rhizobium

prior to infection.

2.1.7.2 Mycorrhizal Associations

The importance of ectomycorrhizal associations in the growth of some woody
perennial species has long been recognized, and during the past two decades
there has been growing recognition of the role of endomycorrhizas in the growth
of many plant species, particularly under conditions of limiting mineral nutrient
supply. So far only limited use of solution culture methods has been made
for such studies. Recent research suggests that it may be important to use
rather dilute culture solutions. Thus OWUSU-BENNOAH and MOSSE (1978) showed
that infection of lettuce roots by a v.a. mycorrhizal fungus was reduced to
zero by a mineral nitrogen concentration of 2,850 11M supplied as nitrate or
an equimolar mixture of nitrate and ammonium. Similarly HOWELER et al. (1981)
showed that formation of vesicular arbuscular mycorrhizas in cassava occurred
at 0.1 11M and 111M phosphorus but not at 10 or 100 11M phosphorus.

2.1.8 Commercial Crop Production (see also Sect. 2.3.5)

The technical feasibility of crop production in solution culture systems was
demonstrated in the U.S.A. in the late 1920's. Since then substantial use of
these techniques, particularly gravel culture, has been made in the production
of high-value greenhouse crops (DOUGLAS 1976). The main constraints to expan-
sion of hydroponic crop production are economic rather than technical. This
form of crop production becomes particularly attractive where centres of popu-
lation exist in highly arid areas in which insufficient water is available for irriga-
tion, or in situations where soil suitable for horticultural production is unavail-
able within a reasonable distance of the population centre. It could be that
increasing costs of transporting produce associated with rising prices of liquid
fuel, will lead to greater use of hydroponic production methods close to major
cities and perhaps within them, flat roof tops being quite suitable for this
purpose.

2.2 Mist Culture

Plants may be grown successfully with their roots suspended in a fine mist
of nutrient solution generated pneumatically or by allowing the solution to
drip onto a rapidly rotating disk. HEWITT (1966) discusses a number of varia-
tions of the technique. INGESTAD and LUND (1979) have described a sophisticated
mist culture apparatus incorporating control of solution temperature, pH, and
electrical conductivity, with provision for automatic nutrient additions. The
main advantage of mist culture over other culture methods is the ease with
which the root systems can be studied. A disadvantage of the method is that
even brief interruptions to mist production due to mechanical problems, e.g.
power failure or blocked atomizer nozzle, can lead to plant damage.
1.3 Modern Solution Culture Techniques 109

Usually concentrated nutrient solutions are supplied at rather low volumes

per unit time. For example MARTIN and HENDRIX (1966) supplied Hoagland's
solution to wheat plants at rates of the order of 65 ml pot - 1 h - 1. At such
low solution supply rates substantial depletion of the solution in contact with
the roots must occur (see Sects. 1.3.1,2.3.1). Hence the method appears unsuited
for critical studies on effects of solution composition on plant growth.

2.3 Flowing Solution Culture

Continuously flowing solution cultures provide the most practical means so

far devised for experimental control of root environment parameters. Flowing
culture techniques allow plants to be grown for extended periods of time with
precise control of root temperature, pH and nutrient ion concentrations in
dilute solutions comparable in composition with soil solutions. Such control
is essential for quantitative studies on effects of various root environment param-
eters on nutrient uptake and growth in higher plants.
EDWARDS and ASHER (1974) made a distinction between those flowing cul-
ture systems in which the nutrient solution enters the pots at the required concen-
tration of each ion under investigation ["Type (a) "systems] and those in which
treatment concentrations are held constant by careful balancing of the rate
of addition of one or more concentrated stock solutions against the rate of
removal of ions by the test plants ["Type (b)" systems]. A well-known example
of a Type (b) system is that devised by VAN DEN HONERT and subsequently
described in detail by BECKING (1956). Type (b) flowing culture systems employ
low flow rates and require only relatively small volumes of nutrient solution.
However, the need to make frequent adjustments to stock solution flow rates
on an individual pot basis means that a single operator can effectively manage
only a small number or pots. It is perhaps for this reason that the majority
of laboratories which have installed flowing culture equipment in recent years
have opted for Type (a) systems.

2.3.1 The Flow Rate Problem

In Type (a) flowing culture systems it is essential that the flow rate through
the pots is high enough to prevent appreciable alteration of solution composition
as a result of contact with the roots. EDWARDS and ASHER (1974) showed that
(F) the lowest acceptable flow rate in I pot -1 min - 1 could be calculated approxi-
mately from the following equation:

F=100R.U (2)
D C

where R is the weight of roots per pot (g fresh wt.), U is the rate of uptake
of the test element (Jlmol g -1 fresh wt. min - 1) at treatment concentration C
(JlM), and D is the maximum acceptable depletion of the test ion (%). Relation-
ships between U and C are such that higher flow rates are required in the
110 C.J. ASHER and D.G. EDWARDS:

Table 4. Nutrient solution flow rates used in experiments with Type (a)a flowing culture
systems

Reference Flow rate Reference Flow rate

(ml poe l (rol pot- l
min-I) min-I)

RORISON (personal communication) 0.5 CLEMENT et al. (1974) 729

FAGERIA (1976) 2 AsHER et aI. (1965) 695--900
SABET et al. (1964) 1.4-16.7 BHAT (1980) Approx
1000
REISENAUER (1969) (H7.4 ASHER and EDWARDS (1978) 1,620
VAN HAl andLAUDEwUT (1966) 3.3--30 EDWARDS and ASHER (1974) 3--3000
TIBBITTS et al. (1978) 200 VAN LUNE and VAN GOOR 32,000
(1975)
TEMPLE-SMITH and MENARY (1977) 704

a For explanation see text

deficiency range than in the adequate to toxic range of external concentrations

(EDWARDS and ASHER 1974; ASHER 1981). Limited data suggest that higher
rates are required for studies with macronutrients than with micronutrients
(ASHER 1981). Both actual experimentation, and substitution into Eq. (2) of
published data on uptake rates at particular concentrations, suggest that flow
rates of the order of hundreds of ml min - 1 will often be required to control
treatment concentrations to within 5% or 10% of their nominal values and
that still higher flow rates will sometimes be needed.
Increases in R with time due to root growth mean that minimum acceptable
flow rates also increase with time. REISENAUER (1969) sought to overcome this
problem by constructing a system with continuously variable flow rates. How-
ever, the highest flow rate used (17.4 ml min - 1) was insufficient to prevent
severe depletion of his nutrient solution by young wheat seedlings. Most modern
flowing culture systems have incorporated constant high flow rates
(:> 700 ml min -1) (Table 4) to ensure adequate control of concentration at the
highest R values likely to be encountered. All of these have been based on
the recirculating principle introduced by ASHER et al. (1965). In most cases
provision has been made for monitoring of the composition of the recirculating
nutrient solution and the slow addition of reagents to maintain control of pH
and test ion concentrations.

2.3.2 Composition of Flowing Culture Solutions

Although much remains to be learned about critical external concentrations
for plant growth (cf. review by ASHER 1978), it is already clear that vigorous
growth can be obtained in quite dilute flowing culture solutions (Table 5, Fig. 3).
However, not all users of the technique have taken full advantage of the oppor-
tunity to work at such low concentrations. Thus in a study of nitrate uptake
by Lolium perenne, CLEMENT et al. (1978) added additional calcium to all treat-
ments to ensure a minimum concentration of 4,000 liM despite the observation
1.3 Modern Solution Culture Techniques 111

Table 5. Concentrations of elements in nutrient solutions employed in flowing culture

experiments (JlM)

Reference ISLAM WILD ASHER and TEMPLE-SMITH CLEMENT

et al. et al. LONERAGAN and MENARY et al.
(1980) (1974) (1967) (1977) (1978)

Element (or Ion)

Nitrogen
NO; 250 700 750 3000 1.4-143,000
NHt 100
Potassium 250 1-33 250 1000 26
Calcium 250 420-470 250 1000 4,000-75,500
Magnesium loa 100 100 400 100
Phosphorus 15 50 0.04-25 0.06-7.7 50
Sulfur 260-560 100 100 400 4125
Chlorine 5 125 100 7.2 1.7
Sodium 40-63 3.6
Silicon 10
Iron 20 5 2 3.6 1.1
Boron 3 2.5 3 18 5
Manganese 0.25 0.5 1 3.6 1
Zinc 0.5 0.05 0.5 0.3 0.08
Copper 0.1 0.02 0.1 0.12 0.03
Molybdenum 0.02 0.005 0.02 0.04 0.05
Cobalt 0.04 0.04
Total nutrients 1114-1437 1454-1536 1658-1682 5837-5844 8311-222,810

a Adequate for plant growth at pH ~ 5.5, but inadequate at pH .::: 4.0

of LONERAGAN et al. (1968) that this species requires only 2.5 11M calcium for
maximum growth.

2.3.3 Research Applications

Flowing solution culture represents a powerful tool for research into certain
aspects of plant physiology and soil-plant relationships. Much greater use of
the technique is likely in the future as increasing numbers of research workers
come to recognize the importance of maintaining accurate control of the root
environment in plant nutrition studies. The recent description of two flowing
culture systems suitable for use in growth cabinets (TmBITTS et al. 1978; BRAT
1980) raises exciting prospects for simultaneous control of root and shoot
environments in plant physiological research.
Flowing solution culture at present offers the only practical means of obtain-
ing quantitative data on the effects on plant growth of external nutrien.t ion
concentrations ranging from deficiency to adequacy. The technique is also highly
suitable for similar studies in the toxicity range, although in the case of some
slowly absorbed elements, comparable results could probably be obtained in
conventional water culture systems (cf. Sect. 2.1.4.3). Much work remains to
be done on these concentration x growth relationships. Thus for many plant
species no reliable information is available concerning limiting external concen-
112 C.J. ASHER and D.G. EDWARDS:

Fig. 3. Vigorous plant growth

obtained in a flowing solution
culture experiment at the Uni-
versity of Queensland, using a
culture solution with a total
nutrient concentration of
3780 11M

trations for the deficiency or toxicity of even a single element. Similarly for
a number of biologically important elements no reliable limiting concentrations
have been established even for a single genotype. In addition, little is known
about interactions amongst ions at concentrations likely to be encountered in
soil solution. In studies with legumes, practically all the existing information
concerns non-nodulated plants. However, there is good evidence that nodulation
is more sensitive to adverse root environments than is the growth of the non-
nodulated host plant (LoNERAGAN and DOWLING 1958; LOWTHER and LONERA-
GAN 1968; CARVALHO et al. 1981). Hence more research is required on limiting
external concentrations for the nodulation and growth of legumes.
Flowing solution culture equipment incorporating efficient temperature
control is ideally suited for studying effects of root temperature on ion uptake
and plant growth. In a recent experiment at the University of Queensland,
WARRINER (unpublished) showed that in some species, above-optimum root
temperature can increase markedly the susceptibility of plants to phosphorus
toxicity (subterranean clover) and manganese toxicity (soybean, pigeon pea).
I.3 Modern Solution Culture Techniques 113

Further research is needed to establish the extent to which critical external

concentrations may be influenced by root temperature.
ASHER and OZANNE (1977) used flowing solution culture to show that indi-
vidual members of apparently uniform sets of Arctotheca calendula seedlings
differed widely in susceptibility to potassium deficiency when grown at a con-
stant low potassium concentration (2.6 ,lM). Hence flowing solution culture
provides a potential method of identifying genotypes with resistance to specific
nutritional disorders (ASHER and EDWARDS 1978). Again, JOHANSEN et al. (1968)
showed that mean rates of potassium uptake by roots of intact plants grown
in flowing culture at constant potassium concentrations were much lower than
uptake rates by potassium-starved roots at similar concentrations in short-term
uptake experiments. Data from flowing culture studies are thus highly relevant
to the important question of the mechanisms by which higher plants regulate
mineral uptake and mineral composition (cf. review by PITMAN and CRAM 1977).
A further potential use for flowing culture technology is in producing plant
material at known and constantly maintained levels of nutrient stress for studies
on the biochemical role of specific mineral nutrients.
Recently, HOWELER et al. (1981) demonstrated the feasibility of using flowing
solution culture to study effects of mycorrhizas on the mineral nutrition of
higher plants. They showed that growth and phosphorus uptake by cassava
from a dilute nutrient solution containing 1 ,lM phosphorus was markedly stim-
ulated by v.a. mycorrhizas. Since no mycorrhizas formed at phosphorus concen-
trations ~ 10 ,lM, study of the root-fungus association would have been ex-
tremely difficult by any other solution culture technique. Further research is
needed to see to what extent the technique is applicable to the study ofmycorrhi-
zas in other plant species.

2.3.4 Likely Future Developments

Considerable scope exists for improving the performance and ease of operation
of existing flowing culture equipment. Although high flow rates can be used
to ensure that only small changes in solution composition occur with each
passage of the solution through the pots, a large volume of solution may be
required to provide an adequate reserve of both treatment and basal nutrient
ions in dilute solutions, e.g. up to 2,400 I unie! (ASHER and EDWARDS 1978).
The need for such large volumes of solution would be reduced greatly if a
system could be devised for equilibrating the nutrient solutions with suitable
cation and anion exchangers. Again, the concentrations of ions corresponding
with nutrient deficiency are in a number of cases too low to be measured directly,
so that time-consuming concentration procedures are needed prior to analysis
of the nutrient solutions (e.g., potassium, ASHER and OZANNE 1967;, zinc,
CARROLL and LONERAGAN 1968; calcium, LONERAGAN et al. 1968; phosphorus,
JINTAKANON et al. 1975).
The long-term solution to the problem of monitoring and controlling solu-
tion composition may come from developments in the field of electrochemistry.
Glass electrode-autotitrator systems have long been in use for controlling solu-
tion pH and have been incorporated into some flowing culture installations
114 C.J. ASHER and D.G. EDWARDS:

(ASHERet al.1965; CLEMENTet al.1974; ASHER and EDWARDS 1978; MORITSUGU

and KAWASAKI 1979). The control of ions other than H+ using appropriate
ion-specific electrodes is foreshadowed by the description of such a system for
controlling potassium and nitrate concentrations in flowing culture (CLEMENT
et al. 1974). However, for some elements of biological interest no electrode
is yet available and for many others the electrodes at present available lack
the necessary sensitivity, selectivity, or speed of response at low concentrations
to be effective in controlling solution composition in the low concentration
range.

2.3.5 Commercial Crop Production

Although close control of the root environment is not essential for hydroponic
crop production, there has been growing interest in continuously flowing culture
methods for this purpose during the last decade.

2.3.5.1 Use of Sewage Effluent

Recent research at the University of California has demonstrated the feasibility
of using secondary treated sewage effluent as a substrate for hydroponic vegeta-
ble production (BERRY et al. 1977; WALLACE et al. 1978). With troughs 6.2 m
long and holding approximately 1,000 I of effluent, some gradients in plant
growth were observed along the length of the bed. Thus with lettuce at a flow
rate of 2.8 x 103 1 day-1 (approx. 2 ml min -1) and with tomato at a flow rate
of 2.8 x 104 1 day-1 (approx. 20 ml min -1) there was some growth reduction
at the outlet end of the bed. The causes of these reductions were not determined.
Clearly further research is needed on the technique but it opens up ewiting
prospects for reducing hydroponic production costs by recycling mineral nu-
trients which might otherwise have contributed to the pollution of natural waters
adjacent to centres of population.

2.3.5.2 Nutrient Film Technique (N.F.T.)

N.F.T. is a Type (a) recirculating flowing culture system (for definition see
Sect. 2.3) in which the nutrient solution flows down shallow, gently sloping
plastic gullies (slopes 1: 30 to 1: 100) in which are placed bare rooted or in
some cases potted seedlings (for review see COOPER 1976). The pH of the circulat-
ing nutrient solution is controlled by additions ofH 3 P0 4 or HN0 3 and periodic
additions of a nutrient mixture are made to keep the electrical conductivity
of the solution between 2 and 3 mS cm - 1. The technique has gained substantial
commercial acceptance, but it is clear that some technical problems remain
and further research is needed. Thus RESH (1976) compared ten variations of
N.F.T. for tomato production and found all to be inferior to standard commer-
cial gravel culture.
In particular, research is needed on effects of flow rate. COOPER (1976)
stresses the need to use flow rates sufficiently low to ensure that at least portion
of the root system is exposed to air to ensure an adequate oxygen supply.
1.3 Modern Solution Culture Techniques 115

However MAHER (1976) favours higher flow rates which result in complete
submergence of the root mat in the bottom of the gully and has shown that
for even short gullies (7 m long) an oxygen deficit develops towards the outlet
of the gully at flow rates below 4 I min - 1. Higher flow rates such as those
advocated by Maher should also reduce gradients along the gully iIi mineral
nutrient concentration and solution temperature.

3 Summary and Conclusions

The wide range of solution culture techniques available to plant scientists today
carries with it a responsibility to select techniques consistent with the objectives
of each particular study. In general the simplest acceptable method should be
used.
Flowing solution culture is the only practical means available at present
for studying relationships between plant growth and external nutrient ion con-
centrations in the deficiency range, and will usually be the best method also
for studies in the toxicity range. However the large volumes of solution and
large amounts of surface (pots, piping, tanks, pumps) in contact with the solu-
tion make the method unsuitable for studies on nutrient essentiality. Such studies
are usually best done in simple non-renewed water culture. Simple water culture
and sand culture methods are well suited to the study of symptoms of nutritional
disorders. However, recent research raises doubts concerning the suitability of
any water culture method for establishing critical tissue concentrations that
may be applied to diagnosis in the field. Flowing solution culture has the advan-
tage over other water culture methods that an adequate total supply of nutrients
can be provided at low and controlled concentrations comparable with those
commonly occurring in soil solutions. Programmed nutrient addition (cf. Sect.
2.1.6) offers a means by which an adequate total supply of nutrients can be
provided without exposure of the roots to the high initial concentrations charac-
teristic of conventional water culture and sand culture systems. However, the
method is not suitable for studies in which control of external concentrations
is required.
Despite the long period of time over which existing solution culture methods
have been evolving, substantial scope still exists for improvement in these meth-
ods. Considerable opportunities exist also for novel applications of present tech-
niques to problems in plant physiology and soil-plant relationships.

References

Addoms RM (1937) Nutritional studies on loblolly pine. Plant Physiol12: 119-205

Ahmed S, Evans HJ (1960) Cobalt - a micronutrient element for the growth of soybean
plants under symbiotic conditions. Soil Sci 90: 205-21 0
Andrew CS (1974) Automatic sub-irrigation sand culture. technique for comparative
studies in plant nutrition. Lab Pract 23: 20-21
116 C.J. ASHER and D.G. EDWARDS:

Andrew CS (1976) Effect of calcium, pH and nitrogen on the growth and chemical
composition of some tropical and temperate pasture legumes. I Nodulation and
growth. Aust J Agric Res 27:611-623
Amon DI, Johnson CM (1942) Influence of hydrogen ion concentration on the growth
of higher plants under controlled conditions. Plant PhysioI17:525-539
Asher CJ (1978) Natural and synthetic culture media for spermatophytes. In: Rechcigl
M Jr (ed) CRC handbook series in nutrition and food, section G: diets, culture
media, food supplements, vol III. CRC Press, Cleveland, pp 575-609
Asher CJ (1981) Limiting external concentrations of trace elements for plant growth:
use of flowing solution culture techniques. J Plant Nutr 3:163-180
Asher CJ, Cowie AM (1970) Programmed nutrient addition - a simple method for con-
trolling the nutrient status of plants. Proc Aust Plant Nutr Conf, Mt Gambier, S
Aust Sect 1(b), pp 28-32
Asher CJ, Edwards DG (1978) Relevance of dilute solution culture studies to problems
of low fertility tropical soils. In: Andrew CS, Kamprath EJ (eds) Mineral nutrition
of legumes in tropical and subtropical soils. CSIRO, Melbourne
Asher CJ, Loneragan JF (1967) Response of plants to phosphate concentration in solution
culture. I. Growth and phosphorus content. Soil Sci 103:225-233
Asher CJ, Ozanne PG (1967) Growth and potassium content of plants in solution cultures
maintained at constant potassium concentrations. Soil Sci 103: 155-161
Asher CJ, Ozanne PG (1977) Individual plant variability in susceptibility to potassium
deficiency: some observations on capeweed [Arctotheca calendula (L.), Levyns]. Aust
J Plant PhysioI4:499-503
Asher CJ, Ozanne PG, Loneragan JF (1965) A method for controlling the ionic environ-
ment of plant roots. Soil Sci 100: 149-156
Barber DA, Martin JK (1976) The release of organic substances by cereal roots into
soil. New Phytol 76:69-80
Barber SA (1962) A diffusion and mass-flow concept of soil nutrient availability. Soil
Sci 93: 39-49
Bates TE (1971) Factors affecting critical nutrient concentrations in plants and their
evaluation: a review. Soil Sci 112: 116-130
Becking JH (1956) On the mechanism of ammonium ion uptake by maize roots. Acta
Bot Need 5: 1-79
Berry WL, Wallace A, Lunt OR (1977) Recycling municipal wastewater for hydroponic
culture. Hort Sci 12:186
Bhat KKS (1980) A low-cost, easy-to-install flow culture system for use in a constant
environment cabinet. J Exp Bot 31: 1435-1440
Bond G (1951) The fixation of nitrogen associated with the roots of Myrica gale L.
with special reference to its pH relation and ecological significance. Ann Bot
15:447-459
Bowen GD, Rovira AD (1976) Microbial colonization of plant roots. Ann Rev Phyto-
pathoI14:121-144
Brewster JL, Bhat KKS, Nye PH (1976) The possibility of predicting solute uptake
and plant growth response from independently measured soil and plant characteristics.
IV The growth and uptake of rape in solutions of different phosphorus concentration.
Plant Soil 44: 279-293
Burd JS, Martin JC (1924) Secular and seasonal changes in the soil solution. Soil Sci
18:151-167
Carroll MD, Loneragan JF (1968) Response of plant species to concentrations of zinc
in solution. I Growth and zinc content of plants. Aust J Agric Res'19:859-868
Carroll MD, Loneragan JF (1969) Response of plant species to concentrations of zinc
in solution. II Rates of zinc absorption and their relation to growth. Aust J Agric
Res 20:457-463
Carvalho de MM, Edwards DG, Andrew CS, Asher CJ (1981) Aluminum toxicity, nodu-
lation and growth of Stylosanthes species. Agron J 73: 261-265
Chen CH, Lewin JC (1969) Silicon as a nutrient element for Equisetum arvense. Can
JBot47:125-131 .
1.3 Modern Solution Culture Techniques 117

Clement CR, Hopper MJ, Canaway RJ, Jones LHP (1974) A system for measuring
the uptake of ions by plants from flowing solutions of controlled composition. J
Exp Bot 25: 81-99
Clement CR, Hopper MJ, Jones LHP (1978) The uptake of nitrate by Lotium perenne
from flowing nutrient solution. I. Effect of NO; concentration. J Exp Bot 29: 453-464
Cooper AJ (1976) Crop production with nutrient-film technique. Proc 4th'lnt Congr
Soilless Cult (IWOSC), Las Palmas
Dixon NE, Gazzola C, Blakely RL, Zerner B (1975) Jackbean urease (EC 3.5.1.5). A
metalloenzyme. A simple biological role for nickel? J Am Chern Soc 97:4131-4133
Douglas JS (1976) Advanced guide to hydroponics. Drake, New York
Edwards DG, Asher CJ (1974) The significance of solution flow rate in flowing culture
experiments. Plant Soil 41: 161-175
Epstein E (1972) Mineral nutrition of plants: principles and perspectives. Wiley and
Sons, New York
Fageria NK(1976) Effect ofP, Ca, and Mg concentrations in solution culture on growth
and uptake of these ions by rice. Agron J 68: 726-732
Forno DA, Asher CJ, Edwards DG (1979) Boron nutrition of cassava and the boron
x temperature interaction. Field Crops Res 2: 265-279
Gericke WF (1929) Aquaculture - a means of crop-production. Am J Bot 16:862
Gillman GP, Bell LC (1978) Soil solution studies on weathered soils from tropical North
Queensland. Aust J Soil Res 16: 1-11
Hai van T, Laudelout H (1966) Phosphate uptake by intact rice plants by the continuous
flow method at low phosphate concentrations. Soil Sci 101 :408-417
Hallsworth EG, Wilson SB, Greenwood EAN (1960) Copper and cobalt in nitrogen
fixation. Nature 187:79-80
Hewitt EJ (1966) Sand and water culture methods used in the study of plant nutrition.
Commonw Bur Hortic Plant Crops Tech Commun 22 (revised)
Hoagland DR, Arnon DI (1950) The water culture method for growing plants without
soil. ColI Agric UC Berkeley, USA. Cal Agric Exp Stn Circ 347: 1-32
Howeler RH, Edwards DG, Asher CJ (1981) Application of the flowing solution culture
techniques to studies involving mycorrhizas. Plant Soil 59: 179-183
Ingestad T, Lund AB (1979) Nitrogen stress in birch seedlings. I. Growth technique
and growth. Physiol Plant 45: 137-148
Islam AKMS, Edwards DG, Asher CJ (1980) pH optima for crop growth. Results of
a flowing solution culture experiment with six species. Plant Soil 54: 339-357
Jintakanon S, Kerven GL, Edwards DG, Asher CJ (1975) Measurement of low phospho-
rus concentrations in nutrient solutions containing silicon. Analyst 100:408-414
Johansen C, Edwards DG, Loneragan JF (1968) Interactions between potassium and
calcium in their absorption by intact barley plants. II. Effects of calcium and potassi-
um concentration on potassium absorption. Plant Physiol 43: 1722-1726
Johnson CM, Stout PR, Broyer TC, Carlton AB (1957) Comparative chlorine require-
ments of different plant species. Plant Soil 8: 337-353
Knop W (1865) Quantitative Untersuchungen iiber die Ernahrungsprozesse der Pflanzen.
Landwirtsch Vers Stn 7:93-107 (cited by Hewitt 1966)
Knypl JS (1976) Culture of Spirodela otigorrhiza in ammonium-media buffered with
calcium carbonate or calcium phosphate. Biochem Physiol Pflanz 170: 243-252
Lee MT, Asher CJ, Whiley AW (1981) Nitrogen nutrition of ginger (Zingiber officinale).
I. Effects of nitrogen supply on growth and development. Field Crops Res 4: 55-68
Leonard LT (1944) Method of testing legume bacteria cultures and results of tests of
commercial inoculants in 1943. US Dep Agric Circ 703
Loneragan JF, Asher CJ (1967) Response of plants to phosphate concentration in solution
culture: II. Rate of phosphate absorption and its relation to growth. Soil Sci
103:311-318
Loneragan JF, Dowling EJ (1958) The interaction of calcium and hydrogen ions in
the nodulation of subterranean clover. Aust J Agric Res 9: 464-472
LoneraganJF, Snowball K, Simmons WJ (1968) Response of plants to calcium concentra-
tion in solution culture. Aust J Agric Res 19: 845-857
118 C.J. ASHER and D.G. EDWARDS:

Lowther WL, Loneragan JF (1968) Calcium and nodulation in subterranean clover (Trifo-
lium subterraneum L.). Plant Physiol 43: 1362-1366
Lune van P, Goor van BJ (1975) A method of growing young fruiting apple trees in
water culture. J Hortic Sci 50: 129-133
Maher MJ (1976) The use of hydroponics for the production of greenhouse tomatoes
in Ireland. Proc 4th Int Congr Soilless Cult (IWOSC). Las Palmas
Malik MNA, Evenson JP, Edwards DG (1978) The effect of level of nitrogen nutrition
on earliness in upland cotton (Gossypium hirsutum L.). AustJ Agric Res 29: 1213-1221
Martin NE, Hendrix JW (1966) An apparatus for the mist culture of wheat. Plant Disease
Rep 50:369-371
Miyake Y, Takahashi E (1978) Silicon deficiency of tomato plant. Soil Sci Plant Nutr
24:175-189
Moritsugu M, Kawasaki T (1979) A new system of automatic pH regulation in solution
culture. Ber Ohara Inst Landw BioI 17: 171-178
Mulder EG (1948) Importance of molybdenum in the nitrogen metabolism of micro organ-
isms and higher plants. Plant Soil 1: 94-119
Munns DN (1968) Nodulation of Medicago sativa in solution culture. I. Acid sensitive
steps. Plant Soil 28:129-146
Norris DO (1964) Techniques used in work with Rhizobium. In: Some concepts and
methods in sub-tropical pasture research, Bull 47. Commonw Bur Pastures Field
Crops
Norris DO, Date RA (1976) Legume bacteriology. In: Shaw NH, Bryan WW (eds)
Tropical pasture research principles and methods, Bull 51. Commonw Bur Pastures
Field Crops
Owusu-Bennoah E, Mosse B (1978) In: Development of VA mycorrhiza (E3 and YU)
in plants fed with nutrient solution in sand and nutrient film culture. Rothamsted
experimental station annual report, Part 1
Pitman MG, Cram WJ (1977) Regulation of ion content in whole plants. In: Jennings
DH (ed) Integration of activity in the higher plant. Soc Exp Bioi Symp 31, Cambridge
Reisenauer HM (1960) Cobalt in nitrogen fixation by a legume. Nature 186:375-376
Reisenauer HM (1966) Mineral nutrients in soil solution. In: Altman PL, Dittmer DS
(eds) Environmental biology. Fed Am Soc Exp BioI Bethesda Md USA
Reisenauer HM (1969) A technique for growing plants at controlled levels of all nutrients.
Soil Sci 108:350-353
Resh HM (1976) A comparison of tomato yields, using several hydroponic methods.
Proc 4th Int Congr Soilless Cult (IWOSC). Las Palmas
Sabet SA, Abdel Salam MA, Lagerwerff JV (1964) Growth and ion uptake by maize
seedlings on solutions variable in phosphate and flow rate. Plant Soil 21:94-100
Sachs J (1860) Berichte liber die physiologische Tiitigkeit an der Versuchsstation in Tha-
randt. IV. Vegetationsversuche mit Ausschluss des Bodens liberdie Niihrstoffe und
sonstigen Erniihrungsbedingungen von Mais, Bohnen und anderen Pflanzen. Land-
wirtsch Vers Stn 2:219-268 (cited by Hewitt 1966)
Saussure de T (1804) Recherches chimiques sur la vegetation. Paris (cited by Hewitt
1966)
Schwarz K (1974) Recent dietary trace element research, exemplified by tin, fluorine
and silicon. Fed Proc 33:1748-1757
Sideris CP, Young HY, Krauss BH (1943) Effects of iron on the growth and ash constitu-
ents of Ananas comosus (L) Merr. Plant Physiol18: 608-632
Spear SN, Edwards DG, Asher CJ (1978) Effects of nutrient supply on critical nutrient
concentrations in cassava plants. In: Ferguson AR, Bieleski RL, Ferguson IB (eds)
Plant nutrition 1978. Proc 8th Int Colloq Plant Anal Fert Probl, Auckland, NZ.
NZ DSIR Inf Ser 134. Gov Printer, Wellington
Spear SN, Edwards DG, Asher CJ (1979) Response of cassava (Manihot esculenta Crantz)
to potassium concentration in solution: critical potassium concentrations in plants
grown with a constant or variable potassium supply. Field Crops Res 2: 153-168
Stiles W (1916) On the interpretation of the results of water culture experiments. Ann
Bot 30 :427-436
1.3 Modern Solution Culture Techniques 119

Temple-Smith MG, Menary RC (1977) Growth and phosphate absorption in lettuce

and cabbage plants in dilute solution culture. Aust J Plant Physiol4: 505-513
Tibbits TW, Palzkill DA, Frank HM (1978) Constructing a continuous circulation system
for plant solution culture. Dniv Wisconsin ColI Agric Life Sci, Res Bull R 2963
Trelease SF, Trelease HM (1935) Physiologically balanced culture solutions with stable
hydrogen ion concentration. Science 78: 438--439
Wallace A, Patel PM, Berry WL, Lunt OR (1978) Reclaimed sewage water: A hydroponic
growth medium for plants. Res Rec Conserv 3: 191-199
Wild A, Skarlou V, Clement CR, Snaydon RW (1974) Comparison of potassium uptake
by four plant species grown in sand and flowing culture solutions. J Appl Ecol
11 :801-812
Williams DE, Vlamis J (1957) Manganese and boron toxicities in standard culture solu-
tions. Soil Sci Soc Am Proc 21 : 205-209
Wilson DO, Reisenauer HM (1963) Cobalt requirement of symbiotically grown alfalfa.
Plant Soil 19 : 364-373
Woodward J (1699) Thoughts and experiments on vegetation. Philos Trans R Soc London
21:193 (cited by Hewitt 1966)
1.4 Diagnosis of Mineral Deficiencies Using
Plant Tests
D.BoUMA

1 Introduction

When agricuftural scientists in the last century began to realize that mineral
elements in a plant were taken up from the soil in which the plants grow,
it was a logical step to suggest that chemical analysis of plants could be used
as a means of assessing the nutrient supply of the soil. At the time it also
appeared reasonable to suggest, as von Liebig did in the last century in his
Law of Restitution (GOODALL and GREGORY 1947), that plant analysis could
be used to determine the quantities of nutrients removed from the soil by a
crop and, therefore, the amounts needed to maintain the supplying power of
the soil.
Much of the early work on chemical analysis of plants was prompted by
the desire to develop techniques that would supplement, or even replace, soil
analysis and thus provide a biological method of soil analysis. A review contain-
ing many references on the history of plant analysis has been presented by
GOODALL and GREGORY (1947). The idea of using plant analysis to determine
the nutrient requirements of the soil has dominated this area of plant nutrition
for many years. The previous edition of the Encyclopedia of Plant Physiology
contains a section of six pages on plant and leaf analysis as part of a chapter
on the determination of nutrient requirements of the soil (BERGMANN 1958).
However, in the past 20 years or so a great deal of experimental work has
caused a gradual but fundamental change in thinking, as it became progressively
apparent that in many situations, and for a range of agricultural and horti-
cultural plant species, the chemical composition of plants or plant tissue can
directly reflect the nutrient status or the nutrient requirement of plants them-
selves.
In recent years, probably as a result of advances in knowledge and under-
standing of the role and function of nutrient elements, new approaches to diag-
nosis are being developed which differ in principle from plant analytical tech-
niques. These are based on specific physiological or biochemical changes caused
by deficiencies or, alternatively, on specific responses that can be induced in
plants or plant tissue by the addition of a deficient element.
This chapter discusses the main features of plant analysis as a diagnostic
technique and then discusses some of the newer approaches.
1.4 Diagnosis of Mineral Deficiencies Using Plant Tests 121

2 Plant Analysis
2.1 Physiological Basis

Within the limits imposed by genetic potential, plant growth depends on envi-
ronmental factors including light, temperature and water supply, but also on
the supply of essential nutrient elements. If all environmental factors except
the supply of one essential element were optimal, then plant growth would
be a function of the supply of this element. Increased plant growth resulting
from increased supply of the element is usually accompanied by an increase
in uptake of that element into the plant. Depending on many factors, some
of which are discussed below, this increase may in tum be reflected in a changed
concentration of the element in the plant dry matter. In plant analysis, our
aim is to establish the relationship between nutrient concentration and growth
(or yield), and then use this relationship in comparable situations to establish
the nutrient status of a plant or crop. In this way, the nutrient requirement
may be assessed (BATES 1971, BOULD 1968, GOODALL and GREGORY 1947, SMITH
1962, ULRICH 1952).
A relationship between plant nutrient concentration and yield may be estab-
lished in pot experiments, in field experiments in which varied levels of one
or more nutrient elements are applied, or in surveys of commercial fields. An
example of the latter approach is the use of a relationship between root weight
and petiolar nitrate content of sugar beet to determine the nitrogen requirements
of the crop (ULRICH 1952).
The relationship is often curvilinear, but may also be linear or a combination
of the two. An idealized example of the relationship between yield and nutrient
supply is shown in Fig. 1 a, where the full line represents the yield response
to a nutrient element with all other elements present at optimal supplies, and
the broken line the response when another less deficient element limits the
full response. A relationship between yield and nutrient concentration in the
dry matter is shown in Fig. 1 b. Central to the diagnosis of nutrient deficiencies
by plant analysis is the concept of a "critical concentration" for each element
in the tissue. This concentration is the one separating the regions on the curve
of sufficiency or "luxury consumption", and deficiency or "poverty adjust-
ment" (arrow Fig. 1 b; MACY 1936). A similar definition is as follows: that
concentration which is just sufficient or just deficient for maximum growth
or yield. In practice it is not often possible to define the critical concentration
as sharply as implied above, and it is therefore often defined as that concentra-
tion where yield is 5% or 10% below the optimum (ULRICH and HILLS 1967).
Under conditions of extreme deficiency, small increases in yield that result
from the application of the deficient element may sometimes be accompanied
by a decrease in the concentration of the element in the dry matter (Fig. 1 b,
region I on the curve). PIPER (1942) first drew attention to this effect. It was
later described in detail by STEENBJERG (1951) for the Cu concentration in barley
grown in a pot experiment, and it is often referred to as the Piper-Steenbjerg
effect. Several explanations have been given for this effect: BATES (1971) sug-
122 D.BoUMA:
(a)

2
-----------

c
~
Curve 1: no other
co 'limiting factors'
a::
Curve 2: another factor
limits full response to
nutrient element

Supply of nutrient element

(b)
IV
Luxury consumption

~
c
co
[L

Critical concentration

1
Concentration of nutrient element in plant tissue

Fig. 1 a, b. Idealized curves representing the relationship between plant growth or yield
and nutrient supply without and with a "limiting factor" (a). b shows relationship be-
tween nutrient concentration and growth or yield. See text for explanation

gested variations in physiological age caused by the selection of tissue. In wheat

LONERAGAN (1978) has shown that Cu is readily lost from ageing leaves of
plants grown with adequate Cu but not in Cu-deficient plants. Deficiency of
Cu may delay the loss ofCu from leaves to such an extent that Cu concentrations
in the corresponding leaves of non-deficient plants may fall below them, thus
giving a marked Piper-Steen bjerg effect.
1.4 Diagnosis of Mineral Deficiencies Using Plant Tests 123

As the nutrient supply is further increased, there may be a region where

there is increased growth (dry matter) without a marked change in the concentra-
tion of the deficient element (region II, Fig. 1 b). With still higher levels of
supply, growth as well as concentration of the element increase (region III,
Fig. 1 b) until the critical concentration is reached, when further additions cease
to affect growth. However, in this range uptake of the element usually continues
to increase, resulting in a further increase in the concentration of the element
in the dry matter. This is the zone of "luxury consumption". Finally, with
still higher nutrient applications growth may decline due to toxicity effects,
while the concentration continues to increase (region V, Fig. 1b).
In most cases it is the total concentration of each element in the dry matter
which has been used for diagnosis. However, there are many cases where the
concentration of a specific fraction, either in fresh or in dried tissue, has been
been found more useful, e.g. nitrate-N, sulfate-S, soluble K, to mention a few
(FRENEY et al. 1978, SPENCER et al. 1977, ULRICH 1952). Occasionally, the ratio
of one element to another has been used. In subterranean clover the ratio
total N /total S has been suggested for the detection of S deficiency (SPENCER
et al. 1977). In wheat, the ratio S01- /total S was least affected by N supply
or plant age and was suggested as a suitable index of S status (FRENEY et al.
1978).

2.2 Choice of Tissue

Many different plant organs and tissues have been used for diagnosis (see
GOODALL and GREGORY 1947 for a table of these). The tissues include roots,
stems, bark, leaves of different age and position, laminae, petioles, midribs,
seed, fruit and grain. In general, changes in nutrient concentration, caused by
varying the nutrient supply, are greater in leaves than in other organs (GOODALL
and GREGORY 1947). Although the entire leaf is usually sampled, the concentra-
tion in the petiole dry matter may occasionally be very different to that of
the laminae, so that the inclusion of variable lengths of petiole will give variable
results (BOULD 1961). Occasionally the petiole will give a better indication of
the nutrient status than the lamina - this, however, may be true of one element
but not of another. Thus in grapes the petioles gave a better index of K status
than the laminae (SHAULIS 1961). This was also found for the K status of
raspberry, while for N the reverse was the case (BOULD 1964). In sugar beet,
nitrate-N in the petioles provided the best indication of the N status while
sulfate-S in the laminae was preferred for the S status (ULRICH 1961).
In choosing an organ for assay, several criteria may be used (EMMERT 1959),
but generally the two most important ones are the sensitivity of the response
and its stability towards factors other than the supply of the particular nutrient
in question. Change in the supply of nutrients can markedly affect plant mor-
phology, thereby altering the proportion of dry matter distributed to different
organs. It is important that this does not distort the correlation between nutrient
content and yield or growth (GOODALL and GREGORY 1947). It will become
apparent from the following pages that there are no general recommendations
and that each crop, element and situation requires ifs own extensive study in
order to develop a reliable diagnostic system based on plant analysis.
124 D. BOUMA:

2.3 Factors Affecting the Relationship Between Nutrient Concentration

and Yield

2.3.1 Plant Development

Within a given environment, the nutrient uptake by a plant depends on the
supply to the roots and on the demands for the nutrient set up within the
plant by the growth and normal functions of plant organs, including the roots
(WILLIAMS 1955). From his studies of the growth of gramineous plants WILLIAMS
(1948) concluded that each vegetative organ of a plant passes through phases
of intake, a relatively constant content, and then export of many of its mineral
nutrients. Each organ therefore has a certain capacity to accumulate nutrients,
and may subsequently during senescence constitute a potential source of nu-
trients for younger plant parts. The demand of each organ for nutrients is
usually met in part, if not entirely, by uptake from the medium, but the rate
of uptake depends to a certain extent on the availability of the nutrient element
within the plant. The importance of this principle was clearly illustrated for
P, which is a mobile element, in an experiment with oats. Plants were grown
at three levels of P supply. The plants grown with the lowest level (deficient)
derived only 30% of their inflorescence P from other plant parts whereas those
grown at the highest (=non-deficient) level derived no less than 93% from
these sources. Thus in the latter case, though the external supply was plentiful,
an abundant and more accessible supply was made available by senescent break-
down in leaves and roots, and later in the stems. For the P-deficient plants,
remobilization was less significant, and P was preferentially obtained from the
deficient medium. It should be pointed out that the pattern of redistribution
of P in the non-deficient plants was caused by the demand of the inflorescences
for N rather than for P. By contrast, the P-deficient plants had a much higher
N status and any senescent breakdown in the leaves was probably initiated
by a demand for P rather than for N (WILLIAMS 1948).
Substantial retranslocation, even over short periods, has also been shown
for K (GREENWAY and PITMAN 1965). In young barley plants retranslocation
was most evident in the mature leaf, which showed little K-uptake. Rapid rises
in K concentration were evident in the developing youngest leaf in which about
50% of the K taken up during the 3-day experimental period had been derived
from within the plant.
When one considers the effects of plant deVelopment on the mineral composi-
tion of plant organs, a distinction should be made between annual and perennial
crops. In annuals, with their shorter period of growth and often with a more
rapid development, changes in nutrient content tend to be more rapid and
extensive. With them, it is generally difficult to use plant analysis to remedy
current nutritional disorders, unless the diagnosis is made early enough in the
life of the plant. Sometimes diagnosis of annual crops can help to regulate
the quality of the harvested product. An example is the use of the nitrate content
of petioles of sugar beet to monitor the N nutrition of the plant and through
it to modify the yield of sugar by changing the supply of N, thereby altering
the balance between top and root growth (ULRICH and HILLS 1967).
1.4 Diagnosis of Mineral Deficiencies Using Plant Tests 125

GREGORY (1953) made the point, worth repeating in the present context,
that in a developing cereal plant over 90% of the nitrogen and phosphorus
can be taken up during the time when dry matter is attaining only 25% of
its final value. Subsequent development, which is accompanied by marked pat-
terns of retranslocation, depends on this store of accumulated nutrients and
determines final yields. Therefore, in the early stages of growth of annual plants
the tissue concentration of nutrient elements tends to be high. As the plant
develops, nutrient uptake continues, but for some elements, particularly mobile
ones like Nand P, the concentration in the plant tops or in the whole plant
as a percentage of dry matter usually declines. The process is usually referred
to as dilution and is partly due to an increase in the proportion of structural
and storage materials relative to protoplasm (ULRICH 1952). Other factors can
be a decline in the nutrient supply to the roots, in an absolute sense or relative
to the rapidly increasing demands for growth. In the case of mobile elements
there is also the redistribution between organs initiated by internal competition
for nutrients or by senescence, even at adequate levels of external supply, re-
ferred to earlier.
Not all elements are mobile in the plant: for example, Ca is not (MILLIKAN
and HANGER 1964). Some of the complications that may arise in the diagnosis
of a deficiency of an immobile element are clearly illustrated in the work of
LONERAGAN and SNOWBALL (1969), who studied the Ca nutrition of a number
of grasses, herbs and legumes. Plants transferred from solutions containing
luxury Ca levels to Ca-deficient solutions developed Ca deficiency symptoms,
even though the plant tops had Ca concentrations three to ten times higher
than the tops of healthy plants grown in a low but constant Ca supply. Even
at a low and constant supply, older leaves may accumulate sufficient Ca to
mask a deficiency in younger parts if analysis of whole plant tops is used as
a criterion of Ca requirement. These considerations apply to any element with
a restricted mobility in the plant. The difficulties in diagnosing deficiencies
of these elements can be lessened by avoidance of old leaves and a careful
selection of young tissues, e.g. leaves or even roots (LONERAGAN and SNOWBALL
1969).
In tree crops, differences in age are relatively unimportant, judging by the
similarities in the nutrient concentration of comparable tissues from trees of
different ages (SMITH 1962). In citrus, which can have two or more growth
flushes in one season, the composition of leaves sampled from different flushes
at comparable stages of development varied little, although the summer flush
reached a relatively stable nutrient concentration sooner than the spring flush
leaves (SMITH and REUTHER 1950).

2.3.2 Effects of Changes in Age of Tissue

The composition of each organ or tissue changes during its development under
the influence of the same factors discussed before for the whole plant. Each
element often shows a characteristic pattern of change in a given tissue as
it develops, matures and finally senesces. Data that have been tabulated by
SMITH (1962) indicate that concentrations of N, P, and K in the dry matter
126 D. BOUMA:

always declined with age, while others, like Ca and Mg, usually increased.
These differences probably reflect differences in mobility, as discussed before.
Such developmental changes with respect to leaves have been extensively studied
and recorded for many crops, particularly horticultural ones, e.g. blackcurrants
(BouLD 1961), peaches (EpSTEIN and LILLELAND 1942), oranges (JONES and
PARKER 1950, SMITH and REUTHER 1950), and apples (REUTHER and BOYNTON
1939). EMMERT (1959) concluded from a survey of the literature that a knowledge
of such trends may help in selecting the best sampling time for a tissue analysis
test. In general, there are three more or less distinct growth periods. In the
first period, when leaves are rapidly expanding, changes in concentration and
day-to-day variations may be relatively large. Significant changes can also occur
towards the end of the growing season, when senescence is often accompanied
by retranslocation of mobile elements to woody tissues. The middle period
is usually one of relative stability and may last from 3 to 6 months. Most
of the diagnostic standards for fruit trees have been developed for leaves of
this age (SMITH 1962). In some cases it is possible to reduce variation due
to differences in leaf age by sampling leaves of the same physiological age,
for example the youngest fully mature or expanded leaf. This is particularly
useful in comparisons of nutritional treatments or when samplings during the
growing season have to be done.
The selection criteria are usually governed by considerations of variability,
reproducibility and the obvious need to obtain a reliable indication of the nu-
trient status of the plant, crop or tree. Requirements often conflict and a com-
promise is often necessary. Thus MASON (1958) concluded that for apples, the
most suitable leaf position and sampling time were those for which variations
in leaf composition were least. The best tissues are not necessarily those which
show the biggest differences in composition. In addition, optimum tissues and
conditions vary among elements and need to be determined separately.

2.3.3 Plant Age and Critical Levels

Although critical concentrations generally depend on the age of the selected
tissue used for establishing the relationship between yield and nutrient concen-
tration, the age of the plant itself is not always important. As pointed out
before, the difficulty is one primarily confined to annuals and possibly biennial
plants. ULRICH (1952, 1961) established a critical value of 1,000 ppm nitrate-N
in the dry matter of petioles of the youngest mature leaves of sugar beets.
This value did not vary during the growing season. Similar results were obtained
for P in sugar beets (ULRICH 1961) and in Ladino clover (ULRICH 1948). SPENCER
et al. (1977), in pot experiments on S requirements of subterranean clover, found
a decrease in critical total S concentrations in young leaves from 0.19% 33
days from sowing to 0.15% and 0.10% for responses 61 and 133 days from
sowing respectively. Similar results were obtained for sulfate-So GREENWOOD
(1966) found a very marked decrease in total N, soluble N and free ninhydrin-N
content of the youngest fully expanded leaves of young wheat plants, sampled
at 3 and 5 weeks from emergence, particularly in the "critical" range of the
yield response curve to nitrogen. In a review, BATES (1971) concluded that
1.4 Diagnosis of Mineral Deficiencies Using Plant Tests 127

the critical concentration in the whole plant changes with age, but that it is
an open question whether a tissue selected to maintain a common physiological
age at different sampling times does have a changing critical concentration.
This again emphasizes the need for careful evaluation of the effects of tissue
or plant age in the nutritional relationships of diagnosis.

2.3.4 Interactions Between Nutrient Elements

The statement was made in the beginning of this section that plant growth
is a function of that nutrient present in suboptimal amounts, if other nutrient
elements and environmental conditions are non-limiting. This is often referred
to as the Law of the Minimum, first proposed by von Liebig (GOODALL and
GREGORY 1947). Although true in principle, in practice there are few occasions
where nutrients other than the limiting one are present at optimal levels. Nu-
trient interactions, involving a change in the concentration of one element in
plant tissue caused by another, are quite common. They are often difficult
to unravel, but need to be taken into account in diagnosis based on plant
analysis. A shift in the content of one element invariably is accompanied by
secondary changes in tissue content of other elements, even though there is
no change in the availability of the interacting nutrient(s) (EMMERT 1961). Nu-
trient interactions may be positive or negative (PRilVOT and OLLAGNIER 1961).
Alternative expressions for the same phenomena are synergism and antagonism
(SMITH 1962). Antagonism can occur during ion uptake, during translocation
and accumulation in the tissues, or in metabolism [e.g. in enzyme systems,
Fe versus Mn (PREVOT and OLLAGNIER 1961)]. It can involve competition be-
tween two or more elements, but also precipitation of nutrients or other phenom-
ena.
Antagonisms during uptake may occur between cations but also between
anions. Well-known examples of cation antagonism are those between K and
Ca, K and Mg, Fe and Mn, to mention only a few (BOULD 1964, SMITH 1962).
N supplied as the ammonium ion is antagonistic to other cations, but when
supplied as nitrate it may compete with other anions, e.g. phosphate. Increasing
the level of K to fruit trees has been shown to induce Mg deficiencies, accompa-
nied by increases in the K concentration of the leaves (REUTHER et al. 1958).
Conversely, changes in Mg concentration ofleaf tissue are usually accompanied
by opposite changes in leaf K (EMMERT 1961). Antagonistic effects have also
been extensively documented for micronutrients. High levels of Cu in the soil
can induce Fe deficiency. Pot cultures have shown that high levels of Cu, Zn
or Mn can induce marked antagonistic effects between pairs of these ions on
concentrations in leaves (REUTHER et al. 1958).
Antagonism during translocation is often caused by precipitation in the
root tissues or elsewhere in the plant. Excess phosphate in the root environment
causes precipitation of Zn and also of Fe along the veins, and this may lead
to iron chlorosis or to "mottle leaf" as an expression of Zn deficiency (WEST
1938). In both cases a precipitation of the trace elements as phosphates appears
to be involved. Another case of induced Zn deficiency was reported for subterra-
nean clover growing in a soil low in Zn. An increase in N supply, regardless
128 D. BOUMA:

of N source, decreased the Zn content of the tops. This was attributed to

a retention of Zn in the roots as a result of the formation of immobile Zn-protein
complexes (OZANNE 1955). In water cultures, the critical Zn concentration of
a number of subterranean clover cultivars showed a wide range of values,
strongly dependent on the age of the plants and the phosphate supply. Plants
in some treatments showed Zn deficiency symptoms, but had a higher Zn con-
centration than in others without these symptoms, again depending on phos-
phate supply. The severity of the Zn deficiency was related to the PjZn ratio
and not to the Zn concentration in the tissues (MILLIKAN 1963).
There is another type of interaction referred to as pseudo-antagonism by
LUNDEGARDH(1966) or induced deficiency (EMMERT 1961), which does not
involve direct ionic competion, and which can have various causes. A common
interaction of this kind exists when two elements are deficient but different
in severity, so that the deficiency of one element is masked by the more deficient
element. When the supply of the latter is increased, a stage is reached where
the first element becomes more deficient and then begins to determine plant
response. This principle is of considerable importance in diagnosis based on
plant analysis and is often referred to as the Law of the Minimum (see also
Fig. 1 A). The interaction between Nand P is a well-known example (BOUMA
1956, 1959a, b, 1961, WILLIAMS 1955). In field experiments designed to develop
criteria for the detection of S deficiency in subterranean clover pastures, BOUMA
et al. (1969) found that correlations between pasture yield and total S or sulfate
concentration in clover tops were much higher when plots had received phos-
phate than without applied phosphate. Similarly, correlations between yield
responses to phosphate applications and the total P concentration in clover
tops were higher when S deficiencies had been corrected. The implication of
these examples is, as CHAPMAN (1966) and others have pointed out, that plant
analysis can only detect a deficiency of one nutrient at a time. It also appears
that, in general, analysis of plant tissue for a single element is of limited value
in diagnosis, unless it is known that the supply of other elements is adequate.
An example of an integrated approach to diagnosis is the one developed by
BEAUFILS, referred to as Diagnosis and Recommendation Integrated System
- DRIS (SUMNER 1974). This system attempts to take into account as many
factors affecting yield as can be expressed quantitatively, including plant compo-
sition with respect to as many elements as are likely to affect crop performance.
Composition is expressed in as many ways as possible, e.g. concentration (%)
in the dry matter, or as ratios or products, in order to obtain the most appro-
priate comparison between different yield populations. By selecting the appro-
priate sets of expressions all elements can be related in the form of a statement
listing them in decreasing order of requirement by the plant.

2.3.5 Environmental Factors

Maximum rates of plant growth depend on environmental factors such as tem-

perature, light and moisture supply. When present at levels below the optimum
for maximum growth, or yield, any of these factors may become "limiting"
1.4 Diagnosis of Mineral Deficiencies Using Plant Tests 129

and thus cause a reduced nutrient requirement by the plant. As a result the
nutrient concentration in the dry matter tends to increase. In subterranean
clover plants grown at three temperatures (15°C day/10 °C night, 21°/16 DC,
27° /22 0C), and at two P levels the response in growth to P was least at the
lowest temperature. The P concentration in all plant parts (leaves, petioles and
roots) was highest at the lowest temperature and decreased with rising tempera-
ture. This occurred at both P levels (BOUMA and DOWLING 1969a, b). Similar
results were obtained for S (BOUMA unpublished).
There is potentially a limit where further decreases in temperature will not
cause a further rise in nutrient concentration and may even cause a decrease.
This will depend on the relative extent to which growth and nutrient uptake
are reduced by lowering the temperature. ZURBUCKI (1961), for example, showed
that the N, P and K concentrations of tomato plants grown at 12°C were
lower than of plants grown at 20 DC. At 50 days from sowing there was a
fourfold difference in dry weight. In a case like this, where the temperature
of 12°C was much lower than the optimum, nutrient uptake may be reduced
to an even greater extent than growth, resulting in a decrease in nutrient concen-
tration.
Light influences nutrient concentrations in a similar manner. In an experi-
ment (BOUMA unpublished) with subterranean clover plants grown at 4 P levels,
concentrations of P in all plant parts (leaves, petioles and roots) were higher
under a light intensity of 1300 ft.c. than under 2600 ft.c. These differences
were relatively small at the lower P levels, but increased with the P supply
due to the greatly reduced growth response to P at the low light level compared
with that at the high level. This suggests a reduced P requirement at lower
light levels. In cherry trees critical values for N were higher in sun leaves than
in shade leaves (PROEBSTING and KENWORTHY 1954).
The effects of changes in soil moisture on nutrient concentrations in plant
tissues are complex. Some obvious effects of excessive moisture are a loss of
nutrients by leaching, erosion, and a decreased availability for some elements
due to poor aeration (WADLEIGH and RICHARDS 1951). Some elements may
become more available due to reduction processes in a poorly aerated soil,
e.g. Fe and Mn. Uptake of nutrients, particularly anions, may also be reduced
by a lack of oxygen in saturated soils. Changes in aeration may also affect
root morphology. In general, all these factors can influence the nutrient concen-
tration in plant tissues, depending on the extent to which they affect nutrient
uptake, utilization and growth.
Reports on the effects of low soil moisture supplies on nutrient concentra-
tions in the plant are conflicting. WADLEIGH and RICHARDS (1951) concluded
that, for a given level of fertility, decreasing soil moisture causes an increase
in N concentration in plant tissue, a decrease in K concentration, and a variable
effect on P, Ca and Mg contents. On the other hand, WILLIAMS and SHAPTER
(1955) concluded that in barley and rye, moisture stress gave increased leaf
concentrations of K, N, Mg, Ca and Mn, although treatment effects on rye
were relatively small initially.
It has often been suggested that drying of the soil causes an increase in
concentration in the soil solution of at least some nutrient elements, and that
130 D. BOUMA:

a resulting increase in uptake by the plant may cause a corresponding increase

in the tissues (WADLEIGH and RICHARDS 1951). However, GATES (1957, 1974)
concluded that there was no evidence for this in his experiments, and that
wilting both restricted growth anq gave lower values for concentrations of a
wide range of nutrient elements, including the principal determinants of protein
synthesis, Nand P. Moisture stress was considered to cause hydrolysis of organic
compounds in older tissues, resulting in an increase in levels of soluble nutrients
in the plant, so that the plant reduced uptake of these elements in response
to a decline in internal demand, rather than because of changes in nutrient
availability (GATES 1974). Upon rewatering, a complete reversal of these trends
was found, in which the main factor controlling the absorption was suggested
to be the internal demand caused by the renewal of synthetic processes (GATES
1957).

2.3.6 Other Factors Affecting Nutrient Composition

The nutrient concentration of plant tissue is strongly influenced by the fruit
or crop load (EMMERT 1959, SMITH 1962). In general, the concentrations of
N, Ca and Mg increase with the fruit load, while the K concentration usually
decreases. The effects on P are variable. Leaf scorch caused by K deficiency
has been reported to increase with the crop load by a number of workers.
There are also marked interactions between the effects of fertilizer additions
on crop load and the concentrations of other elements in leaf tissue. BOUMA
(1959a) found a marked increase in crop load of navel oranges due to N applica-
tions, which was accompanied by a decrease in K concentration of the leaves,
even with adequate K supplies in the soil (GROENEWEGEN and BOUMA 1960).
This was attributed, at least in part, to a dilution effect following yield and
growth responses to N. SATO (1961) considered a decline in leafK with increas-
ing fruit load in Satsuma oranges to be largely due to translocation of K to
the fruit, possibly because the fruit had a relatively high K requirement. Related
to this may be the finding in Florida that there was an almost linear increase
in fruit drop when leaf K was less than 0.8% of the dry matter (WILSON 1961).
The nutrient composition of the leaves also depends on their position relative
to the fruit. In oranges, for example, Nand P concentrations are, as a rule,
lower in leaves from fruiting shoots than in leaves from vegetative shoots, prob-
ably because the fruit competes for available nutrient elements (BOUMA 1959b,
1961). For this reason leaf sampling is usually standardized, some workers pre-
ferring leaves from fruiting shoots and others those from vegetative shoots
(see CHAPMAN 1966 for a tabulation of tissues sampled).
Other factors which can affect the nutrient composition of plant tissues
are pests and diseases, and these need to be taken into account to av~id mislead-
ing interpretations of tissue composition (BERGMANN et al. 1974, KIRKPATRICK
et al. 1964). Genetic factors can be important (see Chap. IV.1, this Vol.). A
special case is the horticultural one where one genotype (scion) is grafted on
to another (stock) and where the nutrient-absorbing and -utilizing parts of
the plant are different in origin and character (EpSTEIN 1972, KENWORTHY 1967,
REUTHER and SMITH 1954).
1.4 Diagnosis of Mineral Deficiencies Using Plant Tests 131

3 Physiological and Biochemical Approaches to Diagnosis

3.1 Introductory Remarks

Significant and impressive though the past progress in the field of plant analysis
has been, there are indications that research along traditional lines in this area
and its application in diagnosis have reached a point of diminishing returns.
BAR-AKIVA (1971) expressed the view that many of the difficulties of plant
analysis stem from its empirical nature. GREENWOOD (1976) concluded that
nutrient response curves based on plant analysis as outlined above give very
little information about the current intensity of deficiency. Because of the many
factors that can affect response curves, they may also be quite misleading.
It is not surprising, therefore, that in recent years there has been a search
for other approaches based on specific metabolic, enzymatic or physiological
changes caused by deficiencies, or on responses that can be induced in plants
or in plant tissue by the addition of a deficient element. They can be broadly
grouped, for ease of presentation, as physiological or biochemical, although
the division is an arbitrary one in some cases.

3.2 Physiological Approaches

3.2.1 Physiological Assessment

In a series of papers BOUMA and DOWLING (1962, 1966a, b, 1967) and BOUMA
et aI., (1969) presented a diagnostic method, to which they referred as "the
physiological assessment of the nutrient status of plants". It is based on the
differential leaf area responses induced by transfer of deficient young subterrane-
an clover plants to complete nutrient solutions and to other solutions each
without a different element. Leaf areas of plants were measured by comparing
individual leaves with a set of photographic standards. This process is quick
and simple. Responses in leaf area were relatively greater than those in dry
weight, thus enabling an earlier identification of induced responses. For
example, P-deficient plants transferred to solutions without P showed smaller
increases in leaf area than corresponding plants transferred to complete solutions
or to solutions lacking an element other than P. When plants not deficient
in P were transferred, there were no significant differences in leaf area increase
between any of the solutions. In this way it was possible to make a clear distinc-
tion between P-deficient and non-deficient plants. Similar results were obtained
with plants raised at different levels of S, K or B. With an unknown deficiency,
involving anyone of these elements, therefore, the deficient element can be
identified by the element omitted from that solution in which the transferred
plants showed the smallest leaf area increases during the test period. In most
cases the relative differences in leaf area between plants in complete solutions
and in solutions without the deficient element increased with the severity of
the deficiency. This was therefore suggested as a measure of the degree of
the deficiency in the field.
132 D. BOUMA:

Obviously the approach only lends itself to those plants which can be trans-
ferred to test solutions and which can subsequently continue or resume growth.
The approach has been further developed by JANSSEN (1974) and by MULLER
(1974), and used to advantage in several countries for a number of crops, includ-
ing wheat and maize, and also for perennials such as Pinus caribea and cacao.
The emphasis has been somewhat different in that the authors were primarily
assessing the likelihood of finding nutrient deficiencies in soils to be used for
agricultural development and testing the suitability of those soils for certain
crops. They therefore grew seedlings in small pots containing the soil concerned.
These pots had a gauze bottom so that the roots could extend into nutrient
solutions on which the pots were placed. These solutions induded, as above,
a complete nutrient solution and others each without a different element. A
deficiency was identified by a difference in growth, using a suitable parameter
such as leaf area or leaf length, between plants on a solution without a certain
element and on the other solutions. JANSSEN (1974) used a "sufficiency quotient"
as an index of nutrient supply in the soil, this being defined as the ratio of
the relative growth rate of plants in pots on the deficient solution to that on
the complete solution.

3.2.2 Nutrient Stress

In a review summarizing the development of an approach having features in
common with the one discussed above, GREENWOOD (1976) suggested the
concept of "nutrient stress" to quantify the current nutrient status of plants.
He defined this as the proportion by which the growth rate of the plant or
crop is limited by a particular nutrient element under the prevailing conditions.
GREENWOOD et al. (1965) developed this concept to estimate N deficiency in
a grass sward, and expressed N stress (SN) as the percentage reduction in the
relative growth rate of deficient plants compared with that of plants growing
with non-limiting N. To measure SN they used a split-plot technique, applying
sufficient N for maximum growth to one half and leaving the other half un-
treated. Dry matter on these plots was measured at intervals of 18-14 days.
Since there are some obvious disadvantages in the use of dry matter as an
index of growth, particularly under field conditions, they also investigated the
concentration of total and ninhydrin-N in the youngest fully expanded leaf
of the tillers of annual ryegrass and wheat, and also leaf elongation as para-
meters of N status that could be used to determine SN. SN could be estimated
reliably by all three indices over the whole range of experimental treatments
employed, but leaf elongation was found to be the best. In wheat and ryegrass,
plant symptoms of N deficiency appeared when SN was greater than 40%
(GREENWOOD 1966, GREENWOOD and TITMANIS 1966).

3.2.3 Approaches Based on Photosynthesis

In a further development of physiological assessment, BOUMA aimed at inducing
responses in detached leaves rather than in intact plants. This would have some
obvious advantages, one being the possible simplification of any routine diag-
1.4 Diagnosis of Mineral Deficiencies Using Plant Tests 133
4

------~
co
3
/~
.J::

/~'
"-
'"c
Q)

c
g
2
~
:::
'"
E
c:-
o
•
P, Pasture response in the field

o O~I--~~--4~O-0---J----8~O-0--~---1-2~OO----~--1~6bo
Superphosphate (kg/ha)

2.1 Leaf response after pretreatment 2.1

_ _ _ with phosphate
_____ without phosphate

2.0
•
2.0
c;

,
c;
E
1
E
~
Q;
:::
•
Q)
:::
'"
E 1.9
I
19 '"E
>- I
,:; I
S! I
Cl
I
.2 I
I

•
1.8 I 1.8
'I

•
P, P2 P3
1.7 O~----~ I I I 1.7
0 7 0 o
Days under light

Fig. 2. Effect of phosphate supply on yield of a subterranean clover-dominant pasture

(top). Bottom graph shows differences in dry matter production by leaves sampled in
the four phosphate levels when placed under fluorescent lights after a brief pretreatment
with and without inorganic phosphate (8 h; 6 mM). Significant differences (p=0.05 and
0.01) shown by vertical lines on day 7. (BOUMA and DOWLING 1980)

nostic application. Use was made of a previous finding that P deficiency causes
a marked reduction in photosynthesis of subterranean clover plants, and that
a transfer of plants to complete nutrient solutions is followed by a return to
normal levels of photosynthesis within 24-48 h (BOUMA 1967). A similar re-
sponse was found in leaves detached from P-deficient plants by placing them
134 D. BOUMA:

for about 8 h in a 6 mM inorganic phosphate solution (approximately 40 times

the concentration required for intact plants). The leaves were then returned
to water and photosynthesis was measured the following day (BOUMA 1975).
Subsequently it was shown (BOUMA and DOWLING 1976) that a corresponding
response in dry matter increase could be obtained when leaves were placed
under fluorescent lights for a period of 5-6 days after the initial phosphate
uptake period. By comparing dry matter increases of leaves treated with phos-
phate after detachment with those of comparable untreated leaves, a simple
and direct method for the detection of P deficiell'CY is being developed. In a
variety of pot experiments (BOUMA and DOWLING 1976) and field studies (BOUMA
unpublished), it has been found that dry matter differences between treated
and untreated leaves depend on the P status of the plants from which the
leaves were detached. These differences were not significant in the case of plants
of an adequate P status, but increased with a decline in the P status of the
plant. This is illustrated in Fig. 2, which shows the response of a subterranean
clover dominant pasture to superphosphate (top portion). Dry matter differences
induced in clover leaves as outlined above were closely related to the phosphate
status of the pasture plots from which the leaves were sampled (bottom portion).
The measurement of photosynthesis to determine the critical concentrations
of K and Mg of corn plants was suggested by PEASLEE and Moss (1966). They
plotted net rates of photosynthesis against extractable leaf K or Mg and ob-
tained the usual response curves for both elements, with critical concentrations
close to those found by other workers for field-grown plants. They suggest
that this approach is likely to give a more accurate and more easily obtained
estimate of the critical concentration than the usual correlative approach. Photo-
synthesis reflects the instantaneous performance of the leaf or plant that can
be related to its concomitant nutrient concentration. However, not all nutrient
elements affect photosynthesis in the same manner. There may even be differ-
ences in the effect of an individual element between plant species. BOUMA et al.
(1979), for example, investigated the effects of potassium and magnesium on
net photosynthesis of intact subterranean clover plants and found that K had
little effect on net photosynthesis in early stages of a deficiency, while Mg
deficiency caused an immediate decline. This is somewhat at variance with the
results of PEASLEE and Moss (1966), who found an early decline in photosynthesis
with a K deficiency while Mg caused a decrease at a more advanced stage
of deficiency, as judged by plant size. Nevertheless, their suggestion remains
valid as a principle, but would require investigation in each case.

3.2.4 Other Approaches

RANDALL (1969a) used detached leaves for the estimation of the S status of
subterranean clover. He incubated young leaves from plants grown at five S
levels, in solutions with and without sulfate. Over a period of 4 days alcohol-
insoluble N of leaves in solutions without sulfate declined, irrespective of S
status. The presence of sulfate in the treating solution arrested or reversed
the decline in leaves from S-deficient plants, but had no such effect on leaves
from non-deficient plants.
1.4 Diagnosis of Mineral Deficiencies Using Plant Tests 135

GREENHAM et al. (1972) investigated the possible use of electrical impedance

methods in the detection of Ca and P deficiency in subterranean clover. This
is based on the fact that some deficiencies can cause injuries of plant tissue.
They impaled petioles with electrodes and studied the changes in capacitive
impedance of an alternating current bridge. They were able to establish some
clear and consistent differences in impedance characteristics between deficient
and non-deficient plants.

3.3 Biochemical Approaches

3.3.1 Nitrogen and Molybdenum

Mo is a constituent of nitrate reductase, an adaptive enzyme which catalyzes
the reduction of nitrate to nitrite (EVANS and NASON 1953, NASON and EVANS
1953, NICHOLAS and NASON 1955). The activity of the enzyme can be measured
colorimetrically from nitrite formed after incubation of leaf or root tissue with
a buffer containing nitrate. Very low nitrate reductase activities are found in
Mo-deficient plants. When Mo is added to the deficient leaf tissue, induction
of the enzyme is rapid, maximum activity values occurring within 4 h (MULDER
et al. 1959). N-deficient plants also show a low riitrate reductase activity. In
this case, addition of nitrate to the rooting medium or to detached leaves causes
a rapid rise in activity. It is not surprising, therefore, that there are several
reports on the use of changes in nitrate reductase activity for the diagnosis
of Nand Mo deficiencies.
BAR-AKIVA and STERNBAUM (1965) grew young grapefruit trees in soil at
five levels of ammonium sulfate and found that nitrate reductase activity in
newly detached leaves, incubated for 2 h with nitrate ("initial activity"), in-
creased approximately 4.5 times from the lowest to the highest level ofN supply.
They also measured the enzyme activity of leaf samples incubated with nitrate
for 24 h prior to the enzyme assay, calling it inducible activity. This treatment
caused considerable increases in activity of the deficient leaves, but not of the
N-sufficient leaves. As a result, the gap between initial and inducible activity
increased from nearly zero at the highest level to quite high values at the lowest
level. The procedure was therefore suggested as a useful tool to assess the
N status of citrus trees.
Citrus usually does not accumulate nitrate. In subsequent work BAR-AKIVA
et al. (1970) investigated this approach for annual ryegrass, which can accumu-
late considerable amounts of nitrate. They found that, in spite of high endoge-
nous nitrate levels in normal tissues, activity could be induced further, and
that the response was negatively correlated with the N supply to the plant.
They calculated the ratio of induced to endogenous enzyme activity, and sug-
gested that a value of 1.5 or higher indicated N deficiency. This was similar
to the value found in citrus. The results of the ryegrass work are shown in
Fig. 3.
WITT and JUNGK (1974), however, found this ratio unsatisfactory for cauli-
flower and poinsettia plants without modifying the approach. Instead they made
use of the fact that additional nitrate reductase activity can be induced by
incubating leaf tissue with nitrate in the light as compared with the dark. This
136 D. BOUMA:

10

Ct1
.r::;
';::
Q)
:::Ct1
«0:" E
,..
z 5
...... ~
"0
c:
B

~0------~30~0------~60~0======::90~0======~12~OJO 2
kg N/ha

Fig. 3. Ratio of inducible/endogenous nitrate reductase and yield as affected by nitrogen

supply. (BAR-AKIVA et al. 1970)

"additional" nitrate reductase activity was negatively correlated with nitrate

supply to the plants. It was suggested that this relationship provided a good
indication of the N status of the plant.
The diagnosis of Mo deficiency on the basis of nitrate reductase activity
has been suggested by several workers. SHAKED and BAR-AKIVA (1967) found
a direct relationship between activity and Mo supply to sweet lime seedlings
growing in solution cultures. They also compared enzyme activity before and
after treatment of leaf tissue with Mo. There was no response by leaves from
plants grown at the highest Mo level. The differences in activity between leaves
incubated with and without Mo increased as the Mo supply to the plants de-
creased, and reached quite a high value for the plants grown without Mo.
RANDALL (1969b) found that light stimulated the induction of nitrate reduc-
tase in Mo-treated leaf tissue from Mo-deficient wheat plants. To assess the
Mo status of plants he incubated leaf fragments with buffered KN0 3 with
or without Mo for 4 h in the light to induce the reductase, followed by 1 h
in darkness to allow accumulation of nitrite, and permit in vivo assay of the
enzyme. There was a good quantitative relationship between the magnitude
of the response in enzyme activity of the incubated leaf tissue to Mo and the
yield responses to Mo of the plants from which the leaf fragments came.

3.3.2 Phosphorus
In tissues of many plant species the activity of acid phosphatase is enhanced
markedly by phosphate deficiency (HEWITT and TATHAM 1960, BESFORD 1978b,
1979a, b, c). The enzyme catalyzes the hydrolysis of phosphate esters. Assay
rests on the hydrolysis of p-nitrophenyl phosphate by tissue extracts, the yellow
colour of the reaction product, p-nitrophenol, being proportional to enzyme
activity. BESFORD has shown that in tomatoes (1979 a) increased enzyme activity
was specific for P deficiency. In this and further work with seven other plant
1.4 Diagnosis of Mineral Deficiencies Using Plant Tests 137

species (BESFORD 1979b) enzyme assay provided a rapid and sensitive index
of the plant's P status. After subsequent experiments BESFORD (1980) concluded
that visual comparison of the yellow colour formed during incubation of tomato
leaf discs may eliminate the need for laboratory facilities in the diagnosis of
P deficiency.

3.3.3 Potassium and Magnesium

In the glycolytic pathway the enzyme pyruvate kinase (PK) catalyses the transfer
of phosphate from phosphoenolpyruvate (PEP) to adenosine diphosphate
(ADP), yielding adenosine triphosphate (ATP) and pyruvate. The enzyme re-
quires K as well as Mg ions (BONNER and VARNER 1965) although BESFORD
and MAW (1975) were unable to demonstrate an absolute requirement for Mg
by PK in tomato plants. In enzyme preparations there is also present a compli-
cating reaction caused by phosphatase which hydrolyzes PEP in the absence
of ADP. The latter enzyme is non-selectively inhibited by monovalent cations
but stimulated by Mg. Activity of PK can be found by measuring the amount
of pyruvate formed in presence and absence of ADP, then subtracting pyruvate
found in the absence of ADP (=phosphatase activity) from the activity pro-
duced in the presence of ADP. In tomato plants PK activity was greatest in
the leaves, and the activity increased with maturity of the tissues (BESFORD
and MAW 1975).
In experiments with K in tomatoes, and K and Mg in cucumbers, BESFORD
(1975) measured the PK activity of leaf extracts in the presence of optimum
concentrations of Mg when assessing the K status of the plants, and in the
presence of optimum K in the assessment of the Mg status. These PK activities
were then compared with the PK activities of corresponding leaf extracts assayed
in the presence of optimum concentrations of both K and Mg (controls). The
assay clearly distinguished between plants receiving adequate or inadequate
K supplies. The test also detected sub-optimal K levels that existed in the leaves
before deficiency symptoms had developed. The assay of phosphatase, as well
as of PK activity, reflected the Mg status of the cucumber plants. Subsequent
more detailed work with K and Mg in tomato plants fully confirmed previous
results and showed that a simultaneous assessment of the K and Mg status
of plants can be completed within 2 h of receiving the leaf samples (BESFORD
1978 a). The relationships between leaf K or leaf Mg and PK activities in toma-
toes are shown in Fig.4a andb respectively.
Potassium deficiency leads to characteristic deficiency symptoms in many
crops. In barley the well-defined necrosis which accompanies K deficiency
appears to be caused by the accumulation of the amine putrescine (RICHARDS
and COLEMAN 1952). COLEMAN and RICHARDS (1956) were able to induce a
leaf necrosis characteristic of K-deficiency in high-K barley plants by feeding
putrescine. They were also able to cause a disappearance of putrescine accumu-
lated in K-deficient barley by supplying K. BAR-AKIVA et al. (1974) suggested
that the knowledge of biochemical effects of nutrient excesses or deficiencies
may help in alleviating harmful effects on crop yields. They were able to over-
come at least partly the undesirable effects of high K on fruit quality of citrus
138 D. BOUMA:

60 (b)
90
(a)
/e.-. e-
>- a
:~~
.,co e...c: 40
>_80 /'" "'0 II)
.~~

I
'fl.!!!.
co e
70
, II)
<00
.,...
c: f c: 0
II)
:';;;:0
c:
:.;20
<00
0 60 ....,'".,
.,'" ~.:=: 20
2~
10 ~ 0 e
2~ >.,

,,$-
c..~
5
,t ~ 40

o
o 2 4 6 o 0.5 1.0 1.5
Potassiu m in leaf tissue Magnesium in leaf tissue
(g K per 100g d.wt) (g Mg per 100g d.wt)

Fig. 4. Relationship between pyruvate kinase activity and leaf potassium concentration
(a) and leaf magnesium concentration (b). The enzyme activity was measured in the
presence of added Mg (for K) or added K (for Mg) and expressed as a percentage
of the activity in the presence of K+Mg. (BESFORD 1978a)

(skin coarseness) by supplying putrescine to the leaves in orchard trees, or

to the media in tissue culture. Although there is no published information avail-
able, putrescine may be a useful index ofK-deficiency.

3.3.4 Iron and Manganese

Fe is possibly the classical example of an element for which it is often difficult,
if not impossible, to identify deficiency on the basis of plant or leaf analysis.
REUTHER et al. (1949) found no relationship between visible deficiency symptoms
and Fe concentration of citrus leaves. AGARWALA et al. (1965) found a poor
correlation between Fe supply and concentration in the tops of corn plants.
An additional difficulty is that symptoms of Fe and Mn deficiencies are some-
times difficult to differentiate (BAR-AKlVA 1961, O'SULLIVAN et al. 1969).
Is is not surprising, therefore, that alternative methods are being investigated
for the detection of Fe deficiency. Fe is a constituent of several metallo-enzymes
in which it is chelated by a porphyrin group. These enzymes include several
oxidases, peroxidase being one, which catalyze oxidation reactions in respiration
(DAVIES et al. 1964). Peroxidase activity is measured by the oxidation ofpyrogal-
101 by the leaf extract,in the presence of hydrogen peroxide (BAR-AKlVA et al.
1967). The method is simple and close relationships have been established be-
tween enzyme activity and Fe status of species as diverse as citrus (BAR-AKlVA
1961,1964, BAR-AKlVA and LAVON 1968) and Yorkshire fog, oats and tomatoes
(O'SULLIVAN et al. 1969).
Estimation of peroxidase activity has also been proposed as a means of
distinguishing between Fe and Mn deficiency in citrus (BAR-AKIVA 1961, 1964),
peroxidase activity showing a marked decrease when Fe is deficient, but an
increase with Mn deficiencv. This has also heen estahlished for timothv. oats
I.4 Diagnosis of Mineral Deficiencies Using Plant Tests 139

_ >- 2.0
co ...
'.j:; ";;:
:~ .~
'-co
~- - ~C~h~lo-rO-p~h~YI~1--~.--f J::
U)

"C 0.15 ~
Q) Q)
() U)
:::l co ?f2.
"C"C
.~§ 1.5 >-
Q)
LL~
~
0.10 g-
J::

Peroxidase ratio
o
J::
U

1.0 0.05
1 2
Fe in substrate (ppm)

Fig. 5. Ratio Fe-induced/initial peroxidase activity and chlorophyll content as affected

by Fe supply in substrate. (BAR-AKIVA and LAVON 1968)

and tomatoes, but not for Yorkshire fog (O'SULLIVAN et al. 1969). As an addi-
tional test for Mn deficieny, pentose (xylose) accumulation, which is specific
for this deficiency and is simple to measure colorimetrically, has been suggested
(BAR-AKIVA 1961,1964).
Enzyme activation by incubation of leaves with ferrous sulphate has been
used as a test for Fe deficiency in lemon trees (BAR-AKIVA and LAVON 1968).
A close relationship was found between the level of Fe supply to the trees
and the peroxidase activity of leaves ("initial activity"), and a close negative
relationship between Fe supply and the ratio induced/initial peroxidase activity.
This is illustrated in Fig. 5.

3.3.5 Copper
Cu is a constituent of the enzyme ascorbic acid oxidase which catalyzes the
oxidation of ascorbic acid. The enzyme is widely distributed in plants (DAVIES
et al. 1964). BROWN (1953) compared a range of crop plants and found that
ascorbic acid oxidase was a good index of available Cu in most plants, whether
they showed visual Cu deficiency symptoms or not. An increased enzyme activity
was found in all of the plants studied after addition of Cu to the soil. Similar
results were obtained with apple seedlings grown in solutions at different copper
levels (PERUMAL and BEATTIE 1966). In lemon trees the relationship between
growth and Cu concentration in the leaves was poor, but between growth and
enzyme activity quite close (BAR-AKIvA et al. 1969). In the same work it was
demonstrated that the enzyme could be induced by infiltration of Cu-deficient
leaf tissue with Cu, and that the response was greater the lower the Cu status
of the plants from which the leaves had been detached. The difference in re-
sponse between leaf tissue treated with and without Cu was, therefore, inversely
related to Cu status.
Such an approach overcomes many of the difficulties of leaf analysis in
the diagnosis of Cu deficiency. Enzyme activity is relatively easy to measure,
140 D. BOUMA:

either manometrically using ascorbic acid as substrate (BROWN 1953) or colori-

metrically as the residual ascorbic acid after incubating leaf tissue with a known
amount of ascorbic acid (BAR-AKIVA et al. 1969).

3.3.6 Zinc
Zn is a component of the enzyme carbonic anhydrase (CA; EVANS and SORGER
1966), which catalyzes the reaction: CO 2 +H 2 0-+H++HCO;, and is thus
thought to function in the transfer of CO 2 through the liquid phase of the
cell to the chloroplast surface (HATCH and SLACK 1970). In a wide range of
crops, including tomatoes, oats, spinach, citrus and beans, there is a close rela-
tionship between Zn concentration in leaf tissue and CA (BAR-AKIVA and LAVON
1969, EDWARDS and MOHAMED 1973, OHKI 1976, RANDALL and BOUMA 1973,
WOOD and SILBY 1952). While for some crops there is also a reasonably close
relationship between photosynthesis and CA, in others photosynthesis does not
decline except at very low CA levels, e.g. spinach (RANDALL and BOUMA 1973).
CA activity has been proposed as a diagnostic test for Zn deficiency in
a number of agricultural species, including citrus, wheat, mustard and maize
(BAR-AKIVA and LAVON 1969, DWEVIDI and RANDHAWA 1974). In Zn-deficient
citrus leaves, CA activities decreased to values only 25%-30% of those for
non-deficient leaves. Although deficiencies of some other minor elements also
caused reductions in CA activity, these were relatively small (BAR-AKIVA and
LAVON 1969). DWEVIDI and RANDHAWA (1974) used CA activity as a measure
of the Zn status of several crops at progressive stages of Zn deficiency. They
compared several indices of CA activity, referred to as "quick tests", with
a standard extraction procedure for CA and found correlation coefficients higher
than 0.9. The "quick tests" involved the estimation of CO 2 liberated by leaf
tissue incubated in the presence of NaHC0 3 • Although the activity values deter-
mined by these "quick tests" were lower than those for the extraction of the
enzyme, the relationship between these values was close enough to permit the
use of "quick tests" in the detection of Zn deficiency before visual symptoms
appeared, and earlier than could be detected by the Zn concentrations of leaf
tissue.

4 Prospects for the Future

The successful diagnosis of the nutrient status of a crop or a plant based on

plant analysis depends on the closeness of the relationship between nutrient
concentration at a certain stage of development and growth or-yield responses
to nutrients at that time or subsequently, established in prior field or pot experi-
ments. It also depends on the ability to recognize, evaluate and take into account
the many factors, discussed in preceding pages, which can modify these relation-
ships.
Although there is no doubt that plant analysis has made significant, and
often spectacular, progress in the past, further progress appears likely to remain
1.4 Diagnosis of Mineral Deficiencies Using Plant Tests 141

slow and painstaking. The plant physiological and nutritional principles underly-
ing this approach have been known for many years and there appear to be
no reasons to expect, or hope for, significant changes in the trends which have
characterized progress in the more recent past. These trends, which can be
summarized as the further extension and refinement of plant analysis as a diag-
nostic technique, are likely to continue. The best prospects for progress here
appear to lie in the further development of the nutritional relationships in diag-
nosis, in understanding the factors that can affect these, and finally in the
integration of these factors in diagnostic functions that can cope with the routine
application in providing reliable fertilizer advice for the farmer.
The physiological and biochemical techniques discussed in the second part
of this chapter are less empirical and, because they are based on well-known
physiological or biochemical functions or activities in the plant, they should
provide a more direct approach to diagnosis. This in itself could be a significant
step forwards.
Unfortunately, in much of the work reported above, the induced responses,
or the physiological or biochemical indices used to assess the nutrient status,
were correlated with nutrient supply, often in pot experiments, rather than
with crop or plant responses. To assess the value of these approaches, which
are often claimed to be more reliable and more rapid, they need to be tested
more rigorously than in the past under conditions applying in the real world.
It is difficult to escape the impression that at least some of this work is little
more than superficial applied biochemistry. In spite of the fact that work in
this area has been in progress for 10-20 years, as far as the author is aware,
relatively little is being applied in routine diagnosis of deficiencies.
It is suggested that the greatest challenge for future progress in this area
lies in the field testing and evaluation of these approaches to establish beyond
doubt their advantages as alternative diagnostic techniques.

Acknowledgements. Thanks are due to Dr. E.A.N. Greenwood, Division of Land Re-
sources Management, CSIRO, Wembley, W.A., and Mr K. Spencer, Division of Plant
Industry, CSIRO, Canberra, A.C.T., for their criticism of the manuscript.

References

Agarwala SC, Sharma CP, Farooq S (1965) Effect of iron supply on growth, chlorophyll,
tissue iron and activity of certain enzymes in maize and radish. Plant Physiol
40:493-499
Bar-Akiva A (1961) Biochemical indications as a means of distinguishing between iron
and manganese deficiency symptoms in Citrus plants. Nature 190:647-648
Bar-Akiva A (1964) Visible symptoms and chemical analysis versus biochemical indicators
as a means of diagnosing iron and manganese deficiencies in Citrus plants. In: Bould
C, Prevot P, Magness JR (eds) Plant analysis and fertilizer problems, vol IV. Am
Soc Hortic Sci, E Lansing, Mich
Bar-Akiva A (1971) Functional aspects of mineral nutrients in use for the evaluation
of plant nutrient requirement. In: Samish RM (ed) Recent advances in plant nutrition,
Proc 6th Int Colloq Plant Anal Fert Probl, vol I. Gordon & Breach New York
142 D. BOUMA:

Bar-Akiva A, Lavon R (1968) Peroxidase activity as an indicator of the iron requirement

of Citrus plants. Isr 1 Agric Res 18: 145-153
Bar-Akiva A, Lavon R (1969) Carbonic anhydrase activity as an indicator of zinc defi-
ciency in Citrus leaves. 1 Hortic Sci 44: 359-362
Bar-Akiva A, Sternbaum 1 (1965) Possible use of the nitrate reductase activity of leaves
as a measure of the nitrogen requirement of Citrus trees. Plant Cell PhysioI6:575-577
Bar-Akiva A, Kaplan M, Lavon R (1967) The use of a biochemical indicator for diagnos-
ing micronutrient deficiencies of grapefruit trees under field conditions. Agrochim
11 :283-288
Bar-Akiva A, Lavon R, Sagiv 1 (1969) Ascorbic acid oxidase activity as a measure
of the copper nutrition requirement of Citrus trees. Agrochim 14:47-54
Bar-Akiva A, Sagiv 1, Leshem 1 (1970) Nitrate reductase activity as an indicator for
assessing the nitrogen requirement of grass crops. 1 Sci Food Agric 21 :405-407
Bar-Akiva A, Sagiv 1, Renveni 0 (1974) Physiological approaches to plant nutritional
problems. In: Wehrmann 1 (ed) Proc 7th Int Colloq Plant Anal Fert Probl, vol I.
Germ Soc Plant Nutr, Hannover
Bates TE (1971) Factors affecting critical nutrient concentrations in plants and their
evaluation: A review. Soil Sci 112: 116-130
Bergmann EL, Boyle IS, Fries RE, Ferretti PA (1974) Influence of virus infection on
elemental content of selected vegetable crops. In: Wehrmann W (ed) Proc 7th Int
Colloq Plant Anal Fert Probl, vol I. Germ Soc Plant Nutr, Hannover
Bergmann W (1958) Die Ermittlung der Niihrstoffbediirftigkeit des Bodens. In: Ruhland
W (ed) Encyclopedia of plant physiology, vol IV. Springer, Berlin Gottingen Heidel-
berg
Besford RT (1975) Pyruvate kinase and a phosphatase as potential indicators of potassi-
um and magnesium status of tomato and cucumber plants. 1 Sci Food Agric
26: 125-133
Besford RT (1978a) Use of pyruvate kinase activity of leaf extracts for the quantitative
assessment of potassium and magnesium status of tomato plants. Ann Bot 42: 317-324
Besford RT (1978b) A phosphatase as a potential indicator of the phosphorus status
of the glasshouse cucumber (Cucumis sativus). 1 Sci Food Agric 29:87-91
Besford RT (1979 a) Nutrient imbalances in tomato plants and acid phosphatase activity
in the leaves. 1 Sci Food Agric 30:275-280
Besford RT (1979b) Phosphorus nutrition and acid phosphatase activity in the leaves
of seven plant species. 1 Sci Food Agric 30:281-285
Besford RT (1979c) Quantitative aspects of leaf acid phosphatase activity and the phos-
phorus status of tomato plants. Ann Bot 44: 153-161
Besford RT (1980) A rapid tissue test for diagnosing phosphorus deficiency in the tomato
plant. Ann Bot 45: 225-227
Besford RT, Maw GA (1975) Some properties of pyruvate kinase extracted from Lycoper-
sicon esculentum. Phytochemistry 14:677-682
Bonner 1, Varner IE (1965) The path of carbon in respiratory metabolism. In: Bonner
1, Varner JE (eds) Plant biochemistry. Academic Press, London New York
Bould C (1961) Leaf analysis as a guide to the nutritional status of soft fruit crops.
In: Reuther W (ed) Plant analysis and fertilizer problems. Am Inst BioI Sci, Washing-
ton DC
Bould C (1964) Leaf analysis in relation to raspberry nutrition. In: Bould C, Prevot
P, Magness lR (eds) Plant analysis and fertilizer problems, vol IV. Am Soc Hortic
Sci, E Lansing, Mich
Bould C (1968) Leaf analysis as a diagnostic method and advisory aid in crop nutrition.
Exp Agric 4: 17-27
Bouma D (1956) Studies in citrus nutrition. II. Phosphorus deficiency and fruit quality.
Aust 1 Agric Res 7:261-271
Bouma D (1959a) Growth, yield, and fruit quality in a factorial field experiment with
citrus in relation to changes in phosphorus nutrition. Aust 1 Agric Res 10:41-51
Bouma D (1959b) The development of the fruit of the Washington Navel orange. Aust
1 Agric Res 10:804-817
1.4 Diagnosis of Mineral Deficiencies Using Plant Tests 143

Bouma D (1961) The development of cuttings of the Washington Navel orange to the
stage of fruit set. IV. The effect of different nitrogen and phosphorus levels on fruiting
cuttings. Aust J Agric Res 12: 1089-1099
Bouma D (1967) Growth changes of subterranean clover during recovery from phospho-
rus and sulphur stresses. AustJ BioI Sci 20:51-66
Bouma D (1975) Effects of some metabolic phosphorus compounds on rates ofphotosyn-
thesis of detached phosphorus-deficient subterranean clover leaves. J Exp Bot
26: 52-59
Bouma D, Dowling EJ (1962) The physiological assessment of the nutrient status of
plants. I:Preliminary experiments with phosphorus. Aust J Agric Res 13: 791-800
Bouma D, Dowling EJ (1966a) The physiological assessment of the nutrient status of
plants. II. The effect of the nutrient status of the plant with respect to phosphorus,
sulphur, potassium, calcium, or boron on the pattern of leaf area response following
the transfer to different nutrient solutions. Aust J Agric Res 17: 633-646
Bouma D, Dowling EJ (1966b) The physiological assessment of the nutrient status of
plants. III. Experiments with plants raised at different nitrogen levels. Austr J Agric
Res 17: 647-655
Bouma D, Dowling EJ (1967) The physiological assessment of the nutrient status of
plants. IV. The effect of the interaction between nutrient elements on leaf area re-
sponses. Aust J Agric Res 18:223-233
Bouma D, Dowling EJ (1969a) Effects of temperature on growth and nutrient uptake
in subterranean clover during recovery from phosphorus stress. I. Growth changes.
Aust. J BioI Sci 22:505-514
Bouma D, Dowling EJ (1969b) Effects of temperature on growth and nutrient uptake
in subterranean clover during recovery from phosphorus stress. II. Phosphorus uptake
and distribution. Aust J BioI Sci 22:515-522
Bouma D, Dowling EJ (1976) The relationship between the phosphorus status of subterra-
nean clover plants and the dry weight responses of detached leaves in solutions with
and without phosphate. Aust J Agric Res 27: 53-62
Bouma D, Dowling EJ (1980) Field evaluation of a test for phosphorus deficiency
in pastures based on dry matter responses, induced in detached subterranean clover
leaves. Commun Soil Sci Plant Anal 11 : 861-872
Bouma D, Dowling E1. Wahjoedi H (1979) Some effects of potassium and magnesium
on the growth of subterranean clover (Trifolium subterraneum L.). Ann Bot
43:529-538
Bouma D, Spencer K, Dowling EJ (1969) Assessment of the phosphorus and sulphur
status of subterranean clover pastures. 3. Plant tests. Aust J Exp Agric Anim Husb
9:329-340
Brown JC (1953) The effect of the dominance of a metabolic system requiring iron
or copper on the development of lime-induced chlorosis. Plant Physiol 28 :495-502
Chapman HD (ed) (1966) Diagnostic criteria for plants and soils. Univ Cal, Div Agric
Sci, pp 1-793
Coleman RG, Richards FJ (1956) Physiological studies in plant nutrition. XVIII. Some
aspects of nitrogen metabolism in barley and other plants in relation to potassium
deficiency. Ann Bot 20:393-409
Davies DD, Giovanelli J, Ap Rees T (1964) Plant biochemistry. Blackwell, Oxford
Dwevidi RS, Randhawa NS (1974) Evaluation of a rapid test for the hidden hunger
of zinc in plants. Plant Soil 40: 446--450
Edwards GE, Mohamed AK (1973) Reduction in carbonic anhydrase activity in zinc-
deficient leaves of Phaseolus vulgaris L. Crop Sci 13: 351-354
Emmert FH (1959) Chemical analysis of tissue as a means of determining nutrient require-
ments of deciduous fruit plants. Proc Am Soc Hort Sci 73: 521-547
Emmert FH (1961) The bearing of ion interactions on tissue analysis results. In: Reuther
W (ed) Plant analysis and fertilizer problems. Washington, D.C.: Am Inst BioI Sci,
Washington DC
Epstein E (1972) Mineral nutrition of plants: Principles and perspectives. Wiley and
Sons. New York
144 D. BOUMA:

Epstein E, Lilleland 0 (1942) A preliminary study of the manganese content of the

leaves of some deciduous fruit trees. Proc Am Soc Hort Sci 41 : 11-18
Evans HJ, Nason A (1953) Pyridine nucleotide-nitrate reductase from extracts of higher
plants. Plant Physiol 28: 233-254
Evans HJ, Sorger GJ (1966) Role of mineral elements with emphasis on the univalent
cations. Annu Rev Plant PhysioI17:47-76
Freney JR, Spencer K, Jones MB (1978) The diagnosis of sulphur deficiency in wheat.
Aust J Agric Res 29: 727-738
Gates CT (1957) The response of the young tomato plant to a brief period of water
shortage. III. Drifts in nitrogen and phosphorus. Aust J BioI Sci 10:125-146
Gates CT (1974) Water shortage and agriculture: Some responses. J. Aust Inst Agric
Sci 40:121-142
Goodall DW, Gregory FG (1947) Chemical composition of plants as an index of their
nutritional status. Tech Commun 17, Imp Bur Hortic Plant Crops, E Malling
Greenham CG, Randall PJ, Ward MM (1972) Impedance parameters in relation to
phosphorus and calcium deficiencies in subterranean clover (Trifolium subterraneum
L.). J Exp Bot 23:197-209
Greenway H, Pitman MG (1965) Potassium retranslocation in seedlings of Hordeum
vulgare. Aust J BioI Sci 18:235-247
Greenwood EAN (1966) Nitrogen stress in wheat - its measurement and relation to
leaf nitro gen. Plant Soil 24: 279-288
Greenwood EAN (1976) Nitrogen stress in plants. Adv Agron 28: 1-35
Greenwood EAN, Titmanis ZV (1966) The effect of age on nitrogen stress and its relation
to leaf nitrogen and leaf elongation in a grass. Plant Soil 24: 379-389
Greenwood EAN, Goodall DW, Titmanis ZV (1965) The measurement of nitrogen defi-
ciency in grass swards. Plant Soil 23: 97-116
Gregory FG (1953) The control of growth and reproduction by external factors. In:
Rep 13th Int Hortic Congr London, R Hortic Soc
Groenewegen H, Bouma D (1960) The chemical composition of the soil in a factorial
experiment with Citrus. I. Exchangeable metal cations and their effect on the cation
content of citrus leaves. Aust J Agric Res 11 : 208-222
Hatch MD, Slack CR (1970) Photosynthetic COrfixation pathways. Annu Rev Plant
Physiol21: 141-162
Hewitt EJ, Tatham P (1960) Interaction of mineral deficiency and nitrogen source on
acid phosphatase activity in leaf extracts. J Exp Bot 11 : 367-375
Janssen BH (1974) A double pot technique for rapid soil testing. Trop Agric (Trinidad)
51 :161-166
Jones WW, Parker ER (1950) Seasonal variations in mineral composition of orange
leaves as influenced by fertilizer practices. Proc Am Soc Hort Sci 55: 92-100
Kenworthy AL (1967) Plant analysis and interpretation of analysis for horticultural crops.
In: Hardy GW et al. (eds) Soil testing and plant analysis, part II, plant analysis.
Soil Sci Soc Am Madison, Wisc
Kirkpatrick JD, Gundy van SD, Mai WF (1964) Interrelationships of plant nutrition,
growth, and parasitic nematodes. In: Bould C, Prevot P, Magness JR (eds) Plant
analysis and fertilizer problems, vol IV. Am Soc Hortic Sci, E Lansing, Mich
Loneragan JF (1978) Anomalies in the relationship of nutrient concentrations to plant
yield. In: Ferguson AR, Bieleski RL, Ferguson IB (eds) Plant nutrition 1978. DSIR
Inf Ser 134. Government Printer, Wellington
Loneragan JF, Snowball K (1969) Rate of calcium absorption by plant roots and its
relation to growth. Aust J Agric Res 20:479-490
Lundegardh H (1966) Plant Physiology. Edinburgh and London: Oliver and Boyd, Edin-
burgh London
Macy P (1936) The quantitative mineral nutrient requirements of plants. Plant Physiol
11 :749-764
Mason AC (1958) The concentration of certain nutrient elements in apple leaves taken
from different positions on the shoot and at different dates through the growing
season. J Hortic Sci 33: 128-138
1.4 Diagnosis of Mineral Deficiencies Using Plant Tests 145

Millikan CR (1963) Effects of different levels of zinc and phosphorus on the growth
of subterranean clover (Trifolium subterraneum L.). Aust J Agric Res 14: 180 - 205
Millikan CR, Hanger BC (1964) Effect of calcium level in the substrate on the distri-
bution of 45Ca in subterranean clover (Trifolium subterraneum L.). Aust J BioI Sci
17: 823-844
Mulder EG, Boxma R, Veen van WL (1959) The effect of molybdenum and nitrogen
deficiencies on nitrate reduction in plant tissues. Plant Soil 10: 335-355
Muller A (1974) Nutrient deficiencies in a volcanic ash soil from Ecuador. In: Wehrmann
J (ed) Proc 7th Int Colloq Plant Anal Fert Probl, vol II. Hannover, W. Germ.:
Germ Soc Plant Nutr Hannover
Nason A, Evans HJ (1953) Triphosphopyridine nucleotide-nitrate reductase in Neurospo-
ra. J BioI Chern 202: 655-673
Nicholas DJD, Nason A (1955) The role of molybdenum as a constituent of nitrate
reductase from soybean leaves. Plant Physiol 30: 135-138
Ohki K (1976) Effect of zinc nutrition on photosynthesis and carbonic anhydrase activity
in cotton. Physiol Plant 38: 300-304
O'Sullivan M, Flynn MJ, Codd FJ (1969) A biochemical method for diagnosing micronu-
trient deficiencies in plants. Ir J Agric Res 8: 111-119
Ozanne PG (1955) The effect of nitrogen on zinc deficiency in subterranean clover.
Aust J BioI Sic 8:47-55
Peaslee DE, Moss DN (1966) Photosynthesis in K- and Mg-deficient maize (Zea mays
L.) leaves. Soil Sci Soc Am Proc 30:220-223
Perumal A, Beattie JM (1966) Effect of different levels of copper on the activity of
certain enzymes in leaves of apple. Proc Am Soc Hortic Sci 88 :41-47
Piper CS (1942) Investigations on copper deficiency in plants. J Agric Sci 32: 143-178
Prevot P, Ollagnier M (1961) Law of the minimum and balanced mineral nutrition.
In: Reuther W (ed) Plant analysis and fertilizer problems. Am Inst BioI Sci, Washing-
ton DC
Proebsting EL, Kenworthy AL (1954) Growth and leaf analysis of Montmorency cherry
trees as influenced by solar radiation and intensity of nutrition. Proc Am Soc Hortic
Sci 63 :41-48
Randall PJ (1969 a) Restoration of protein level in detached leaves by supply of sulphate.
Plant Soil 31: 385-388
Randall PJ (1969b) Changes in nitrate and nitrate reductase levels on restoration of
molybdenum to molybdenum-deficient plants. Aust J Agric Res 20: 635-642
Randall PJ, Bouma D (1973) Zinc deficiency, carbonic anhydrase, and photosynthesis
in leaves of spinach. Plant Physiol 52: 229-232
Reuther W, Boynton D (1939) Variations in potassium content of foliage from certain
New York apple orchards. Proc Am Soc Hortic Sci 37: 32-38
Reuther W, Smith PF (1954) Leaf analysis of Citrus. In: Childers NF (ed) Fruit nutrition.
Hortic Publ Rutgers Univ, New Brunswick, New Jersey
Reuther W, Embleton TW, Jones WW (1958) Mineral nutrition of tree crops. Annu
Rev Plant Physiol 9: 175-206
Reuther W, Smith PF, Specht A (1949) A comparison of the mineral composition of
Valencia orange leaves from the major producing areas of the United States. Proc
Fla Hortic Soc 60: 38-45
Richards FJ, Coleman RG (1952) Occurrence of putrescine in potassium-deficient barley.
Nature 170:460
Sato K (1961) The effects of growth and fruiting on leaf composition of citrus trees.
In: Reuther W (ed) Plant analysis and fertilizer problems. Am Inst BioI Sci, Washing-
ton DC
Shaked A, Bar-Akiva A (1967) Nitrate reductase activity as an indication of molybdenum
level and requirement of citrus plants. Phytochemistry 6:347-350
Shaulis NJ (1961) Associations between symptoms of potassium deficiency, plant analysis,
growth and yield of Concord grapes. In: Reuther W (ed) Plant analysis and fertilizer
problems. Am Inst BioI Sci, Washington DC
Smith PF (1962) Mineral analysis of plant tissues. Annu Rev Plant Physiol13:81-108
146 D. BOUMA: 1.4 Diagnosis of Mineral Deficiencies Using Plant Tests

Smith PF, Reuther W (1950) Seasonal changes in Valencia orange trees. I. Changes
in leaf dry weight, ash, and macro-nutrient elements. Proc Am Soc Hortic Sci 55: 61-72
Spencer K, Jones MB, Freney JR (1977) Diagnostic indices for sulphur status of subterra-
nean clover. Aust J Agric Res 28:401-412
Steenbjerg F (1951) Yield curves and chemical plant analyses. Plant Soil 3:97-109
Sumner ME (1974) An evaluation of Beaufils' physiological diagnosis technique for deter-
mining the nutrient requirements of crops. In: Wehrmann J (ed) Proc 7th Int Colloq
Plant Anal Fert Probl, vol II. Hannover, W. Germ. Germ Soc Plant Nutr, Hannover
Ulrich A (1948) Plant analysis - methods and interpretation of results. In: Kitchen
HB (ed) Diagnostic techniques for soils and crops. Am Pot Inst, Washington DC
Ulrich A (1952) Physiological bases for assessing the nutritional requirements of plants.
Annu Rev Plant Physiol 3: 207-228
Ulrich A (1961) Plant analysis in sugarbeet nutrition. In: Reuther W (ed) Plant analysis
and fertilizer problems. Am Inst BioI Sci, Washington DC
Ulrich A, Hills FJ (1967) Principles and practices of plant analysis. In: Hardy GW
et al. (eds) Soil testing and plant analysis, part II, plant analysis. Madison, Wisc
Wadleigh CH, Richards LA (1951) Soil moisture and the mineral nutrition of plants.
In: Truog E (ed) Mineral nutrition of plants. Univ Wisc Press, Madison
West ES (1938) Zinc-cured mottle leaf in citrus induced by excess phosphate. J Counc
Sci Ind Res Aust 11: 182-184
Williams RF (1948) The effects of phosphorus supply on the rates of intake of phosphorus
and nitrogen and upon certain aspects of phosphorus metabolism in gramineous
plants. Aust J Sci Res Ser B 1: 336: 361 .
Williams RF (1955) Redistribution of mineral elements during development. Annu Rev
Plant PhysioI6:81-108
Williams RF, Shapter RE (1955) A comparative study of growth and nutrition in barley
and rye as affected by low water treatment. Aust J BioI Sci 8:435-466
Wilson AE (1961) Leaf analysis and fertility control in commercial citrus production.
In: Reuther W (ed) Plant analysis and fertilizer problems. Am Inst BioI Sci, Washing-
ton DC
Witt HH, Jungk A (1974) The nitrate inducible nitrate reductase activity in relation
to nitrogen nutritional status of plants. In: Wehrmann J (ed) Proc 7th Int Colloq
Plant Anal Fert Probl, vol II. Hannover, W. Germ. Germ Soc Plant Nutr Hannover
Wood JG, Silby PM (1952) Carbonic anhydrase activity in plants in relation to zinc
content. Aust J Sci Res Ser B 5:244-255
Zurbucki ZI (1961) Dependence of mineral composition of plants on environmental
conditions. In: Reuther W (ed) Plant analysis and fertilizer problems. Am Inst BioI
Sci Washington DC
1.5 Interactions Between Nutrients in Higher Plants
A.D. ROBSON and M.G. PITMAN

1 Introduction

Interactions between nutrients in higher plants occur when the supply of one
nutrient affects the absorption, distribution or function of another nutrient.
Thus, depending on nutrient supply, interactions between nutrients can either
induce deficiencies or toxicities and can modify growth response. Interactions
between nutrients can be assessed by examining both the relationship between
nutrient supply and growth and the relationship between nutrient concentrations
in plants and growth. Additionally where nutrient supply is neither deficient
nor toxic for plant growth, interactions can be assessed by considering nutrient
concentrations and contents within the plant.
There are several ways in which nutrients can interact either within the
soil or plant to affect nutrient absorption or utilization. However, these interac-
tions can be classified in two major categories. Firstly there are interactions
which occur between ions because the ions are able to form a chemical bond
(either ionic or covalent). Hence interactions can occur due to the formation
of precipitates or complexes. This form of interaction is generally most marked
when the interacting ions have very different chemical properties [for example
anions interacting with cations, a small cation with a high charge density (elec-
tron acceptor or hard acid) with a small anion with a high charge density
(electron donor or hard base), or a large cation with a low charge density
(soft acid) with a large anion with a low charge density (soft base)].
Interactions between nutrients relating to the formation of precipitates or
complexes (either in the soil or the plant) can be predicted from solubility
products and binding or stability constants. However, to use this chemical data
to define the role of these processes in interactions among nutrients, concentra-
tions and forms of nutrients (oxidation state, complexed anion or cation) in
each of the compartments (soil solution, cell wall of root, cytoplasm of leaf,
vacuole in leaf and phloem) must be known. Our knowledge of the concentra-
tions and chemical states of nutrients in many of these compartments is very
limited. Many micronutrients do not exist as divalent cations in either the soil
solution or the xylem (Table 1). The formation of complex anions between
micronutrients and organic ligands may have several important consequences.
For example, from a consideration of the solubility of eu and Zn compounds
alone it would be predicted that liming soils would decrease the concentrations
of both eu and Zn in soil solutions to a similar extent and hence decrease
the uptake of eu and Zn (LINDSAY 1978). However, because eu is more com-
plexed by soluble organic matter than Zn (HODGSON et al. 1966), effects of
148 A.D. ROBSON and M.G. PITMAN:

Table 1. Concentrations and forms of mineral nutrients in soil solutions, xylem and
phloem

Nutrient Soil solution a Xylem b Phloem c

Conc (IlM) Form Conc Form Conc Form

(IlM) (IlM)
Range Median
values

Calcium 0-5,000 1,500 Ca2+ 400 to Ca2+ 250 to nd

4,500 2,650
Potassium 0-5,000 1,300 K+ 500 to K+ 20,000 to nd
11,500 85,000
Magnesium 0-6,000 4,000 Mg2+ 80-1,125 Mg2+ 2,100 to nd
23,000
Sodium ndd 125 Na+ 1,600 to Na+ 60 to nd
2,600 19,000
Phosphorus 0-2 1 H2PO;, 30-1,500 H 2PO; 10,000 nd
HPO;-
Sulphur 0-2,000 500 SO~- 1,400 SO~- 900 to no SO~-
1,200
Nitrogen 0-3,000 1,500 NO~:' 700 to NO~, noNO~
NH4 30,000 amino-N
Chlorine nd 200 Cl- 1,300 to Cl- 7,900 to nd
3,300 11,900
Iron <0.03-3 Complexed 15-30 Fe-complex 75-220 nd
(anion)
Copper <0.01 Complexed 1.5-2.0 Cu-complex 3-140 nd
to 0.6 (89%-100%) (anion)
Zinc <0.03-5 Complexed 5-10 Zn-complex 3-170 Zn-
(28%-99%) (anion) complex
(anion)
Manganese <0.02 Complexed 7-11 Cation 9-25 Mainly
(84%-99%) cation
Cobalt <0.007 Complexed 0.4-14 nd nd Co-
to 0.2 (8%-50%) complex
(anion)
Boron 3-1,000 H 3B0 3 60-75 H3B03(?) nd nd
Molyb- 0.02 MoO~- 0.5-4 MoO~-(?) 0.1 nd
denum to 0.08
pH 4-8 4.9-6.3 7.9-8.2

a Values are from REISENAUER (1966) for macronutrients and from LONERAGAN (1975)
for micronutrients. Values for sodium and chloride are those cited by RUSSELL (1973)
for typical soils
b Values are collated from BOLLARD (1960), HUSA and McILRATH (1965), TIFFIN (1967),
JONES and ROWE (1968), PATE (1975) and KIRKBY and ARMsTRONG (1980)
C Values are collated from TAMMES and VAN DIE (1964), MACROBBIE (1971), PATE (1975)

and VAN GOOR and WIERSMA (1974)

d nd no data available
1.5 Interactions Between Nutrients in Higher Plants 149

increasing soil pH are more marked on Zn uptake than on Cu uptake by plants

(PIPER and BECKWITH 1949, WILLIAMS 1977). An understanding of the forms
and concentrations of nutrients in soil solution is also required if studies of
interactions in solution culture are to be applicable to interactions occurring
within soil. Many experiments examining interactions involving micronutrients
are conducted using concentrations rarely, if ever, found in soil solutions and
using cations even though many of the micronutrients occur as complex anions
in soil solutions (Table 1).
To avoid both precipitation either in the soil or within plant cells and immo-
bilization by adsorption to cell walls, ions may be absorbed and transported
in different forms to that required for metabolism within the cell. Similarly
nutrients may be stored in cells in forms (including as insoluble salts) other
than as simple anions or cations to avoid interactions between nutrients affecting
metabolic roles.
The second form of interactions is between ions with chemical properties suffi-
ciently similar to compete for sites of adsorption (on clay minerals and oxides
in soil and in cell walls of plants), absorption (both influx and efflux), transport
(within xylem and phloem) and function (at active sites). For these interactions,
competition will be greatest for nutrients with similar size, charge, geometry
of co-ordination and electronic configuration. If competition occurs between
ions in absorption and transport but at the site of function, there is more
marked specificity, the effect of one ion in reducing the concentration of the
required ion at the site of action may decrease the rate of that reaction. However,
some roles of nutrients are sufficiently non-specific (for example maintenance
of electroneutrality or osmotic regulation) that one nutrient can substitute for
another (for example, Na for K).
Interactions between nutrients related to competition between ions occur
because there is insufficient specificity. The degree of specificity or selectivity
varies widely for different processes. For adsorption of cations onto surfaces
of clay minerals in soils, competition is related largely to the concentration
in solution and the ratio of charge to radius (see RUSSELL 1973). Similarly
the adsorption of ions by fixed carboxylates in the cell walls of plants is a
relatively non-selective process. Root surfaces are thought to act as cation ex-
changers with little selectivity for particular cations other than that related
to charge density (LEDIN and WIKLANDER 1974); however, their selectivity has
been examined for relatively few cations (mainly alkalis and alkaline earth
cations). However, for active absorption of ions across the plasmalemma only
ions with extremely similar chemical properties compete at low concentrations
(SeOi- jSOi-, K +jRb+ but not K +jNa+, Ca2 +jSr2 +).
For incorporation into metalloproteins and for the functioning of those
metalloproteins, selectivity is generally more marked. Indeed for some metallo-
proteins only one metal ion is incorporated into their structure. Additionally,
if more than one ion can be incorporated into a metalloprotein, biological
activity may only be retained with one ion. It is interesting that the activity
of metal ions with enzymes does not correlate with strength of binding or
stability of metal chelates (PRICE 1970, SUELTER 1974).
In competitive binding the amount of any ion bound will be related to
the concentration of that ion relative to its competitors and the affinity constants
150 A.D. ROBSON and M.G. PITMAN:

for the ion-ligand pairs (DA SILVA and WILLIAMS 1977). The affinity constants
depend upon both properties of the ions (charge, size, electronic configuration)
and of the ligand (size of cavity, charge, extent of chelation, co-ordination
geometry). Additionally the stability of complexes formed by ions and ligands
will vary with the solution pH (affect dissociation of the ion and ligand), the
ionic strength of the solution, the effective redox potential (affect oxidation
state of the ion) and the nature of the solvent (affect the hydration of the
ion). Hence since these chemical properties differ within different compartments
within the plant, selectivity of ligands could be different in different compart-
ments (for example comparison of pH in xylem and phloem (Table 1).
DA SILVA and WILLIAMS (1977) also point out that incorporation of metal
ions into specific sites may be achieved by different rates of reactions (kinetic
factors) as well as by thermodynamic considerations. They suggest that the
extremely high specificity of incorporation of Mg into chlorin, Cu into uropor-
phyrin, Fe into protoporhyrin and Co into corrin (in vivo which is difficult
to achieve in vitro) is due to differences in rates of reactions involving intermedi-
ate products.
Cations differ in their affinity for ligands due to differences in their electronic
configuration (electropositivity and electronegativity). Sodium and K have high
affinities for ether, alcohol or carbonyl-O-ligands; Ca and Mg have high affini-
ties for carboxylate or phosphate-O-ligands, while the transition metals have
high affinities for -N- and -S-ligands (DA SILVA and WILLIAMS 1977). Biological
systems must have a wide variety of ligands separated in space to concentrate
and to separate the inorganic elements. The degree of strength of complexing
differs markedly between the cations. Potassium, Na, Mg, Ca and Mn exist
mainly as free or reversibly bound ions, whereas Fe, Cu, Zn and Mo are present
mainly as structural chelates or metalloproteins (MENGEL and KIRKBY 1978).
Anions also differ in their affinity for ligands. Ions like CI- are only weakly
bound at hydrophobic centres, whereas ions like H 2 PO; may form moderately
strong complexes through H bonding and condensation. Of the anions, P, B
and Si are present largely as oxyanions or undissociated molecules, Nand
S are covalently bonded, whereas CI is present as free or reversibly bound
Cl- (MENGEL and KIRKBY 1978). Molybdenum which exists largely as MoOr-
(oxidation state VI) within the soil solution is reduced within the cell to Mo
(V) and behaves like a metal with affinity for both Nand S ligands.
As well as interactions which occur because of effects of one element on
the absorption, transport or utilization of another there may be interactions
relating to growth responses. Where two nutrients correct separate deficiencies,
the interaction between them on growth is positive, i.e. the effect of one essential
element on growth is greatest when all other nutrients are in optimum supply.
There are many examples of positive interactions between essential nutrients
on growth (see ANDERSON 1956). Where two nutrients correct the same defi-
ciency the interaction between them on growth is negative (for example see
Sect. 2.2.2.3).
Many interactions between nutrients have been noted in the literature (Ta-
ble 2). As well as the interactions listed, competitive interactions can occur
1.5 Interactions Between Nutrients in Higher Plants 151

Table 2. Interactions occurring between elements which affect the absorption and utiliza-
tion of nutrients by plants. (+, interaction recorded; 0, no interaction observed after
investigation; no satisfactory data for other combinations)

'"
Nt{

K+
+ l'"
No" + 1"'-

'"'"
(0 2+
+ +
Mg2+
+ +
Mn 2• + ~

'"'"
(u 2+

Zn2+
+ + +
Fe 3", Fe 2+
+ + ~

'"'"
A13'
+ +
(1- NO) HlO;: SeO!- SO~- HoO~- H3B0 3H4SiO,

(1-
+ ~

'"
N0 3
+
H2Po;; IHPOt- + + + + + 0 0 ~

'"'"
SeOl-

S02-
4 0 0 +

I"
Mo°I- + + + + I'"
H3B03 + +
H4Si04 + + ~

between essential nutrients and other elements in the environment (particularly

trace elements with other heavy metals, for example Zn-Cd interactions). Be-
cause of the chemical basis of the interaction between ions, many interactions
which affect the nutrition of plants also affect the nutrition of animals (see
review by UNDERWOOD 1977).
In this review we have decided not to consider all of the interactions that
occur between elements which affect the absorption and utilization of nutrients.
Instead we have chosen specific examples that show the mechanisms that give
rise to the interaction. Interaction between nutrients is covered also in Chap.
IV.2 and in Chap. V.3, this Volume.
152 A.D. ROBSON and M.G. PITMAN:

2 Interactions Between Nutrients in Monoculture

2.1 Interactions Between Nutrients Affecting the Absorption of Nutrients

There are many interactions between nutrients which change the relationship
between nutrient supply and growth without changing the relationship between
nutrient concentrations in the plant and growth. Additionally, where nutrients
are not deficient for plant growth, the application of one nutrient may depress
both the concentration and content of another nutrient within the plant. Interac-
tions of this nature operate either in the soil or at the absorbing sites within
the root.

2.1.1 Interactions Occurring in the Soil

Plants absorb nutrients from the soil solution. However, the concentration of
nutrients in the soil solution depends upon reactions occurring within the soil
(Fig. 1). Interactions may arise from the complex equilibria between the solution
and other components of the soil. The formation of complexes, the precipitation
of insoluble salts and the adsorption by soil colloids can all be affected by
the concentration of other ions. We will consider here the nature of interactions
between ions related to either adsorption or precipitation.

2.1.1.1 Adsorption
Concentrations of both anions and cations in solution can be controlled by
adsorption onto soil colloids (see reviews by RUSSELL 1973, NYE and TINKER
1977). Ion adsorption by soil components is termed either non-specific or specific
(chemisorption). Non-specific adsorption occurs for ions adsorbed onto charged
surfaces of opposite signs (for example cation adsorption on negatively charged
surfaces of clay minerals). Ions of equal valency are adsorbed in simple propor-
tion to their equilibrium activity in solution. In contrast to non-specific adsorp-
tion, specific adsorption can occur on uncharged surfaces or even on surfaces
of like sign, as well as on surfaces of unlike sign (adsorption of cations and
anions on surface of iron oxides). Specifically adsorbed ions are held by chemical
bonds as well as by electrostatic attraction (see BOWDEN et al. 1973 for a fuller
discussion). Competition between ions which are specifically adsorbed is more
complex than competition between non-specifically adsorbed ions because the
adsorption of one ion can reverse the sign of the surface charge.
Effects of soil acidity on nutrient uptake can arise through effects of acidity
on the adsorption of ions. For example, the application of CaC0 3 to acid
soils can markedly increase the growth of legumes on acid soils by effects on
the uptake of Mo by plants (ANDERSON and OERTEL 1946). The adsorption
of MoO~- by soils and by oxides of Al and Fe decreased as pH was increased
above 4 (JONES 1957). The adsorption of both SO~- and H 2 PO';:- was also de-
creased by increasing soil pH although the effects are much less marked than
those observed for MoO~- (BARROW 1970). The effect of pH on the adsorption
of anions can be related to the variable surface charge of Fe and Al oxides
1.5 Interactions Between Nutrients in Higher Plants 153

Nutr ients in
plant shoots
(i) Function
(i i) Re-trans-
location

Transport Transport
in xylem in phloem

Nutrients in
plan t root

absorption efflux, exchange

Soil solution
nutrients

mineralization

dissolution precipitation

immobilization adsorption

Nutrients Nutrients Nutrients

held in preci pi tated adsorbed on
organic as inorganic su rfaces
matter solids

Fig. 1. Processes in soils and plants where interactions between nutrients may occur

(BOWDEN et al. 1973). As the solution pH nears the pKa of the appropriate
acid the proportion of negatively charged ions in solution increases rapidly.
Hence adsorption will increase because only the negatively charged ions are
adsorbed. However, as the pH increases beyond the pKa the tendency for the
surface to carry a positive charge decreases and adsorption will decrease. Hence
maximum adsorption for particular anions tends to occur when solution pH
equals the pKa of their acid (for example at pH 4.0 for molybdate, at pH 9.2
for silicate). For polybasic acids with large differences in pKa (for example
H 3 P0 4 ) there are discontinuities at the pKa values rather than a maximum.
The anions of fully dissociated acids are not adsorbed unless the surface is
positive. Thus SO~ - is adsorbed only on the acid side of the zero point of
charge (the point at which equal amounts of H + and OH - are adsorbed).
Interactions between elements affecting nutrient uptake by plants can arise
through effects of one ion on the adsorption of another. Here we will consider
(1) competition in the adsorption of anions (silicate-phosphate, phosphate-mo-
lybdate), (2) adsorption and incorporation of a cation into an oxide surface
(Co-Mn) and (3) the effect of anion adsorption on the adsorption of a cation
(P-Zn).
154 A.D. ROBSON and M.G. PITMAN:

The application of CaSi0 3 may increase the growth of plants by increasing

P uptake. For example, on highly weathered soils in Hawaii, CaSi0 3 increased
the growth of soybeans only when P supply was limited (Fox 1978). In these
situations where the P concentrations in the soil solution are low, the added
soluble silicates may displace H 2 PO';- adsorbed onto Al and Fe oxides particu-
larly in neutral or alkaline soils. The competition between phosphate and silicate
for adsorption onto the surfaces of goethite depends upon solution pH
(HINGSTON et al. 1971, 1972). Silicate can only be adsorbed in the presence
of H 2 PO';- when it can increase the negative charge on the surface, that is
above pH 7. Competition between other anions in adsorption also occurs. For
example adding H 2 PO';- to an acid soil very low in MoO~- may increase Mo
concentrations in soil solutions and the uptake of Mo by plants by displacing
MoO~- from the surface of soil colloids (STOUT et al. 1951).
The adsorption of cations onto the surfaces of oxides can also lead to interac-
tions affecting nutrient uptake. One particularly striking example is the close
association between Co and Mn in soils. Differences among soils in the availabil-
ity of applied Co to plants was inversely related to their total Mn content
(ADAMS et al. 1969). On soils not containing high contents of montmorillonite,
adsorption capacities for Co were positively correlated with their Mn content
(TILLER et al. 1969). Waterlogging soils results in the dissolution of Mn oxides
[Mn (IV) reduced to Mn (II)] and markedly increases the uptake of Co by
plants (ADAMS and HONEYSETT 1964). These observations are related to the
ability of Mn oxides to adsorb Co from solution and to hold it in a form
unavailable to plants. McKENzm, on the basis of an elegant series of experiments
(McKENzm 1967, 1970, 1971, 1972), has concluded that Co2+ adsorbed onto
Mn oxides can be slowly oxidized to C0 3 +. The trivalent ion is incorporated
into surface layers of the crystal lattice. While other ions are adsorbed by Mn
oxides (for example Cu 2 + , Ni2+) only Co is incorporated into the surface layers
of the oxide. This specificity appears to be due to the high crystal field stabiliza-
tion energy of the low spin Co3+ ion (McKENzm 1970, 1972). Only C0 2 +
could be oxidiz~d by Mn3+ and only the low-spin Co3+ ion has a higher crystal
field stabilization energy than Mn3+.
The adsorption of anions by oxides and hydroxides of Fe and Al can enhance
the adsorption of cations. For example the adsorption ofH 2 PO';- by both amor-
phous hydrous oxides of Fe and Al and goethite enhanced the adsorption of
Zn (STANTON and DU BURGER 1967, 1970, BOLLAND et al. 1977) and resulted
in decreased absorption of Zn by plant roots (STANTON and DU BURGER 1967).
This phenomenon could arise through the formation of complexes between
Zn and H 2 PO';- on the surface of the oxides.

2.1.1.2 Precipitation
Interactions between ions can occur because the addition of one ion either
enhances the precipitation or dissolution of sparingly soluble compounds. Three
examples will be considered here: (1) the effect of soil acidity on the toxicity
of Al to plants (2) the effect of soil acidity on the availability of Mn to plants
and (3) the dissolution of calcium phosphates in alkaline soils.
1.5 Interactions Between Nutrients in Higher Plants 155

Aluminium toxicity may decrease the growth of plants on acid soils (see
reviews by Foy 1974 and HELYAR 1978). The application of compounds which
increase soil pH (CaC0 3, K zC0 3 ) can eliminate Al toxicity (for example see
MUNNS 1965a). In soils the solubility product of Al (OHh (PAI+3pOH) varied
from 32 for some freshly limed soils to 35 for other soils (MARION et al. 1976).
In solution the solubility products of gibbsite and amorphous AI(OH)3 were
34 and 32.3 respectively. At pH 5 the activity of AI in the soil solution would
therefore be 0.1 11M if gibbsite was the predominant form and 5.0 11M if amor-
phous AI(OHh predominated. The growth of cotton roots in both soil and
nutrient solutions was depressed when the activity of the AI ion in solution
exceeded 211M (ADAMS and LUND 1966). Hence while plants differ greatly in
their tolerance of high concentrations of Al in solution (see review by Foy
1974), it is likely that toxic concentrations of Al will only occur at soil pH
values around 5 and less.
Avoidance of Al toxicity by tolerant species is in many cases related to
maintenance of rhizosphere pH above 4.0 (FoY 1974). When anion absorption
exceeds cation absorption, the rhizosphere will become less acid (KIRKBY 1968).
Indeed differences among tropical pasture legumes in tolerance to Al toxicity
(ANDREW et al. 1973) appeared to be related to differences among them in
the excess of anion absorption over cation absorption (HELYAR 1978). HELYAR
further predicted that legumes dependent on symbiotic nitrogen fixation - where
rhizosphere pH falls (NYATSANGA and PIERRE 1973) - will be more susceptible
to Al toxicity than legumes supplied with NO; . The effect of high concentrations
of Al in solution on the growth of plants can be modified by both P and
Ca supply (MUNNS 1965a, b). These interactions, which may involve several
processes, are considered in Section 2.3.1.
The concentration of Mn2+ in the soil solution is dependent on the soil
pH and on its oxidation-reduction potential because of the reaction MnOz +
2H+~Mn2+ +~Oz+HzO. Manganese deficiency in plants occurs mainly on
alkaline soils and Mn toxicity occurs mainly on acid soils because plants probab-
ly only take up Mnz+ (HALSTEAD and BARBER 1968). Reactions occurring within
the rhizosphere may markedly affect the dissolution of insoluble and unavailable
Mn oxides. Firstly exudates from plant roots may solubilize amorphous Mn
oxides (JONES and LEEPER 1951). Secondly bacterial activity in the rhizosphere
may also affect the solubility of Mn oxides. Thus varieties of oats susceptible
to Mn deficiency can have more Mn-oxidizing bacteria around their roots than
those less susceptible (TIMONIN 1946). Thirdly the acidity of the rhizosphere
may vary particularly in relation to the form of N supplied. Plants fed NH:
(which leads to acidification of the rhizosphere) are less likely to be Mn-deficient
than those fed NO; (which leads to increased pH in the rhizosphere) (HoYT
and MYOVELLA 1979).
The availability of P for plants on neutral and alkaline soils may be con-
trolled by the dissolution and precipitation of sparingly soluble Ca phosphates
such as hydroxy-apatite (LARSEN 1967). For example if the concentration of
HzPO i in the soil solution is controlled by the solubility of hydroxy-apatite
[Ca s(P04hOH] the relationship 1O(pH--!pCa)-3(pH+pH zP04) must be
constant [pH ZP04 =constant+2pH+pP0 4 and' p(OH)=constant-pH].
156 A.D. ROBSON and M.G. PITMAN:

Hence it can be calculated that for hydroxy-apatite increasing the pH from

6 to 7 in the presence of 5 x 10- 3 M Ca will decrease the P concentration
in solution from 3.1 x 10- 3 M to 2.5 X 10- 7 M (RUSSELL 1973).
However, it is difficult to use solubility products to predict the concentrations
of H 2 PO; in solution (see discussions by LARSEN 1967, RUSSELL 1973). Firstly
solubility products predict concentrations at equilibrium which may be slowly
reached in soils. Secondly the solubility constant of hydroxy-apatite can depend
on whether the hydroxy-apatite is being precipitated or dissolved (BJERRUM
1949). Thirdly impurities in the hydroxy-apatite such as F and CaC0 3 may
modify its solubility (LARSEN 1967, RUSSELL 1973). Finally the neutral molecule
(CaHP0 4 )O may exist in the soil solution (LARSEN 1967) and further complicate
the use of solubility constants.
Notwithstanding these difficulties, solubility products of various Fe, Al and
Ca phosphates have been used to construct a unified solubility diagram (for
example LINDSAY and VLEK 1977) for P in soils with a wide range of soil
pH. The value of such an approach is not certain.

2.1.2 Absorption from Solution at the Root Surface

Interactions among ions can arise through the effect of one ion on the absorption
of another by plant roots from solution. Absorption of ions from solution
may be affected by other ions through effects due to (1) decreased or enhanced
access to the sites of absorption, to (2) competition at the site of absorption
and (3) interaction with metabolic controls of absorption (such as an ATPase
or an electrogenic H+ transport system). Each of these three types of interaction
that occur within the root will be considered in turn. Unfortunately in many
instances it is not possible to define which of these three mechanisms is involved
in the enhancement or depression of the absorption of one ion by another.

2.1.2.1 Access to the Sites of Absorption

One ion may affect the absorption of another ion by affecting its concentration
at the site of absorption in the plasmalemma. There are fixed negative charges
in the cell walls of plants associated with carboxylate groups on which cations
are adsorbed. While there is little relationship between cation adsorption and
cation absorption, and adsorption is not a necessary prerequisite to absorption
(see review by HAYNES 1980), adsorbed cations may influence the absorption
of both anions and cations.
For example, Ca concentrations in solution may markedly affect the rate
of absorption of both anions and other cations (see review by HAYNES 1980).
Two effects are probably involved. Firstly, low concentrations qf Ca in solution
(about 50 J.1M) are required to maintain the integrity of membranes (EpSTEIN
1962). This effect is specific for Ca (EpSTEIN 1962). Secondly, higher concentra-
tions of Ca in solution may depress cation absorption [for example Mn 2 +
(ROBSON and LONERAGAN 1970) and Zn 2 + (CHAUDHRY and LONERAGAN 1972a)]
and enhance anion absorption [for example, Br- (HOOYMANS 1964); SOi-
(FRANKLIN 1971) and H2POi (EDWARDS 1968, FRANKLIN 1969, 1970, ROBSON
1.5 Interactions Between Nutrients in Higher Plants 157

et al. 1970)], perhaps by affecting the access of these ions to the sites of absorp-
tion. Effects of Ca on anion absorption are established almost instantaneously
(ROBSON et al. 1970).
The effect of Ca is not specific. Trivalent cations enhance the absorption
of HzPO';:- more than divalent cations which have a larger effect than monova-
lent cations (FRANKLIN 1969, 1970, HYDE 1966). Calcium and other cations
may enhance anion absorption by contracting the electrical double layer asso-
ciated with fixed negative charges in the cell wall and thus allowing greater
access to sites of absorption. This effect may partly explain the observation
that CI- uptake by Beta tissue was increased by an increased ratio of Ca2 +
to K + in solutions of equal total cation concentrations (PITMAN 1964). Increas-
ing the solution pH generally increases the rate of absorption of cations and
decreases the rate of absorption of anions (see for example ISLAM et al. 1980).
These effects may at least partly reflect effects of H+ ions on the access of
ions to sites of absorption.
The effect of Ca on the absorption of anions and cations varies with plant
species. Increasing Ca concentration in solution depressed Mn absorption more
and enhanced P absorption more for Medicago truncatula than for Trifolium
subterraneum (ROBSON and LoNERAGAN 1970, ROBSON et al. 1970). Perhaps these
species differ in the distribution of negative charges in the pores of their cell
wall.
2.1.2.2 Competition at Site of Absorption
Ion absorption (i.e. movement across the plasmalemma) by plant roots is a
very selective process at low concentration in solution (see EpSTEIN 1972). For
the univalent cations Na + does not compete with K + for uptake at low concen-
trations «0.2 mM) in solution (see EpSTEIN 1972). However, ions more similar
to K + can compete with it and decrease its absorption. Both Rb + and NH:
can depress the absorption of K + by plant roots. Indeed the effect of NH:
on K + absorption from solution (TROMP 1962) may change markedly the rela-
tionship between potassium concentrations in solutions and growth (see SPEAR
et al. 1978a). For the alkaline earth cations only Sr2+ and Caz + compete in
their absorption from solution (EpSTEIN 1962).
Increases in the external K concentration often reduce the absorption of
Ca (LAZAROFF and PITMAN 1966, OSMOND 1966, JOHANSEN et al. 1968, SPEAR
et al. 1978b) and Mg (yOSHIDA 1966, LEGGETT and GILBERT 1969, FAGERIA
1974). In some instances long-distance transport of Ca is reduced at high concen-
trations of monovalent cations due to a reduction in transpiration rate (LAZAR-
OFF and PITMAN 1966). In other instances other explanations are required. Effects
of Ca on K absorption are complex. Apart from effects of Ca in maintaining
membrane integrity and affecting the access of ions to sites of absorption
(Sect. 2.1.2.1), Ca may affect K absorption through competition for absorption
for the low affinity mechanism operating at high concentrations of ions in
solution.
The absorption of the transition metal cations from concentrations of the
order of 1 JlM is strongly inhibited by the alkali and alkaline earth cations
(Fig. 2) and by other transition metal cations. The effect of alkali and alkaline
158 A.D. ROBSON and M.G. PITMAN:

400 Fig. 2. The effect of other cations on

>.
c " the absorption of zinc by wheat
.~ ~ roots. (CHAUDHRY and LONERAGAN
e-o ;;
- III

0
300 1972a). Zinc added as zinc chloride
III
.0
~ at a concentration of 1 J.lM
"u 3:
LL
c Ol 200
N . .III. .
o E
'" .8
"'0 " 100
a:: Ol
c

Calcium concentration

5 5

-
32
4 e_ 4
~
.c Ol 3 3
26
b.

'"~
;; 2 2
0
.s::.
(/)

a b
0 0
Manganese concentration Manganese concentration
in the media (pg/g) in the plant (pg/g)

Fig. 3. The effect of iron supply (0,0.005; /)., 0.5; and e, 3 J.lg Fe g-I) on the relationships
between a manganese supply and growth and b manganese concentrations in shoots
and growth for soybeans grown in solution. (SOMERS and SHIVE 1942). In b data are
iron concentrations in shoots (J.lg g -I dry wt.)

earth cations may not necessarily involve competition for sites of absorption
across the plasmalemma (see Sect. 2.1.2.2). Nevertheless there may be competi-
tion between some transition metal cations. The absorption of Zn2+ was
strongly inhibited by Cu 2 +, weakly by Co2+ and not at all by Mn 2 + or Fe 2 +
(HAWF and SCHMID 1967, BOWEN 1969, CHAUDHRY and LONERAGAN 1972b).
At low concentrations of Zn 2 + and Cu 2 +, and in the presence of Ca 2 + at
concentrations commonly present in soil solutions, the effect of Cu 2 + was com-
petitive, suggesting that it would be additive to that of Ca2+. Both Zn 2 + and
Mn 2 + inhibited Fe absorption from Fe 2 + by tomato roots (TIFFIN 1967) and
by tobacco leaf cells (KANNAN 1969). Iron added either as an inorganic salt
(SOMERS and SHIVE 1942) or as a chelate (MORAGHAN 1980) depressed Mn ab-
sorption by plants. Indeed the much-cited interaction between Fe and Mn in
plants may result from effects on absorption rather than antagonisms within
the plant (Fig. 3).
1.5 Interactions Between Nutrients in Higher Plants 159

For the anions the absorption of H 2 PO';:- by plant roots is not affected
by the concentrations of CI- (CARTER AND LATHWELL 1967), NO; (ALBERDA
1948), or SOi- (TANADA 1955) in the solution. Similarly the absorption of
NO; by low salt roots of barley was not affected by CI-, Br- or SOi - concen-
trations in solution (RAo and RAINS 1976). For the halides, CI- and Br-
compete for the same transport site where Br- replaces CI- with nearly the
same efficiency (ELZAM and EpSTEIN 1965, SMITH and Fox 1977). By contrast,
F- and 1- compete only weakly with Cl- (ELZAM and EpSTEIN 1965), possibly
due to interference with metabolism. In other plant systems however, there
is clear ability of I - to substitute for CI- (e.g. in beet root slices or certain
algae). The uptake of SOi- is unaffected by the concentration of other anions
(LEGGETT and EpSTEIN 1956) except SeOi- (EpSTEIN 1962). For Mooi-, absorp-
tion by tomato roots was strongly inhibited by SOi- (STOUT et al. 1951). A
particularly puzzling interaction is the marked stimulation of the absorption
of MoOi - by tomato roots associated with increasing the H 2 PO';:- concentration
in solution (STOUT et al. 1951). BARSHAD (1951) suggested that P may stimulate
Mo uptake because of the formation of a phosphomolybdate complex absorbed
more readily by the plant.

2.1.2.3 Metabolic Controls of Ion Absorption

Interactions between ions may be due to activation or deactivation of the driving
forces such as an ATPase or an electrogenic H+ transport system. Interaction
may also occur due to suppression of a transporting compound.
The deficiency or toxicity of one ion may affect the control of absorption
of another. For example Zn-deficient plants appear to be more susceptible to
P toxicity. Zinc-deficient plants may have higher rates of P absorption than
plants adequately supplied with Zn (PARIBOK and ALEKSEEVA-POPOVA 1965,
LONERAGAN et al. 1979). The mechanism of this effect is not understood.
Another example is the apparent lack of control of H 2 PO';:- absorption (leading
to concentration of H 2 PO';:- within the plant high enough to cause P toxicity)
in com plants grown in saline media (NIEMAN and CLARK 1976). High P concen-
trations may have accumulated within the plant because salinity impaired
growth but not P uptake.
Apart from metabolic changes that result from deficiency and toxicity, other
controls of the absorption of one ion may be modified by another. The activity
of anion transport may be regulated by vacuolar contents of the ion concerned,
by OH- jH+ in the cytoplasm or by products of metabolism such as organic
acids. For example the addition of NO; may decrease Cl- influx (and vice
versa) by carrot tissue (CRAM 1973), excised barley roots (SMITH 1973) and
citrus leaf slices (SMITH and Fox 1977). However in wheat roots ther:e is an
inducible NO; uptake system that is separate from the CI- uptake process
(JACKSON et al. 1973). Similarly Cl- did not affect the uptake of NO; by low
salt roots of barley (RAo and RAINS 1976). It is therefore likely that the interac-
tion between Cl- and NO; is related to some general aspect of ion absorption.
High intra-cellular NO; concentrations greatly decreased Cl- influx (and
vice versa), suggesting that the internal Cl- plus NO; concentration regulated
160 A.D. ROBSON and M.G. PITMAN:

the influx of these ions (CRAM 1973, SMITH 1973, SMITH and Fox 1977). Organic
anions within roots (for example malate) did not affect the influx of either
NO; or CI- (CRAM 1973, SMITH 1973). It is therefore necessary to distinguish
between situations where reduced NO; can be removed from the system, as
in transfer from roots to shoots, and studies where the vacuole was a sink
for CI- and NO;. Additionally it may be important to differentiate between
species where NO; is reduced in roots and those in which NO; is reduced
in shoots.
For avocados grown in soil, the N concentration in leaves was unaffected
by the addition of 20 mM NaCl which produced symptoms of CI- toxicity
and reduced growth by nearly 60% (DOWNTON 1978). Chloride concentrations
in leaves increased more than tenfold. Even though the ratio of CI- to NO;
added in solution was about 100: 1 in saline soils there was effective discrimina-
tion in uptake between NO; and Cl-.
Another source of interaction among ions occurs due to the regulation of
cation/anion balance in cells. This process is discussed in detail by SMITH and
RAVEN (1976). Uptake of more equivalents of N as NO; than equivalents
of cation must be accomplished by release of OH - to the solution while uptake
of N as NH; or of more K + than NO; results in H+ excretion. Changes
in rhizosphere pH can affect ion uptake-both by affecting the concentration
of ions in solution (see Sect. 2.1.1) and by affecting the absorption of ions
from solution by plant roots (see Sect. 2.1.2.1).
When cation absorption exceeds anion absorption organic acids may be
synthesized. Synthesis of organic acids is also the regular way of eliminating
OH - formed in reduction of NO; or SOi -. Organic acids are important in
the transport of some divalent cations across the root. In particular citrate
appears to chelate Fe and possibly Zn (TIFFIN 1967). There have been various
observations that organic acid synthesis is affected by deficiencies. DEKOCK
and MORRISON (1958) found that Fe-induced chlorosis reduced concentrations
of malic and citric acids by 50%. By contrast, chlorosis associated with Ni toxi-
city was accompanied by increased citrate and decreased malate concentrations.
DEKOCK has suggested that the ratio of citrate to malate is related to the balance
of P to Fe and K to Ca, but there are other examples where the ratio of
citrate to malate changed, that could not be associated with these particular
ratios (see HEWITT 1963). The increase in citrate could be a response to low
Fe supply to facilitate transport in the plant.

2.2 Interactions Between Nutrients Affecting the Utilization

of Nutrients Within the Plant

Interactions involving effects on nutrient utilization within the plant may involve
either of two distinct components. First the transport of the nutrient (in either
xylem or phloem) to the site of function may be impaired. Second at this site
the function of the nutrient may be impaired or enhanced. Effects of one ion
on the utilization of another is usually assessed by considering its effect on
the relationship between nutrient concentration in whole shoots and growth
I.5 Interactions Between Nutrients in Higher Plants 161

(see BATES 1971). Changed relationships may, however, reflect either of the
two components of nutrient utilization. Antagonism or synergism between nu-
trients in their function can only be assessed if concentrations at the site of
function (functional nutrient requirement; LONERAGAN 1968) can be assessed.
In this section of this review we will consider interactions that are due to effects
on the distribution and function of nutrients.

2.2.1 Distribution

One nutrient may affect the distribution of another in at least three distinct
ways. Firstly, transport in either xylem or phloem may be impaired by precipita-
tion. Secondly, nutrients may not be able to enter the phloem for retranslocation
from old to young leaves because they are immobilized in old leaves. Finally,
one nutrient may modify the distribution of another within the leaf of the
plant. Each of these types of interaction will be considered.

2.2.1.1 Precipitation Within the Plant

High concentrations of P in solution may increase the incidence of chlorosis
associated with Fe deficiency. While these effects may partly relate to effects
of P on the concentrations of Fe available for absorption, there are also interac-
tions between Fe and P within the plant. OLSEN (1935) first suggested that
when Fe is taken up from neutral or alkaline solutions it can be precipitated
as ferric phosphate in the vascular bundles along the veins of a leaf. Autoradio-
graphs of leaves of plants grown with an initial external H 2 PO';- concentration
of 100 11M and an initial Fe concentration (as labelled FeCI 3 ) of 18 11M showed
a heavy deposition of Fe along the veins, while the rest of the leaf blade con-
tained little Fe (REDISKE and BIDDULPH 1953).
Under certain conditions it appears likely that immobilization of Fe, perhaps
as Fe phosphates, can occur in the vascular system to such an extent that
the adjacent tissues become markedly deficient in Fe. In some species Fe is
transported in the xylem as an anion complexed with citrate (TIFFIN 1967).
Soybean varieties susceptible to Fe chlorosis appeared less able to keep the
Fe in their tissues mobile than tolerant varieties when supplied with high H 2 PO';-
concentrations (BROWN et al. 1959). Perhaps varieties differ in the concentra-
tions of citrate available to compete with H 2 PO';- for the Fe.
The formation of precipitates may also limit the transport of ions within
the phloem. When Mn and Ca chlorides were added to phloem sap from Yucca
and Ricinus, precipitation occurred when the Ca concentration in the sap ex-
ceeded 0.2-0.5 mM (depending on species and CalK ratio in growth'media)
and when the Mn concentration exceeded 0.004-0.18 mM (depending on species)
(VAN GOOR and WIERSMA 1974). By contrast, precipitation did not occur until
much higher concentrations of K and Mg chlorides were added. V AN GOOR
and WIERSMA did not identify the compounds precipitated but considered that
Ca phosphates would be precipitated when Ca concentrations in sap exceeded
0.6mM.
162 A.D. ROBSON and M.G. PITMAN:

2.2.1.2 Immobilization in Old Leaves

One nutrient may affect the retranslocation of another from old to young leaves.
When plants are grown with an ample N supply, S deficiency leads to chlorosis
of new leaves (ERGLE 1954). When N supply is low or when plants are relying
on symbiotic activity for the N supply, symptoms ofS deficiency may be indistin-
guishable from N deficiency, developing most strongly in older leaves (ANDER-
SON and SPENCER 1950a, EATON 1966). The effect of N supply is probably
due to an effect on the rate of hydrolysis of proteins. Sulphur present as SOi-
in old leaves moves readily to new growth. However, when cotton plants were
transferred from plus-S to minus-S solutions old leaves lost none of their protein-
S or soluble organic S when amply supplied with N (ERGLE 1954).
Similarly, N supply may influence the retranslocation of Cu from old to
new leaves. The movement of Cu out of old leaves of wheat closely paralleled
the movement of N (Fig. 4; LONERAGAN et al. 1976, 1980b, HILL et al. 1978,
1979b). Copper deficiency, which delayed the retranslocation of Cu from the
oldest leaf of wheat, also delayed the retranslocation of N and leaf senescence.
Moreover when the oldest leaves of Cu-deficient and Cu-sufficient plants were
shaded and induced to senesce, their Cu and N contents declined simultaneously
(HILL et al. 1979a). In Cu-adequate plants Nand Cu concentrations in the
oldest leaf of wheat decreased slowly until the leaf senesced (HILL et al. 1979 b).
After senescence Nand Cu concentrations fell rapidly. In Cu-deficient plants
which did not senesee, Cu and N concentrations decreased slowly. KANNANGARA
and WOOLHOUSE (1968) have concluded that the initial gradual decline in total
protein content after full expansion ofleaves of Perlllafrutescens was associated
with a decline in the content of Fraction 1 proteins, whereas the Fraction 2
proteins declined only during the sharp decline in total protein content, just
prior to abscission. The strong binding of Cu by nitrogen ligands may thus
be involved in the relationship of Cu movement out of old leaves to N movement
out of these leaves.
In the study of HILL et al. (1979b), the movement of Zn out of the oldest
leaf of wheat supplied with adequate Zn also paralleled the movement of N
out of these leaves. The Zn content of these leaves did not decline until the
leaf senesced.

2.2.1.3 Distribution of Nutrients Within the Leaf

Increasing the supply of Si can alleviate Mn toxicity in cereals without affecting

the uptake of Mn by the plant or the concentration of Mn in the entire leaf
(Fig. 5; WILLIAMS and VLAMIS 1957, VLAMIS and WILLIAMS 1967). With high
Mn supply, necrotic, brown spots occurred on the old leaves: The addition
of sodium silicate but not sodium sulphate eliminated these symptoms. Autora-
diography using 54Mn indicated that the brown spots were localized high con-
centrations of Mn. When Si was added the 54Mn was distributed uniformly
throughout the leaf. These effects of Si on Mn toxicity and the distribution
of Mn within the leaf were also observed for a legume (HORST and MARSCHNER
1978 a) even though it is likely that concentrations of Si in the shoots of legumes
1.5 Interactions Between Nutrients in Higher Plants 163

._1200
C
CUo

~
3. 800
C
~
c 400
0
u
z a
0

..
160

C
~ 120
en
c

c 80
~
c
0
U
:l 40
u
b
0

.
'0
~
en
800
E
C
C0 400
C1>

u
C C
N
00
Days from sowing

Fig. 4. The effect of copper supply on the changes in contents of a nitrogen (total nitrogen,
0; alcohol insoluble nitrogen .), b copper and c zinc in the oldest leaf of wheat (HILL
et a1. 1979b). Chlorophyll contents for copper-adequate plants (CU 1600) fell sharply after
day 28, whereas those of copper-deficient plants (Cuo) did not. At each harvest the mean
and the limits of two replicates are shown

.Si
30
en
Q
III
25
Ih~--- __
00
~
20
III
I ------0----------------------0
'0 -Si
.:cen 15
'OJ
3: 10
Fig. 5. The effect of sili-
~
con (0 no silicon, D. sili- Cl 5
con added) in alleviating
manganese toxicity in 0
wheat. (VLAMIS and WIL-
LIAMS 1967) Manganese c:oncentration
164 A.D. ROBSON and M.G. PITMAN:

would be much lower than in the shoots of cereals (see JONES and HANDRECK
1967).
The mechanism by which Si alters the distribution of Mn within the leaf
is not understood. HORST and MARSCHNER (1978b) consider that Si facilitates
the movement of Mn from the xylem vessels into the surrounding tissue and
thereby prevents local accumulation of Mn near the vessels. The addition of
Si to culture solutions increased the proportion of Mn in shoots of ryegrass
associated with the cell walls (JARVIS and JONES, personal communication). Sili-
con may also affect the distribution of Mn within the cell. The addition of
Si increased the Mn concentration in the press sap from bean leaves suggesting
that Si may favour the transport of Mn. into the vacuoles (HORST and
MARSCHNER 1978b). Direct interactions between Si and Mn in the press sap
could not be detected by thin layer chromatography or electrophoresis (HORST
1976). However, Si may affect the form of Mn within the plant. The addition
of Si to the culture solution doubled the amount of Mn extracted from the
leaves of bean plants by methanol but did not affect the amount of Mn extracted
by water, KCI or EDTA (HORST and MARSCHNER 1978c). Water extracted
80% of the total Mn content of leaves, whereas methanol extracted less than
10%.
There may be interactions between Si< and Mn on plant growth which are
not related to effects of Si on Mn distribution within leaves. In studies with
Sudan grass, increasing the Si in solution decreased Mn uptake and either in-
duced Mn deficiency or alleviated Mn toxicity depending on the level of Mn
supply (BOWEN 1972). Similarly, the application of silicic acid to an acid soil
depressed the concentration ofMn in leaves and total uptake ofMn by tomatoes
(PEASLEE and FRINK 1969).

2.2.2 Function
Interactions between nutrients may occur because (1) one nutrient competes
with another at the site of its function for incorporation into active sites, (2)
one nutrient can substitute for another and (3) one nutrient is required for
the assimilation or metabolism of another.

2.2.2.1 Competition at Site of Action

Although antagonisms at the site of function are often invoked to explain inter-
actions between nutrients, there are few examples which clearly demonstrate
this phenomenon in vivo. To demonstrate such an antagonism one ion must
inhibit a physiological process or a metabolic reaction at low but not at high
concentrations of the nutrient at the site of its action. Otherwise the antagonism
affecting plant growth may result from effects either on uptake or distribution
within the plant.
An example of an antagonism is the interaction between WO~ - and MoO~ - .
Tungstate can inhibit the activity of NO; reductase in higher plants (HEIMER
et al. 1969, NOTTON and HEWITT 1971, RAO and RAINS 1976, HARPER and NI-
CHOLAS 1978). When Mo is replaced by W an analogue of NO; reductase
1.5 Interactions Between Nutrients in Higher Plants 165

Fig. 6. The effect of tungstate on the growth

~ 800
of Alnus glutinosa supplied with either nitrate o
(~) or urea (0). (PrzELLE and THIERY 1979)
Ol
E 600
~
o
o
.r:.
lfl

L
~ 200
3:

o
0.1 10
NQZW04 concentration (mM)

is formed (NOTTON and HEWITT 1971). This enzyme, while still displaying dehy-
drogenase activity, is inactive in NO~ reduction. Molybdenum and W were
not interchangeable in vivo.
The effect of this interaction on plant growth depends both on the source
of N and the ratio of WO~ - to MoO~ - in the external media. Tungstate to
MoO~ - ratios of 1,500: 1 did not decrease growth of white clover reliant either
on symbiotically fixed N or NO~ (QUIN and HOGLUND 1976). However, WO~­
inhibited the growth of shoots and roots of Alnus glutinosa (Fig. 6; PIZELLE
and THIERY 1979); the effect was more marked when the source of N was
NO~ than when it was urea. For Alnus, inhibiting NO~ reduction with WO~­
increased nodule number without affecting the weight of each nodule. Hence
with NO~ -fed plants, acetylene reduction activity per plant but not per g nodule
was increased by adding 0.01 mM Na zW0 4 . Further increasing the concentra-
tion of Na zW0 4 to 0.1 mM depressed acetylene reduction activity perhaps due
to incorporation of W in place of Mo in nitrogenase.
The effects of WO~ - on plant growth may also result from competition
between WO~ - and MoO~ - for absorption, as well as for incorporation into
NO~ reductase or nitrogenase. In many studies with WOi-, Mo concentrations
within the plant are not presented.

2.2.2.2 Sparing Effect

Sodium can substitute for part of the requirement for K for cereals (WEHUNT
and COLLINS 1953, DE WIT et al. 1963), grasses (GAMMON 1953, HYLTON et al.
1967, SMITH 1974) and dicotyledons (HOLT and YOLK 1945). The addition of
Na to soil can decrease both the external and internal requirements for K
(for example SMITH 1974), indicating that some species can use Na in place
of K for some but not all physiological or metabolic functions. Plant growth
may decrease if a balance between inorganic cations and anions is not main-
tained (DE WIT et al. 1963). Sodium may have a sparing effect on K requirements
by contributing to the cation content of the cell vacuole in place of K.
Some plant species have a specific requirement for Na (BROWNELL and WOOD
1957, BROWNELL 1965). Increasing the K supply did not eliminate effects of
Na on the growth of several species of Atriplex (BROWNELL 1968). Species
166 A.D. ROBSON and M.G. PITMAN:

with the C4 dicarboxylic acid pathways for CO 2 generally have a requirement

for Na (BROWNELL and CROSSLAND 1972).

2.2.2.3 Involvement of One Nutrient in the Assimilation and Metabolism

of Another
Three examples will be considered here to demonstrate this type of interaction.
The form of N supplied can influence the response in growth by plants
to nutrients. For example the growth of tobacco, rice and soybean cell cultures
was enhanced by Ni when urea was the source ofN (POLACIO 1977). Substituting
NH: for urea eliminated any requirement for Ni. Nickel is a component of
jackbean urease (DIXON et al. 1975) which hydrolyzes urea. Another example
is the greater effect of Mo on algal and plant growth when NO~ rather than
NH: is the source of N (ISHIOKA and ARNON 1955, HEWITT and MCCREADY
1956). Molybdenum is a constituent of nitrate reductase (NICHOLAS and NASON
1955).
The third example relates to the involvement of mineral nutrients in sym-
biotic N fixation by both legumes and non-legumes. Mineral nutrients may
limit symbiotic N fixation by legumes by affecting growth and survival of rhizo-
bia, infection and nodule development and nodule function as well as by affect-

0} VI

VI
10 00 100
0 .<=-
0 VI ~ C
.<= ~ :;J
VI .<=~
O}-
50
:c ._ :;J
., u
.....
.,
0}
~
0}

~ a .<=-
.,
VI
>- 0 0.1
0
~ 150 U: 0
Molybdenum applied Cobalt added (,..moles /I-IJ
(flg sodium molybdate/pot-I)

,
0a. VI
0}
10 00 2
.<=-
VI 1Il"L
00 ~c

.<= 5 .<=
O}_
c
VI 0a; a.
:c ~ 0}

.,
0} .<=
.,
VI d
~ U: 0
Copper applied (flg / pot-I)

Fig. 7. The interaction between the application of nitrogen (0, no nitrogen; e, nitrogen
applied) and a molybdenum (ANDERSON and SPENCER 1950a), b calcium (LONERAGAN
1959), c cobalt (DELWICHE et al. 1961) and d copper (SNOWBALL et al. 1980) on the
growth of legumes. For calcium there is a positive interaction for severely calcium-defi-
cient plants; for calcium when moderately deficient and for the other nutrients the interac-
tion is negative
1.5 Interactions Between Nutrients in Higher Plants 167

ing host plant growth. For particular nutrients where the requirements (either
external or internal) for symbiotic N fixation are greater than for growth of
the host legume, effects of nutrient deficiency on growth can be alleviated by
supplying mineral N (see ROBSON 1978). There are several examples. Firstly
the growth of legumes responds to Co only when reliant on symbiotically fixed
N (DELWICHE et al. 1961). For other nutrients - Mo (ANDERSON and SPENCER
1950b), Ca (LONERAGAN 1959) and Cu (SNOWBALL et al. 1980) - the interaction
between nutrient supply and N on growth may also be negative (i.e. correct
the same deficiency) but less complete than that for Co (Fig. 7). For all other
nutrients the interaction between their supply and N on the growth of legumes
is positive (see summary in ROBSON 1978), indicating that host plant growth
has a higher requirement than symbiotic N fixation for these nutrients.
For Mo, Ca and Cu, nodule function appears to be the component of the
symbiosis which has the highest requirement. Molybdenum is a component
of nitrogenase (see review by DILWORTH 1974). The role of Cu and Ca in
nodule function is less clear (see review by ROBSON 1978).

2.3 Complex Interactions Between Nutrients Involving Several Processes

Some interactions between nutrients involve effects of one nutrient on both

the absorption and utilization of another nutrient. These interactions may
operate through effects both in the external media and within the plant. Two
such interactions will be considered.

2.3.1 Calcium/Aluminium/Phosphate
Plants grown in external media with high concentrations of Al may display
symptoms associated with either P deficiency (stunting, small dark green leaves,
purpling of stems and leaves) or Ca deficiency (cupping or rolling of young
leaves). The interactions between Al and P and between Al and Ca arise through
effects on both absorption and utilization.
The growth of roots is markedly affected by Al concentrations in solution
(see review by JACKSON 1967). At low concentrations of Al in solution the
primary root may elongate normally but lateral growth is inhibited. At severely
toxic concentrations of Al even primary roots are stunted and thickened. The
effects of Al on the uptake of nutrients including Ca and P may reflect effects
on root growth as well as interactions at the site of absorption.
In soil H 2 PO';:- is adsorbed onto the surfaces of Al oxides and may also
be precipitated as Al phosphates (see RUSSELL 1973). Both the application of
CaC0 3 and H 2 PO';:- to soil can decrease Al concentrations in solution and
alleviate Al toxicity by precipitating either Al hydroxides or Al phosphates
(Fig. 8; MUNNS 1965c). Effects of H 2 PO';:- in alleviating Al toxicity probably
result from these reactions in the external media rather than from effects within
the plant. Under conditions where precipitation did not occur, increasing the
H 2 PO';:- concentration in solution from 1 to 20 11M (both adequate concentra-
tions) did not alleviate Al toxicity (MUNNS 1965b). Moreover, high concentra-
168 A.D. ROBSON and M.G. PITMAN:

7
"0
a. 4.0 i 10000
3
0
3-
01 tJ 1000
E ~x
III CI>
a.

-
.8 .!:
0
-0
C;;
~
0
6i
.3
Phosphate added (m moles P pot-I)

Fig. 8. The interaction between the application ofCaC0 3 (e, nil; 0, 15 mmol; l>., 30 mmol)
and KH zP0 4 on a the growth of Medicago sativa on an acid soil and b the concentration
of aluminium in CaCl z extracts. (MUNNS 1965c)

tions of Al in solution can depress plant growth even though there are nomially
adequate concentrations of P in tops (MUNNS 1965b, ANDREW et al. 1973).
Phosphate and Al can form neutral aluminophosphate complexes (WHITE
et al. 1976). Solution pH markedly affects this reaction. The optimum pH for
the formation and polymerization of aluminophosphate complexes in dilute
solutions was about 5 and no complexed P was detected in solution at pH 4.3
or less.
In solutions where Al and H2P0.i would not precipitate, increasing the
concentration of Al in solution to 20 JlM depressed the growth of lucerne more
at pH 4.5 than at pH 5 particularly at high H2P0.i concentrations (20 JlM)
(WHITE 1976). WHITE (1976) suggested that the greater accumulation yet lower
toxicity of Al in plants grown at pH 5 may have been due to passive influx
and transport of the uncharged aluminophosphate polymers into vacuoles.
There are also other effects of Al on P uptake by plant roots. Pre-treatment
of roots with AI2(S04h increased the amount of P in roots (CLARKSON 1966).
However, the accumulation of this additional P, which was readily exchangeable
with non-radioactive P, was unaffected by dinitrophenol and low temperatures.
CLARKSON (1966) suggested that reactions between Al and P occurred in the
free space of the root external to the plasmalemma. Aluminium and P were
also co-precipitated in the same zones of roots of maize grown in nutrient
solutions containing 3,000 JlM Al (RASMUSSEN 1968). The precipitation of Al
in the free space of the root as either hydroxide or phosphate may regulate
the transport of Al and in some instances P to shoots.
Plant species tolerant of Al toxicity differ from sensitive species in the effects
of Al on P transport from roots to shoots. High Al concentrations in solution
markedly depressed P concentrations in the shoots of a sensitive species, Medi-
cago sativa, but did not affect P concentrations in the shoots of a tolerant
species, Trifolium subterraneum (MUNNS 1965b). Similarly for several tropical
and temperate legumes, increasing Al concentrations in solution decreased P
transport from roots to shoots considerably more in sensitive than in tolerant
species (ANDREW and VANDENBERG 1973). .
1.5 Interactions Between Nutrients in Higher Plants 169

Aluminium can also interfere with the metabolism of P within the plant.
Aluminium inhibits the activity ofhexokinases (RORISON 1965, CLARKSON 1966),
acid phosphatases (WOOLHOUSE 1969) and ATPases (WOOLHOUSE 1969) leading
to reduced incorporation of P into phosphorylated hexose sugars (CLARKSON
1966).
The interactions between Al and Ca appear to be less complex than those
between Al and P. Increasing the Ca concentration in solution can often mitigate
the effects oflow concentrations of Al on plant growth (see for example MUNNS
1965b, JACKSON 1967). Part of this effect can be attributed to effects of increas-
ing the Ca concentration in solution in reducing the activity but not the concen-
tration of Al ions in solution (see HELYAR 1978). Increased Ca supply can
also depress Al uptake by plants (MUNNS 1965b). Aluminium can also decrease
Ca uptake (see review by JACKSON 1967) perhaps through effects on root growth
as well as interactions at the root surface. The marked effects of CaC0 3 in
alleviating Al toxicity are not however associated in many instances with effects
of Ca on Al toxicity because K 2 C0 3 may be equally effective (see for example
MUNNS 1965a).

2.3.2 Zinc/Phosphate

The interactions between Zn and P have been extensively studied with several
mechanisms proposed to explain the effect of P application in inducing Zn
deficiency. The effects of P fertilizers on plant response may involve a number
of discrete phenomena. The importance of a particular phenomenon will depend
upon plant species and environmental conditions. It is likely to be simplistic
to consider that all interactions between P and Zn result from a single phenome-
non.
Interactions between P and Zn can be divided into three major categories.
First, there are interactions because of effects of P and Zn on plant growth.
Second, there are interactions between P and Zn which result from both effects
of P on Zn uptake and Zn on P uptake. Finally there may be interactions
which result from interactions between P and Zn within the plant.
Phosphorus applications which increase plant growth may decrease Zn con-
centrations in tops to deficient levels by dilution (BOAWN et al. 1954, LAMBERT
et al. 1979) without decreasing Zn uptake. Phosphorus applications which
correct P deficiencies may also decrease the proportion of total plant weight
which is roots (for example LONERAGAN and ASHER 1967). Zinc deficiency may
induce P toxicity by decreasing growth without decreasing the accumulation
of P (LONERAGAN et al. 1979). Each of these three effects of either P or Zn
on plant growth may lead to effects of high P applications in "inducing" symp-
toms attributed to Zn deficiency.
A complicating factor in the study of interactions between P and Zn is
the high concentrations of Zn in some phosphatic fertilizers. Superphosphate
contains 250-750 J.Lg g-l (WILLIAMS 1977), sufficient to alleviate Zn deficiency,
and leading to negative interactions between the application of superphosphate
and Zn in some instances (ANDERSON 1946).
170 A.D. ROBSON and M.O. PITMAN:

Phosphate applications may also reduce Zn contents and concentrations

within plants. Phosphate enhances Zn adsorption by oxides and hydroxides
of Fe and Al and decreases Zn absorption by plant roots (see Sect. 2.1.1.1).
It is unlikely that the precipitation of Zn phosphates is involved in phosphate-
induced Zn deficiency because Zn phosphate is a readily available source of
Zn for plant growth (BOAWN et al. 1957).
The application of phosphates to soil may decrease Zn uptake by plants
due to effects of the associated cations in decreasing Zn absorption (CHAUDHRY
and LONERAGAN 1972a). Another mechanism by which P applications may de-
crease Zn absorption is through effects of P supply in decreasing the develop-
ment of vesicular arbuscular mycorrhizas (LAMBERT et al. 1979). Increasing the
rate of P application (as CaH 2 P0 4 ) from 75 to 200 ppm in the media, which
increased the growth of non-mycorrhizal corn and soybean but did not affect
the growth of mycorrhizal plants, decreased the percentage of root volume
infected by mycorrhizal fungi. This increase in P supply decreased Zn uptake
more for plants inoculated with mycorrhizal fungi than for non-mycorrhizal
plants. Increasing P supply to plants may decrease both percentage infection
and the weight of mycorrhizal roots (ABBOTT and ROBSON 1977). Mycorrhizal
roots absorb more Zn than non-mycorrhizal roots both from solution (BOWEN
et al. 1974) and from soil (GILMORE 1971, PAIRUNAN et al. 1980).
As well as effects of P application on Zn uptake, Zn deficiency may affect
P absorption. For sugar beets (ROSELL and ULRICH 1964), peas (PARIBOK and
ALEKSEEVA-POPOVA 1965) and for okra (Hibiscus esculentus) (LONERAGAN et al.
1979) Zn deficiency enhanced the accumulation of P by leaves. Coupled with
the reduced growth due to Zn deficiency, this effect may lead to marked P
toxicity in Zn-deficient plants (LONERAGAN et al. 1979).
Interactions between Zn and P within the plant (for example MILLIKAN
1951, 1963, BOAWN and LEGGETT 1964) have often been invoked to explain
the effects ofP supply on Zn deficiency. Effects ofP supply on Zn concentrations
in whole tops were poorly related to effects on symptoms and growth. Apart
from the possible involvement of P toxicity in these experiments, Zn concentra-
tions in whole tops may be a poor guide to the Zn status of plants (REUTER
1980). Increasing the P supply from that deficient for plant growth to that
toxic for plant growth did not affect the critical concentrations of Zn in young
tissues of subterranean clover (REUTER 1980).

3 Interactions Between Nutrients in Mixed Communities

Interactions between nutrients in mixed communities are more complex than

those in monoculture because of genetic variation in absorption and utilization
of nutrients. Additionally, the ability oflegumes to symbiotically fix N produces
marked interactions between nutrients in grass-legume swards.
Fertilizer application may change the bptanical composition of a legume-
grass sward in several ways. A limiting N supply may prevent grasses but not
I.5 Interactions Between Nutrients in Higher Plants 171

7000

:- 6000
'0
.r;

'if 5000
"0
OJ 4000
>-
....
Q) 3000
"0
~ 2000
o
1000

o a b c
o
Applied Phosphorus (kg/hal
Fig. 9. The effect of phosphate supply on the growth of monocultures of a rye grass
b subterranean clover and c a mixed sward of rye grass and subterranean clover. (OZANNE
et al. 1976). Arrows indicate the amount of phosphate required for 90% of maximum
yield

legumes from responding to applications of other nutrients. Fertilizers may

thus initially increase legume growth only. Subsequently the increased N supply
due to N fixation may enable grasses to respond to such an extent that legume
growth is reduced by shading. Hence the effects of fertilizer on the botanical
composition of a pasture can change markedly with time. An example is the
effect of P fertilizers. Continued P applications to pastures generally leads to
non-legume (usually grass) dominance (for example see TRUMBLE and DONALD
1938 and review by ROSSITER 1966). The initial application of P may, however,
increase the proportion oflegume in the pasture (WILLOUGHBY 1954, WILLIAMS
et al. 1956). Grass dominance with continued P application probably reflects
an improved N supply associated with an initial response by legumes to P
(see review by LONERAGAN 1964).
Grass grown with clover responded to a much higher level of applied P
than when it was grown as a single species with high rates of NH 4 N0 3 (Fig. 9;
OZANNE et al. 1976). The response of clover in the mixed sward to applied
P was very similar to its response in a pure sward. The effect of clover on
the response of grass to applied P was attributed to increased availability of
N from symbiotically fixed N due to effects of P in stimulating clover growth.
Potassium-deficient pastures are generally grass-dominant, and K applica-
tion markedly increases the legume content (for example ROSSITER 1947, FITZPA-
TRICK and DUNNE 1956, JONES 1966). In monocultures the growth of most
grasses appears less affected by low K supply than the growth of legumes and
other herbs (ASHER and OZANNE 1967). Both when growing singly (ASHER 1964,
JACKMAN 1965) and when growing in association with a legume (WHELAN and
172 A.D. ROBSON and M.O. PITMAN:

Table 3. The influence of an associated grass species on the re-

sponse of Macroptilium atropurpureum to equivalent amounts of
either sodium or potassium sulphate (HALL 1978)

Treatment Mono- With Chloris With Setaria

culture gay ana anceps

Yield (g plant - 1)

Nil 1.22 1.09 1.21

+Na 2 S0 4 1.20 1.49 1.21
+K 2 S0 4 .1.53 1.76 1.79

EDWARDS 1975) grasses absorbed K at a faster rate per g roots than did legumes.
When a tropical legume and tropical grass were grown with both their tops
and roots competing, the growth of the legume was markedly depressed by
the grass at low but not at high K supply (HALL 1971). When only the tops
were competing, the competitive abilities of the grass and legume were unaf-
fected by K supply. The differential effects of K on the growth of legume
and grasses may lead to complex interactions between K and N on the growth
of mixed swards.
HALL (1978) has examined the effect of Na and K supply on the growth
of a legume either in monoculture or in association with either of two grass
species which differ in their ability to use Na in cation balance. Potassium
application increased the growth of the legume in all three situations (Table 3).
Sodium application did not increase the growth of the legume in monoculture
or in association with a grass unable to use Na in cation balance. However,
when grown in association with a grass able to substitute Na for K for some
functions, Na application increased its growth.
Sulphur-deficient pastures are generally also grass-dominant (for example
CONRAD 1950, HILDER and SPENCER 1954, WALKER et al. 1956). Sulphur applica-
tion to these pastures stimulated legume growth and increased the contribution
of the legume to total production. In all these situations response of the grass
but not the legume may have been limited by N deficiency. In monoculture
there is little evidence for differences between grasses and legumes in response
to S (see review by ROBSON and LONERAGAN 1978).
The effect of Cu and Zn application on botanical composition and growth
of pastures sown on newly cleared land (ANDERSON 1946) provides another
example. Both the application of Cu and Zn increased total yield. When neither
nutrient was supplied Phalaris tuberosa was the dominant species. Addition
of Cu alone gave dominance of Medicago sativa, whereas addition of Zn alone
increased the competitive advantage of P. tuberosa. Addition of both nutrients
increased the growth of P. tuberosa less than the growth of M. sativa so that
M. sativa became dominant. It is not clear whether these responses are produced
by differences in the absorption or the utilization of Cu and Zn by these species.
I.5 Interactions Between Nutrients in Higher Plants 173

4 Conclusion

Interactions between nutrients can arise through several distinct mechanisms;

the importance of a particular mechanism will depend upon environmental con-
ditions. It is important in the study of interactions to identify whether an interac-
tion occurs due to effects on growth or interactions in the media, or to interac-
tions within the plant. Moreover interactions occurring within the plant may
occur due to effects of one nutrient on the absorption, transport or function
of another. A major limitation in the study of interactions is the lack of knowl-
edge of the concentrations and forms of nutrients in different compartments
of the soil and the plant.

References

Abbott LK, Robson AD (1977) Growth stimulation of subterranean clover with vesicular
arbuscular mycorrhizas. Aust J Agric Res 28: 639-649
Adams F, Lund ZF (1966) Effect of chemical activity of soil solution aluminium on
cotton root penetration of acid subsoils. Soil Sci 101: 193-198
Adams SN, Honeysett JL (1964) Some effects of soil waterlogging on the cobalt and
copper status of pasture plants grown in pots. Aust J Agric Res 15:357-367
Adams SN, Honeysett JL, Tiller KG, Norrish K (1969) Factors controlling the increase
of cobalt in plants following the addition of a cobalt fertilizer. Aust J Soil Res 7: 29-42
Alberda T (1948) The influence of some external factors on growth and phosphate uptake
of maize plants of different salt conditions. Rec Trav Bot Neerl41: 541-561
Anderson AJ (1946) Fertilizers in pasture development on peat soils in the lower south
east of South Australia. J Counc Sci Ind Res Aust 19: 394-403
Anderson AJ (1956) Effects of fertilizer treatments on pasture growth. Proc 7th Int
Grassl Congr, pp 323-333
Anderson AJ, Oertel AC (1946) Plant responses to molybdenum as a fertilizer. Counc
Sci Ind Res Bull 198: 25-44
Anderson AJ, Spencer D (1950a) Molybdenum in nitrogen metabolism of legumes and
non-legumes. AustJ Sci Res Ser B 3:414-430
Anderson AJ, Spencer D (1950b) Sulphur in nitrogen metabolism of legumes and non-
legumes. Aust J Sci Res Ser B 3:431-449
Andrew CS, Vandenberg PJ (1973) The influence of aluminium on phosphate sorption
by whole plants and excised roots of some pasture legumes. Aust J Agric Res
24:341-351
Andrew CS, Johnson AD, Sandi and RL (1973) Effect of aluminium on the growth
and chemical composition of some tropical and temperate pasture legumes. Aust
J Agric Res 24:325-339
Asher CJ (1964) Potassium uptake by pasture plants as affected by root development
and the potassium concentration in the root environment. PhD thesis, Univ West
Aust
Asher CJ, Ozanne PG (1967) Growth and potassium content of plants in solution cultures
maintained at constant potassium concentration. Soil Sci 103: 155-161
Barrow NJ (1970) Comparison of the adsorption of molybdate, sulphate and phosphate
by soils. Soil Sci 109:282-289
Barshad I (1951) Factors affecting the molybdenum content of pasture plants. 2. Effect
of soluble phosphates, available nitrogen and soluble sulphates. Soil Sci 71: 387-398
174 A.D. ROBSON and M.G. PITMAN:

Bates TE (1971) Factors affecting critical nutrient concentrations in plants and their
evaluation - a review. Soil Sci 112: 116--130
Bjerrum N (1949) Selected papers. Munksgaard, Copenhagen
Boawn LC, Leggett GE (1964) Phosphorus and zinc concentrations in Russet Burbank
potato tissues in relation to development of zinc deficiency symptoms. Soil Sci Soc
Am Proc 28: 229-232
Boawn LC, Viets FG, Crawford CL (1954) Effect of phosphate fertilizer on zinc nutrition
of field beans. Soil Sci 78: 1-7
Boawn LC, Viets FG, Crawford CL (1957) Plant utilization of zinc from various types
of zinc compounds and fertilizer materials. Soil Sci 83: 219-227
Bolland MDA, Posner AM, Quirk JP (1977) Zinc adsorption by goethite in the absence
and presence of phosphate. Aust J Soil Res 15:279-286
Bollard EG (1960) Transport in the xylem. Annu Rev Plant Physipl11: 141-166
Bowden JW, Bolland MDA, Posner AM, Quirk JP (1973) Generalized model for anion
and cation adsorption at oxide surfaces. Nature Phys Sci 245:81-83
Bowen GD, Skinner MF, Bevege DI (1974) Zinc uptake by mycorrhizal and uninfected
roots of Pinus radiata and Araucaria cunninghamii. Soil BioI Biochem 6: 141-144
Bowen JE (1969) Absorption of copper, zinc and manganese by sugar cane tissue. Plant
PhysioI44:255-261
Bowen JE (1972) Manganese-silicon interaction and its effect on growth of sudan grass.
Plant Soil 37: 577-588
Brown JC, Holmes RS, Tiffin LO (1959) Hypothesis concerning iron chlorosis. Soil
Sci Soc Am Proc 23:231-234
Brownell PF (1965) Sodium as an essential micronutrient element for a higher plant
(Atriplex vesicaria). Plant Physio140:460-468
Brownell PF (1968) Sodium as an essential micronutrient for some higher plants. Plant
Soil 28 : 161-164
Brownell PF, Crossland CJ (1972) The requirement for sodium as a micronutrient by
species having the C 4 dicarboxylic photosynthetic pathway. Plant Physiol 49:794-
797
Brownell PF, Wood JG (1957) Sodium as an essential micronutrient element for Atriplex
vesicaria Howard. Nature 179: 635-636
Carter OG, Lathwell DJ (1967) Effect of chloride on phosphorus uptake by corn roots.
Agron J 59:250-253
Chaudhry FM, Loneragan JF (1972a) Zinc absorption by wheat seedlings. I. Inhibition
by macronutrient ions in short term studies and its relevance to long term zinc nutri-
tion. Soil Sci Soc Am Proc 36: 323--327
Chaudhry FM, Loneragan JF (1972b) Zinc absorption by wheat seedlings. II. Inhibition
by hydrogen ions and by micronutrient cations. Soil Sci Soc Am Proc 36:327-331
Clarkson DT (1966) Effect of aluminium on the uptake and metabolism of phosphorus
by barley seedlings. Plant Physiol 41 : 165-172
Conrad JP (1950) Sulfur fertilization in California and some related factors. Soil Sci
70:43--54
Cram WJ (1973) Internal factors regulating nitrate and chloride influx in plant cells.
J Exp Bot 24:328-341
DeKock PC, Morrison RI (1958) The metabolism of chlorotic leaves. 2. Organic acids.
Biochem J 70: 272-277
Delwiche CC, JohnsonCM, Reisenauer HM (1961) Influence of cobalt on nitrogen
fixation by Medicago. Plant PhysioI36:73--78
Dilworth MJ (1974) Dinitrogen fixation. Annu Rev Plant PhysioI25~81-114
Dixon NE, Gazzola C, Blakeley RL, Zerner B (1975) Jackbean urease (EC 3.5.1.5),
a metalloenzyme. A simple biological role for nickel? J Am Chern Soc 97:4131-4133
Downton WJS (1978) Growth and' flowering in salt-stressed avocado trees. Aust J Agric
Res 29:523--534
Eaton FM (1966) Sulfur. In: Chapman HD (ed) Diagnostic criteria for plants and soils.
Dniv Cal Div Agric Sci, Berkeley
1.5 Interactions Between Nutrients in Higher Plants 175

Edwards DG (1968) Cation effects on phosphate absorption from solution by Trifolium

subterraneum. Aust J BioI Sci 21: 1-11
Elzam OE, Epstein E (1965) Absorption of chloride by barley roots: kinetics and selectivi-
ty. Plant Physiol 40: 620-624
Epstein E (1962) Mutual effects of ions in their absorption by plants. Agrochimica
6:293-322
Epstein E (1972) Mineral nutrition of plants: Principles and perspectives. Wiley and
Sons, New York
Ergle DR (1954) The utilization of storage sulphur by cotton and the effect on growth
and chloroplast pigments. Bot Gaz 115: 225-234
Fageria NK (1974) Uptake of potassium and its influence on growth and magnesium
uptake by groundnut (Arachis hypogaea L) plants. BioI Plant 16:210-213
Fitzpatrick EN, Dunne TC (1956) Potassium for subterranean clover. J Dep Agric West
Aust Ser 3: 321-326
Fox RL (1978) Studies on phosphorus nutrition in the tropics. In: Andrew CS, Kamprath
EJ (eds) Mineral nutrition of legumes in tropical and sub-tropical soils. CSIRO,
Melbourne
Foy CD (1974) Effect of aluminium on plant growth. In: Carson EW (ed) The plant
root and its environment. Univ Press Virg Charlottesville
Franklin RE (1969) Effect of adsorbed cations on phosphorus uptake by excised roots.
Plant Physiol 44: 687-700
Franklin RE (1970) Effect of adsorbed cations on phosphorus absorption by various
plant species. Agron J 62: 214-216
Franklin RE (1971) Cation effects on chloride, sulphate and phosphate uptake by excised
roots. Soil Sci 112:343-347
Gammon N (1953) Sodium and potassium requirements of pangola and other pasture
grasses. Soil Sci 76:81-90
Gilmore AE (1971) The influence of endotrophic mycorrhiza on the growth of peach
seedlings. J Am Soc Hort Sci 96: 35-38
Goor van BJ, Wiersma D (1974) Redistribution of potassium, calcium, magnesium and
manganese in plants. Physiol Plant 31 : 163-168
Hall RL (1971) The influence of potassium supply on competition between Nandi Setaria
and Greenleaf Desmodium. Aust J Exp Agric Anim Husb 11 :415-419
Hall RL (1974) Analysis of the nature of the interference between plants of different
species. II. Nutrient relations in a Nandi Setaria and Greenleaf Desmodium association
with particular reference to potassium. Aust J Agric Res 25: 749-756
Hall RL (1978) The analysis and significance of competitive and noncompetitive interfer-
ence between species. In: Wilson JR (ed) Plant relations in pastures. CSIRO, Mel-
bourne
Halstead EH, Barber SA (1968) Manganese uptake attributed to diffusion from soil.
Proc Soil Sci Soc Am 32:540-542
Harper JE, Nicholas JC (1978) Nitrogen metabolism of soybeans. I. Effect of tungstate
on nitrate utilization, nodulation and growth. Plant PhysioI62:662-664
Hawf LR, Schmid WE (1967) Uptake and translocation of zinc by intact plants. Plant
Soil 27: 249-260
Haynes RJ (1980) Ion exchange properties of roots and ionic interactions within the
root apoplasm: their role in ion accumulation by plants. Bot Rev 46:75-99
Heimer YM, Wray JL, Filner P (1969) The effect of tungstate on nitrate assimilation
in higher plant tissues. Plant Physiol 44: 1197-1199
Helyar KR (1978) Effects of aluminium and manganese toxicities on legume growth.
In: Andrew CS, Kamprath EJ (eds) Mineral nutrition of legumes in tropical and
sub-tropical soils. CSIRO, Melbourne
Hewitt EJ (1963) The essential nutrient elements: requirements and interactions in plants.
In: Steward FC (ed) Plant physiology: A treatise, vol III. Academic Press, London
New York
Hewitt EJ, McCready CC (1956) Molybdenum as a plant nutrient, VII. The effects
176 A.D. ROBSON and M.O. PITMAN:

of different molybdenum and nitrogen supplies on yields and composition of tomato

plants grown in sand culture. J Hortic Sci 31: 284-290
Hilder EJ, Spencer K (1954) The influence of sUlphur on a natural Medicago pasture.
J Aust Inst Agric Sci 20:171-176
Hill J, Robson AD, Loneragan JF (1978) The effect of copper and nitrogen supply
on the retranslocation of copper in four cultivars of wheat. Aust J Agric Res
29:925-939
Hill J, Robson AD, Loneragan JF (1979a) The effect of copper supply and shading
on Cu retranslocation from old wheat leaves. Ann Bot 43: 449--457
Hill J, Robson AD, Loneragan JF (1979b) The effect of copper supply on the senescence
and the retranslocation of nutrients of the oldest leaf of wheat. Ann Bot 44: 279--287
Hingston FJ, Posner AM, Quirk JP (1971) Competitive adsorption of negatively changed
ligands on 'oxide surfaces. Faraday Soc. Disc 52:334-342
Hingston FJ, Posner AM, Quirk JP (1972) Anion adsorption by goethite and gibbsite.
I. The role of the proton in determining adsorption envelopes. J Soil Sci 23: 177-192
Hodgson JF, Lindsay WL, Trierweiler JF (1966) Micronutrient cation complexing in
soil solution. II. Complexing of zinc and copper in displaced solution from calcareous
soils. Soil Sci Soc Am Proc 30:723-726
Holt ME, Yolk NJ (1945) Sodium as a plant nutrient and substitute for potassium.
J Am Soc Agron 37: 821-827
Hooymans JJM (1964) The role of calcium in the absorption of anions and cations
by excised barley roots. Acta Bot Neerl 13: 507-540
Horst WJ (1976) Einfluss von Silizium auf die Mangantoleranz von Buschbohnen (Pha-
seolus vulgaris L). Diss Nr 59 TU, Berlin -
Horst WJ, Marschner H (1978a) Effect of silicon on manganese tolerance of bean plants
(Phaseolus vulgaris L). Plant Soil 50:287-303
Horst WJ, Marschner H (1978b) Symptome von Mangan-Uberschuss bei Bohnen (Pha-
seolus vulgaris L). Z Pflanz Bodenkd 141 :129--142
Horst WJ, Marschner H (1978c) Einfluss von Silizium auf den Bindungszustand von
Mangan im Blattgewebe von Bohnen (Phaseolus vulgaris). Z Pflanz Bodenkd
141 :487-497
Hoyt PB, Myovella OOS (1979) Correction of severe manganese deficiency in wheat
with chemical fertilizers. Plant Soil 52:437-444
Husa JO, McIlrath WJ (1965) Absorption and translocation of boron by sunflower
plants. Bot Oaz 126: 186-194
Hyde AH (1966) Nature of the calcium effect in phosphate uptake by barley roots.
Plant Soil 24: 328-332
Hylton LO, Ulrich A, Cornelius DR (1967) Potassium and sodium interrelations in
growth and mineral content of Italian ryegrass. Agron J 59:311-314
Ishioka PS, Arnon DI (1955) Molybdenum in relation to nitrogen metabolism II. Assimi-
lation of ammonia and urea without molybdenum by Scenedesmus. Physiol Plant
8:552-560
Islam AKMS, Edwards DO, Asher CJ (1980) pH optima for crop growth. Results of
a flowing solution culture experiment with six species. Plant Soil 54: 339--357
Jackman RH (1965) The uptake of rubidium by the roots of some graminaceous and
leguminous plants. NZJ Agric Res 8:763-777
Jackson WA (1967) Physiological effects of soil acidity. In: Pearson RW, Adams F
(eds) Soil acidity and liming. Am Soc Agron, Madison, Wisc
Jackson WA, Flesher D, Hageman RH (1973) Nitrate uptake by dark-grown corn seed-
lings: some characteristics of apparent induction. Plant Physiol51 :'120-127
Johansen C, Edwards DO, Loneragan JF (1968) Interactions between potassium and
calcium in their absorption by intact barley plants. 1. Effect of potassium on calcium
absorption. Plant Physiol 43: 1717-1721
Johansen C, Edwards DO, Loneragan JF (1969) Interactions between potassium and
calcium in their absorption by intact barley plants. II. Effects of calcium and potassi-
um concentration on potassium absorption. Plant Physiol 43: 1722-1726
Jones LHP (1957) The solubility of molybdenum in simplified systems and aqueous
soil suspensions. J Soil Sci 8: 312-327
1.5 Interactions Between Nutrients in Higher Plants 177

Jones LHP, Handreck KA (1967) Silica in soils, plants and animals. Adv Agron
19:107-149
Jones LHP, Leeper GW (1951) The availability of various manganese oxides to plants.
Plant Soil 3:141-153
Jones OP, Rowe RW (1968) Sampling the transpiration stream in woody plants. Nature
219:403
Jones RJ (1966) Nutrient requirements of improved pasture on podzolic soils developed
in phyllite at North Deep Creek. Proc Trop Grassl Soc Aust 6:23-30
Kannan S (1969) Factors related to iron absorption by enzymically isolated leaf cells.
Plant Physiol44: 1457-1460
Kannangara CG, Woolhouse HW (1968) Changes in the enzyme activity of soluble
protein fractions in the course of foliar senescence in Perillafrutescens L. Butt. New
Phytol 67:533-542
Kirkby EA (1968) Influence of ammonium and nitrate nutrition on the cation anion
balance and nitrogen and carbohydrate metabolism of white mustard plants grown
in dilute nutrient solutions. Soil Sci 105: 133-141
Kirkby EA, Armstrong MJ (1980) Nitrate uptake by roots as regulated by nitrate assimi-
lation in the shoot of castor oil plants. Plant Physiol 65: 286-290
Lambert HD, Baker DE, Cole H (1979) The role of mycorrhizae in the interactions
of phosphorus with zinc, copper and other elements. Soil Sci Soc Am J 43:976-980
Larsen S (1967) Soil phosphorus. Adv Agron 19:151-210
Lazaroff N, Pitman MG (1966) Calcium and magnesium uptake by barley seedlings.
Aust J BioI Sci 19:991-1005
Ledin S, Wiklander L (1974) Exchange acidity of wheat and pea roots in salt solutions.
Plant Soil 41 :403-413
Leggett JE, Epstein E (1956) Kinetics of sulfate absorption by barley roots. Plant Physiol
31:222~226
Leggett JE, Gilbert WA (1969) Magnesium uptake by soybeans. Plant Physiol
44: 1182-1186
Lindsay WL (1978) Chemical reactions affecting the availability of micronutrients in
soil. In: Andrew CS, Kamprath EJ (eds) Mineral nutrition of legumes in tropical
and sub-tropical soils. CSIRO, Melbourne
Lindsay WL, Vlek PLG (1977) Phosphate minerals. In: Dixon JB, Weed SB (eds) Miner-
als in the soil environment. Soil Sci Soc Am, Madison, Wisconsin
Loneragan JF (1959) Calcium in the nitrogen metabolism of subterranean clover. Aust
J BioI Sci 12:26-39
Loneragan JF (1964) The nutrition of grasslands. In: Barnard C (ed) Grasses and grass-
lands. MacMillan, London Melbourne
Loneragan JF (1968) Nutrient requirements of plants. Nature 188: 1307-1308
Loneragan JF (1975) The availability and absorption of trace elements in soil-plant
systems and their relation to movement and concentrations of trace elements in plants.
In: Nicholas DJD, Egan AR (eds) Trace elements in soil-plant-animal systems. Aca-
demic Press, London New York
Loneragan JF, Asher CJ (1967) Response of plants to phosphate concentration in solution
culture. II. Rate of phosphate absorption and its relation to growth. Soil Sci
103: 311-318
Loneragan JF, Snowball K, Robson AD (1976) Remobilization of nutrients and its
significance in plant nutrition. In: Wardlaw IF, Passioura JB (eds) Transport and
transfer processes in plants. Academic Press, London New York
Loneragan JF, Grunes DL, Welch RM, Aduayi EA, Tengah A, Lazar DA, Cary EE
(1979) P accumulation and toxicity in leaves in relation to zinc supply. Annu Meet,
Fort Collins. Agron Abstr, pp 17~176
Loneragan JF, Grove TS, Robson AD, Snowball K (1980a) Phosphorus toxicity as
a factor in zinc-phosphorus interactions in plants. Soil Sci Soc Am J 43: 966-972
Loneragan JF, Snowball K, Robson AD (1980b) Copper supply in relation to content
and redistribution of copper among organs of the wheat plant. Ann Bot 45: 621-632
Macrobbie EAC (1971) Phloem translocation, facts and mechanisms: A comparative
study. BioI Rev 46:429-481
178 A.D. ROBSON and M.O. PITMAN:

Marion OM, Hendricks DM, Dutt OR, Fuller WH (1976) Aluminium and silica solubility
in soils. Soil Sci 121 :76-85
Mengel K, Kirkby EA (1978) Principles of plant nutrition. Berne Int Potash Inst
Millikan CR (1951) Diseases of flax and linseed. Dep Agric Vic Aust Tech Bull 9: 1-120
Millikan CR (1963) Effects of different levels of zinc and phosphorus on the growth
of subterranean clover (Trifolium subterraneum). Aust J Agric Res 14: 180-205
McKenzie RM (1967) The sorption of cobalt by manganese minerals in soils. Aust J
Soil Res 5:235-246
McKenzie RM (1970) The reaction of cobalt with manganese dioxide minerals. Aust
J Soil Res 8:97-106
McKenzie RM (1971) The influence of cobalt on the reactivity of manganese dioxide.
Aust J Soil Res 9:55-58
McKenzie RM (1972) The sorption of some heavy metals by the lower oxides of manga-
nese. Oeoderma 8: 29-35
Moraghan JT (1980) Distribution of selected elements within flax plants as affected
by Fe EDDHA. Plant Soil 54:153-158
Munns DN (1965a) Soil acidity and growth of a legume. I. Interactions of lime with
nitrogen and phosphorus on growth of Medicago sativa L. and Trifolium subterraneum
L. Aust J Agric Res 16:733-741
Munns DN (1965b) Soil acidity and growth of a legume. II. Reactions of aluminium
and phosphate in solution and effects of aluminium, phosphate, calcium and pH
on Medicago sativa L and Trifolium subterraneum in solution culture. Aust J Agric
Res 16:743-755
Munns DN (1965c) Soil acidity and growth -of a legume. III. Interaction of lime and
phosphate on growth of Medicago sativa L in relation to aluminium toxicity and
phosphate fixation. Aust J Agric Res 16:757-766
Nicholas DJD, Nason A (1955) Role of molybdenum as a constituent of nitrate reductase
from soybean leaves. Plant Physiol 30: 135-138
Nieman RH, Clark RA (1976) Interactive effects of salinity and phosphorus nutrition
on the concentration of phosphate and phosphate esters in mature photosynthesising
com leaves. Plant PhysioI57:157-161
Notton BA, Hewitt EJ (1971) The role of tungsten in the inhibition of nitrate reduc-
tase in spinach (Spinacia oleracea L) leaves. Biochem Biophys Res Commun 44:702-
710
Nyatsanga T, Pierre WH (1973) Effects of nitrogen fixation by legumes on soil acidity.
Agron J 65: 936-940
Nye PH, Tinker PB (1977) Solute movement in the soil-root system. Blackwell, Oxford
Olsen C (1935) Iron absorption and chlorosis in green plants. CR Trav Lab Carlsberg
Ser Chim 21: 15-63
Osmond CB (1966) Divalent cation absorption and interaction in Atriplex. Aust J BioI
Sci 19:37-48
Ozanne PO, Howes KMW, Petch A (1976) The comparative phosphate requirements
offour annual pastures and twb crops. Aust J Agric Res 27:479-488
Pairunan AK, Robson AD, Abbott LK (1980) The effectiveness of vesicular-arbuscular
mycorrhizas in increasing growth and phosphorus uptake of subterranean clover from
phosphorus sources of different solubilities. New Phytol 84: 327-338
Paribok TA, Alekseeva-Popova NY (1965) Effect of zinc on absorption and utilization
of phosphorus by plants. Sov Plant PhysioI12:514-518
Pate JS (1975) Exchange of solutes between phloem and xylem and circulation in the
whole plant. In: Zimmermann MH, Milburn JA (eds) Transport in plants I. Encyclo-
pedia of plant physiology new ser vol 1. Springer, Berlin Heidelberg New York
Peaslee DE, Frink CR (1969) Influence of silicon acid on uptake of Mn, AI, Zn, Cu
by tomatoes. Soil Sci Soc Am Proc33: 569-571
Piper CS, Beckwith RS (1949) The uptake of copper and molybdenum by plants. In:
Plant and animal nutrition in relation to soil and climatic factors. Br Comm Spec
Conf Agric HMSO, London
Pitman MO (1964) The effect of divalent cations on the uptake of salt by beetroot
tissue. J Exp Bot 15:444-456
1.5 Interactions Between Nutrients in Higher Plants 179

Pizelle G, Thiery G (1979) Influence du tungstate sur la nutrition azotee nitrique et

symbiotique d' Alnus glutinosa. Physiol Veg 17: 769-77 5
Polacio JC (1977) Nitrogen metabolism in soybean tissue culture. II. Urea utilization
and urease synthesis require Ni 2 + . Plant Physiol 59: 827-830
Price CA (1970) Molecular approaches to plant physiology. McGraw-Hill, New York
Quin BF, Hoglund JH (1976) The effects of tungstate and nitrogen source on the dry
weight and nitrogen yields and molybdenum and tungsten content of white clover.
Plant Soil 45: 201-212
Rao KP, Rains DW (1976) Nitrate absorption by barley. II. Influence of nitrate reductase
activity. Plant Physiol 57: 59-62
Rasmussen HP (1968) Entry and distribution of aluminium in Zea mays: Electron micro-
probe X-ray analysis. Planta 81 :2&-37
Rediske JH, Biddulph 0(1953) The absorption and translocation of iron. Plant Physiol
28:576-593
Reisenauer HM (1966) Mineral nutrients in soil solution. In: Environmental biology.
Fed Am Soc Exp BioI BioI Handb, Bethesda Md
Reuter DJ (1980) Distribution of copper and zinc in subterranean clover in relation
to deficiency diagnosis. PhD thesis, Murdoch Univ West Aust
Robson AD (1978) Mineral nutrients limiting nitrogen fixation in legumes. In: Andrew
CS, Kamprath EJ (eds) Mineral nutrition of legumes on tropical and sub-tropical
soils. CSIRO, Melbourne
Robson AD, Loneragan JF (1970) Sensitivity of annual Medicago species to manganese
toxicity as affected by calcium and pH. Aust J Agric Res 21 :223-232
Robson AD, Loneragan JF (1978) Responses of pasture plants to soil chemical factors
other than nitrogen and phosphorus, with particular emphasis on the legume symbio-
sis. In: Wilson JR (ed) Plant relations in pastures. CSIRO, Melbourne
Robson AD, Edwards DG, Loneragan JF (1970) Calcium stimulation of phosphate
absorption by annual legumes. Aust J Agric Res 21 :601-612
Rorison IH (1965) The effect of aluminium on the uptake and incorporation of phosphate
by excised sainfoin roots. New Phytol 64:23-27
Rosell RA, Ulrich A (1964) Critical zinc concentrations and leaf minerals of sugar beet
leaves. Soil Sci 97: 152-167
Rossiter RC (1947) Effect of potassium on the growth of subterranean clover and other
pasture plants on Crawley sand. 2. Field plot experiments .. J Counc Sci Ind Res
20:389-401
Rossiter RC (1966) Ecology of the Mediterranean annual type pasture. Adv Agron
18:1-56
Russell EW (1973) Soil conditions and plant growth. Longman, London
Silva da JJR Fr, Williams RJP (1977) The uptake of elements by biological systems.
Struct Bond 29:67-121
Smith FA (1973) The internal control of nitrate uptake into barley roots with differing
salt content. New Phytol 72: 769-782
Smith FA, Fox AL (1977) Interactions between chloride and nitrate uptake in Citrus
leaf slices. Aust J Plant Physiol 4: 177-182
Smith FA, Raven JA (1976) Nitrogen assimilation and transport in vascular land plants
in relation to intracellular pH regulation. New Phytol 76:415-431
Smith FW (1974) The effect of sodium on potassium nutrition and ionic relations in
rhodes grass. Aust J Agric Res 25:407-414
Snowball K, Robson AD, Loneragan JF (1980) The effect of copper on nitrogen fixation
in subterranean clover (Trifolium subterraneum). New Phytol 85: 63-72
Somers JJ, Shive JW (1942) The iron-manganese relation in plant metabolism. Plant
Physiol 17: 582-602
Spear SN, Asher CJ, Edwards DG (1978a) Response of cassava, sunflower and maize
to potassium concentration in solution. I. Growth and plant potassium concentration.
Field Crop Res 1: 347-361
Spear SN, Edwards DG, Asher CJ (1978b) Response of cassava, sunflower and maize
to potassium concentration in solution. II. Interaction 'between potassium, calcium
and magnesium. Field Crop Res 1: 375-389
180 A.D. ROBSON and M.G. PITMAN: 1.5 Interactions Between Nutrients

Stanton DA, Burger du TR (1967) Availability to plants of zinc sorbed by soil and
hydrous iron oxides. Geoderma 1: 13-17
Stanton DA, Burger du TR (1970) Studies on zinc in selected Orange Free State soils.
V. Mechanics for the reactions of zinc with iron and aluminium oxides. Agrochemo-
physica 2:65-76
Stout PR, Meagher WR, Pearson GA, Johnson CM (1951) Molybdenum nutrition of
crop plants. I. The influence of phosphate and sulphate on the absorption of molybde-
num from soils and solution cultures. Plant Soil 3: 51-87
Suelter CH (1974) Monovalent cations in enzyme-catalyzed reactions. In: Sigel H (eds)
Metal ions in biological systems. Decker, New York
Tammes PML, Die van J (1964) Studies on phloem exudation from Yuccaflaccida flaw.
I. Some observations on the phenomenon of bleeding and the composition of the
exudate. Acta Bot Neerl 13: 76-83 .
Tanada T (1955) Effect of ultra violet radiation and calcium and their interaction on
salt absorption by excised mung bean roots. Plant PhysioI30:221-225
Tiffin LO (1967) Translocation of manganese, iron, cobalt and zinc in tomato. Plant
Physiol42: 1427-1432
Tiller KG, Honeysett JL, Hallsworth EG (1969) The isotopically exchangeable form
of native and applied cobalt in soils. Aust J Soil Res 7:43-56
Timonin MI (1946) Microflora of the rhizosphere in relation to the manganese deficient
disease of oats. Soil Sci Soc Am Proc 11 : 284-292
Tromp J (1962) Interactions in the absorption of ammonium, potassium and sodium
by wheat roots. Acta Bot N eerl 11 : 147-192
Trumble HC, Donald CM (1938) The relation of phosphate to the development of seeded
pasture on a podsolized sand. Counc Sci Ind Res Aust Bull 116:1-47
Underwood EJ (1977) Trace elements in human and animal nutrition, 4th edn. Academic
Press, London New York
Vlamis J, Williams DE (1967) Manganese and silicon interaction in the Grarnineae.
Plant Soil 27 : 131-140
Walker TW, Adams AFR, Orchiston HD (1956) The effect of levels of calcium sulphate
on the yield and composition of a grass and clover pasture. Plant Soil 7: 290-300
Wehunt RL, Collins WD (1953) Response of oats to Na and K on Norfolk sandy
loam at two residual K levels. Soil Sci 76:91-96
Whelan BR, Edwards DG (1975) Uptake of potassium by Setaria anceps and Macropti-
lium atropurpureum from the same standard solution culture. Aust J Agric Res
26:819--829
White RE (1976) Studies on Inineral ion absorption by plants. III. The interaction of
aluminium, phosphate and pH on the growth of Medicago sativa. Plant Soil
46:195-208
White RE, Tiffin LO, Taylor AW (1976) The existence of polymeric complexes in dilute
solutions of aluminium and orthophosphate. Plant Soil 45: 521-529
Williams CH (1977) Trace metals and superphosphate: toxicity problems. J Aust Inst
Agric Sci 43: 99--109
Williams DE, Vlamis J (1957) The effect of silicon on yield and manganese 54 uptake
and distribution in the leaves of barley plants grown in culture solutions. Plant Physiol
32:404-409
Williams WA, Love RM, Conrad JP (1956) Range improvement in California by seeding
annual clovers, fertilization and grazing management. J Range Manage 9:28--33
Willoughby WM (1954) Some factors affecting grass-clover relationships. Aust J Agric
Res 5:157-180
Wit de CT, Dijkshoorn W, Noggle JC (1963) Ionic balance and growth of plants. Versl
Landbouwkd Onderz 69: 15
Woolhouse HW (1969) Differences in the properties of the acid phosphatases of plant
roots and their significance in the evolution of edaphic ecotypes. In: Rorison IH
(ed) Ecological aspects of the Inineral nutrition of plants. Blackwell, Oxford
Yoshida F (1966) Inter-relationships between potassium and magnesium absorption by
oats (Avena sativa L). Thesis State Agric Univ Wageningen
1.6 Import and Export of Mineral Nutrients
in Plant Roots 1
U. LUrTGE

1 Introduction: The Dual Role of Roots in the Evolution

of Higher Land Plants

The aqueous medium of water plants provides two essentials; firstly, support,
which is obtained by various buoyancy mechanisms, and secondly, a nutrient
supply where the solution is in contact with the whole plant. With the evolution
of higher land plants roots acquired the dual role of anchoring the plant in
the ground and serving to exchange material with the soil medium. Root-like
structures, rhizoids, are frequently observed in lower plants. They are often
lengthy single cells or may be multicellular such as in kelp. In benthic algae
these rhizoids normally only function for attachment. In liverworts and mosses
the unicellular rhizoids may serve both anchoring and transport functions. But
only in higher land plants has there been an evolution of complex structures
to satisfy both requirements.
These structures, the roots, are the major absorptive organs of higher plants,
and it is no surprise that they have been the material most frequently used
for studies of general mechanisms of ion uptake by plant tissues (reviews, e.g.,
PITMAN 1976, EpSTEIN 1976). From the nutritional point of view, however,
uptake and accumulation of ions in the root tissue itself are often less pertinent
than the transport through roots, followed by export and translocation of ions
to the shoot system (reviews, e.g. PITMAN et al. 1976, ANDERSON 1976).
The dual role of roots needs to be stressed even in a review that focuses
only on the physiological function. Root structure must serve both functions.
Although highly optimized in the evolutionary process, the structural features
that have evolved for one function may interfere with the other function and
vice versa. A typical example is the occurrence of secondary growth of roots,
which is particularly developed in larger dicotyledonous plants and trees, and
necessary in providing their special anchoring requirements. The associated lig-
nification, periderm formation and suberization of the major part of the root
system prevents uptake of water and solutes and restricts it to the very tips
of roots, i.e. to a quite small fraction of the entire root system. It is conceivable
that ectotrophic mycorrhizas of trees have evolved as a consequence of this.
The hyphae of mycorrhizal fungi extend several centimeters from the r60t sur-
face into the soil, and intrude the cell walls of the root cortex up to the endoder-
mis. In this way the surface area which is in contact with the soil and which
acts as a pathway for transfer of nutrients from the soil to the root is much
increased (see PITMAN 1976, CLARKSON and HANSON 1980).
1 Dedicated to the memory of NOE HIGINBOTHAM, distinguished promoter of this field
182 U. LUrrGE:

2 Relations Between Structure and Transport Functions Along

the Length of Roots

2.1 The Phenomenon of Variations in Transport Functions Along

the Length of Roots

Variations in transport functions along the length of roots have long been
documented experimentally, in particular by the use of radioactive tracers ad-
ministered to different root zones (e.g. WIEBE and KRAMER 1954, SHONE et al.
1969, BOWEN 1969, 1970, ROVIRA and BOWEN 1970, SMITH 1970, MARSCHNER
and RICHTER 1973). A general simple picture of the most effective root zone(s)
in transport has not emerged, because there is considerable diversity of behav-

_f_ ..f. .-f- .c:

(j)
Fig. 1. Absorption and translocation of K + ,
c
N a + and Ca 2+ in various root zones of 9 day
OJ old maize seedlings (Zea mays L. cv. Velox).
1 mEq 1-1 solutions of these ions (labelled
with 42K, 22Na, and 4SCa) were administered
. to various root zones as indicated by the
arrows, and the remainder of the roots were
bathed with non-labelled nutrient solution of
otherwise identical composition. After 24 h
plants were harvested, roots were rinsed for
15 min in non-labelled solution and then sepa-
rated in 3-cm segments; root segments, grains
(endosperm) and shoots were analyzed sepa-
base rately. The amounts of labelled K +, Na + or
Ca2+ found after 24 h in the various zones
are given in the histograms as nEq per 12
E
OJ (j) plants calculated on the basis of the specific
0.-
(j)
o 0
0 activity of the labelling solution. The lines
"C .c: touching the arrows give the sums, i.e. the total
c (j)
OJ amount of label taken up. Note the logarith-
mic scale of the ordinate. The reader can make
the various comparisons allowed by the in-
structive experiment by himself. Some interest-
ing aspects are as follows: K + and Na + ab-
root base
sorption are slightly smaller in the root zones
.l!! closer to the tip than further up; conversely
o
o Ca 2+ absorption is larger in the tip region.
_f_ ....._,_
...E -ffi
OJ ......
K + and Na + are translocated towards the root
IF tip and the root base, grain and shoot but the

~
o profIles are different. Na + transport towards
"C
3

t - -, ,--Hl-::~~
C
a. 10 '--':,!:::, :.......:r._., OJ the tip seems to be less than K + transport.
N a + export seems to be similar in all root
zones, whereas K + transport out of the tip
region is smaller than export from the other
zones. The latter is also observed with Ca2+,
but Ca 2+ is only transported upwards to the
grains and shoots and not at all downwards
o 3 6 9 12 15 18 21 24 27cm to the tip. (Data from MARSCHNER and
tip root base RICHTER 1973)
1.6 Import and Export of Mineral Nutrients in Plant Roots 183

Fig. 2. Phosphate uptake and 50

translocation in various root
zones of 6-day-old maize seed-
lings (Zea mays cv. Golden Cross
Bantam). (BURLEY et al. 1970)

..
... ."
..
o 20 60 100 140 180 220
Distance of uptake zone from root tip [mmJ
'0
o
ti 60
.£ 50
£ 40
.~30
..
... ..
>
t; 20
CII
~ 10 ~.

f! ~4-"1,o ..t~,...,·. . . . . ...,..-.-.--,.-.-.--,.-..-.---'-T"'---,

;;'! 0 20 60 100 140 180 220
Distance of uptake zone from root tip [mm]

iour (summarized in the introduction of MARSCHNER and RICHTER 1973, CLARK-

SON and HANSON 1980). This is due to wide differences between plant species,
ionic species (Fig. 1), environmental and nutritional conditions (BOWEN 1970,
CLARKSON et al. 1978c) and positions in the root-branching system, i.e. primary
root, lateral root or adventitious root (RUSSELL and SANDERSON 1967). Distinc-
tion also must be made between uptake and accumulation in the root tissue
on the one hand and translocation on the other (Figs. 1 and 2, CLARKSON
et al. 1968, SHONE et al. 1969). Clearly both structural and physiological varia-
tions along the length of the root must be responsible for the observed diversity.

2.2 Structure-Function Relations in Various Root Zones

A three-dimensional view of the root is required in order to understand the

complex structural relations along and across the root in respect to the transport
functions of uptake and translocation and also with regard to the role of the
roots in anchoring the plant and exploiting the soil (Fig. 3).

2.2.1 The Root Surface

2.2.1.1 The Mucigel of the Tip
Glycoprotein secretion (GREEN and NORTHCOTE 1979) and degeneration of the
root cap cells form a slimy material at the surface of the very tip of roots.
Fig. 3 A-D. Three dimensional scheme of a root (lower right) with details of four successive
root zones, as discussed in Sections 2.2.1- 2.2.4, describing transport across the root
in the various zones and along the length of roots. Anatomy of grass roots was taken
as a basis for the scheme; distances along the length of the root are not indicated because
the differentiation can vary greatly, depending on species and growth conditions. A Zone
of primary root differentiation, B Root hair zone with primary endodermis, C Zone
of secondary endodermis, D Zone of lateral root formation, hypodermis and tertiary
endodermis. C cortex; CS Casparian strip; En endodermis (pEn primary, sEn secondary,
tEn tertiary endodermis); Ep epidermis; Hy hypodermis; LR lateralroot; MX metaxylem;
P pericycle; PC passage cell; Ph phloem; PX protoxylem; RC root cap; RH root hair;
SL suberin lamella; XP xylem parenchyma
1.6 Import and Export of Mineral Nutrients in Plant Roots 185

But also further up the primary root the epidermis is covered by an amorphous
material, a mucigel, containing free carboxyl groups and other material of hy-
drophilic nature (see Chap. 1.2, this Vol. for a full description). This peripheral
mucigel considerably facilitates the uptake of ions by ion exchange with solid
soil particles (see LXUCHLI 1976a, PITMAN et al. 1976). Autoradiographs demon-
strate that there can be a preferential adsorption of polyvalent ions in the
mucigel, a process which is passive and thus independent of temperature
(CLARKSON and SANDERSON 1969). Presumably this adsorption is also responsi-
ble for the relatively high uptake and retention of Ca2+ in the apical 3 cm
of maize roots (Fig. 1).

2.2.1.2 The Primary Root Epidermis

The dominating role of the outermost cell layer of the root, the primary root
epidermis, in ion adsorption, absorption and uptake can be documented by
autoradiographs (Fig. 4) and X-ray microanalysis (LXUCHLI 1967, LXUCHLI et al.
1971). Pectic substances in the relatively thin cell walls again may be important
for adsorption of cations prior to absorption. The formation of root hairs
is essential in providing a much increased external surface for these processes.
Recently it was discovered that, in response to various situations of stress,
the root epidermis also can increase its inner surface. The primary roots thicken
in a zone which contains the root hairs. The cell walls of the root hairs and
the outer tangential walls of the epidermal cells form protuberances or in-
growths, leading to a considerable increase of the plasmalemma surface. Thus,
these cells develop to transfer cells, which are generally characterized by such
cell wall labyrinths frequently occurring at apoplast-symplast boundaries in
sites where there is considerable transport across the plasmalemma (GUNNING
1977). One example is salt stress of Atriplex hastata plants which develop root
thickenings and peripheral transfer cells 1-3 mm behind the tips when grown
in media with 100-600 mM NaCI (Fig. 5, KRAMER et al. 1978). Similarly, Helian-
thus annuus plants form root thickenings and transfer cells 3-15 mm behind
the tips when Fe is absent from the nutrient solution for 24 h to 48 h, i.e.
under conditions of stress due to Fe deficiency (Fig. 6, KRAMER et al. 1980).
Helianthus has been characterized as belonging to the "Fe-efficient" plants
which respond to Fe deficiency by H + extrusion and production of reducing
substances. These two processes considerably increase the capacity of the plants
for uptake of Fe, since Fe is largely transported in the form ofFe2+ (MARSCHNER
et al. 1974, BROWN 1978). This behaviour is related to adaptation to calcareous
soils, i.e. "Fe-efficient" plants are calcicole, whereas" Fe-inefficient" ones are
calcifuge (BROWN 1978, CLARKSON and HANSON 1980).
Conceivably, the increased plasmalemma surface of root epidermis transfer
cells formed under conditions of Fe deficiency serves increased H + pumping
out of the cells. The ubiquitous occurrence of H + extrusion pumps in higher
plant cells including roots has been stressed repeatedly (PITMAN et al. 1975,
ANDERSON et al. 1977, reviews MARRE 1979, SMITH and RAVEN 1979, see Sect.
2.2.4.3). Additional experiments with A. hastata (KMM£R unpublished) show
that NaCI stress induces symptoms of Fe deficiency, although Fe is present
186 U . Lij'TTGE:

Fig. 4. Adsorption and absorption of sulphate by the peripheral cell layers of Zea mays
(a) and Acer platanoides (b) seminal roots, and by a single root hair of A. platanoides
(c). 20 mM 35SOi- was applied for 1 h (a) and 2 h (b, c), respectively. The root in
(a) was inhibited with 1 mM NaN 3 (see Sect. 2.2.3). a from W EIGL and LUTTGE (1962),
b from LUTTGE (1964). C cortex ; Ep epidermis; RH root hairs

in the medium. This can be due to effects of ion competition. The formation
of an increased plasmalemma surface in the transfer cells may be a response
to particular passive and active transport problems in the presence of high
NaCI concentrations.
Aerial roots of orchids often have a special velamen radicum, which morpho-
genetically corresponds to a multilayered root epidermis. The cells of the func-
tional velamen are dead; they have complex cell wall structures with pores
and cracks creating capillary forces, which allow the velamen to rapidly imbibe
any water sprayed on the plant (rain) (CAPESIUS and BARTHLOTT 1975). The
outermost layer of the cortex, i.e. the hypodermis forms a sheath which resem-
bles endodermal differentiations having Casparian strips and secondary and
1.6 Import and Export of Mineral Nutrients in Plant Roots 187

a c

Fig. 5. a Atriplex hastata root grown in 400 mM NaCI with root thickening behind
the tip and with numerous root hairs. b Cell wall ingrowths of a root hair from a
root grown in 400 mM NaC!. c Cell wall of a root epidermis cell of the same ontogenetic
level as in b but from a root grown in the absence of NaCI. (KRAMER et al. 1978)

tertiary developmental stages including formation of passage cells (see Sect.

2.2.3). Tritiated water and 32P-Iabelled phosphate are readily taken up from
the velamen by the underlying living root tissues (CAPESIUS and BARTHLOTT
1975).

2.2.1.3 Root Hypodermis and Periderm

Above the zone of the root hairs, the epidermis often deteriorates and the
outer cells of the cortex begin to suberize, forming a hypodermis (or exodermis)
on the root surface. Suberization of cell walls may also occur in the epidermal
cells and even in the root hairs. In Zea mays it was shown that the deposition
188 U . LUTTGE :

Fig. 6a, b. Root-epidermal transfer cells with cell wall protuberances due to iron defi-
ciency. Note the abundance of mitochondria. a Helianthus annuus, Fe omitted from
culture solution for several days. (Unpublished micrograph, see KRAMER et al. 1980.)
b Capsicum annuum, Fe omitted for several weeks (unpublished electron-micrograph by
courtesy of Dr. D . KRAMER)

of suberin lamellae in hypodermal walls led to a marked reduction of phosphate

uptake (FERGUSON and CLARKSON 1976b). Similar results were obtained with
Carex arenaria, where several suberized cell layers in the outer cortex largely
prevent the passage of H 2 0 , Ca2+ and phosphate (CLARKSON et al. 1978a).
But this conclusion cannot be generalized. In another species, Allium cepa, suber-
ized cells walls of the hypodermal cell layers seem to be of a porous nature
remaining highly permeable for H 2 0, ions and small organic molecules. In
this species, obviously, hydrophilic channels pass across the sllberin lamellae,
and solutes are selected on the basis of their molecular weight (CLARKSON et al.
1978b). In salt-stressed Atriplex hastata roots and in Fe-stressed Helianthus
roots, hypodermal cells even form cell wall labyrinths on their outer tangential
walls when epidermal cells die, and thus seem to serve special transport functions
(see Sect. 2.2.1.2., KRAMER et al. 1978, KRAMER et al. 1980).
1.6 Import and Export of Mineral Nutrients in Plant Roots 189

Fig. 7. Electrical analogue depicting alternative ion uptake Medium

pathways across the root. RH root hair; Ep epidermis; Plasmalemma
C cortex; En endodermis; XP xylem parenchyma cells; CS barriers
Casparian strip RH+Ep
-iii iii
rl'" >->JVVVVVVV\J--< ~u
;:0. _
~o
C o UI
U'"
00. u III
0.
E
CS En (ij~

-
Ci]
]i g. >--.IVINX'V'p>IVV---""
]i
UI

UI~

xylem vessel
lumen

In Iris roots, where the cortical cells have storage functions, a peridenn
develops at the root surface. This peridenn has been found to be entirely imper-
meable to sulphate (ZIEGLER et al. 1963).
Thus in these root zones uptake is often much reduced or negligible. Lateral
roots, initiated in the stele and breaking through the cortex and root surface,
begin to exploit the soil in these zones. It has often been suspected that rupture
of the outer root tissues by the lateral roots opens routes for uncontrolled
passive entry of solutes into the stele, but quantitatively this seems to be not
important (see also Sect. 2.2.3).

2.2.2 The Cortex

The cortex usually comprises several layers of large, rather non-differentiated
and highly vacuolated parenchyma cells. When a root fulfils storage functions,
the cortex is particularly thick. In other cases, such as in the roots of Calluna
vulgaris, the cortex can be very much reduced and is only represented by the
endodermis, which morphogenetically corresponds to the innennost layer of
the cortex (pp. 379-380 in TROLL and HOHN 1973).
The cortex must be traversed by water and solutes during their transport
from the medium to the stele where they can then be exported from the root.
In principle this passage can occur via either apoplastic or symplastic pathways,
and the situation can be depicted by an electrical analogue (e.g. GLINKA 1977,
Fig. 7). Apoplast and symplast constitute two parallel pathways of different
resistances. The apoplastic pathway is blocked at the endodennis by the Caspar-
ian strip (see Sect. 2.2.1.4). The symplastic pathway can be entered by passage
through the plasmalemma in the root hair and epidennal cells, in the cortical
cells and in the endodennal cells. The relative importance of the vari~us path-
ways for transport across the cortex depends on their relative resistances, which
may vary for plant species, solute species, environmental conditions (e.g. solute
concentration in the medium), and intrinsic factors (e.g. honnonal balance,
state of ionic relations, etc.). Perhaps this consideration can help to resolve
a long-lasting controversy on whether or not the cortical apoplast is an impor-
190 U. LOrrOE:

tant and rather freely accessible pathway ("free space", see LXUCHLI 1976a)
for entry of solutes deeper into the root. On the one hand, for example, in
the Crafts-Broyer hypothesis of ion transport across the root, the entire cortical
apoplast is taken as supplying a considerably increased surface for membrane-
controlled ion uptake (CRAFTS and BROYER 1938, LATIES 1967, 1969). Conver-
sely, VAKHMISTROV (1967), BANGE (1973), and VAN IREN and VAN DER SLUIJS
(1980) maintain that uptake is exclusively limited to the root surface and thus
"the epidermis is the only actively absorbing part of the root, the salt solution
in the free space representing in a sense a ballast volume". Histochemical local-
ization of ATPase activity in corn roots suggests that the epidermis and outer
cortex must have a primary energy-linked role in ion absorption by the root
(MALONE et al. 1977).
There is much evidence that, as the electrical analogue suggests, both the
apoplastic pathway up to the endodermis and the symplastic pathway are in
fact used. Cytological labelling techniques have demonstrated contributions of
both the apoplastic (see LXUCHLI 1976a) and the symplastic pathways of trans-
port across the root. The endoplasmic reticulum extending from cell to cell
through plasmodesmata is used as a route in symplastic transport (STELZER
et al. 1975, VAN IREN and VAN DER SPIEGEL 1975, ROBARDS and CLARKSON
1976). Visible cytoplasmic streaming in the1'oot cortex apparently is not impor-
tant for radial transport (GLASS and PERLEY 1979). GRUNDWALDT et al. (1979)
suggest that at low ion concentrations in the external medium uptake is largely
restricted to the root surface, but that uptake from the cortical apoplast begins
to playa role as ion concentrations are increased. PITMAN (1965) argued that
in a solution of 0.5 mM K + +9.5 mM Na + the transport ofK + and presumably
also of Na + across the cortex of barley roots was too large to be explained
solely by diffusion in the apoplast and that the symplastic pathway must play
an important role. Since the cell wall carries a negative charge, the apoplastic
pathway would be even less efficient for anions than for cations.
In conclusion it seems that, while it is not always superfluous, the cortex
is possibly not essential for ion absorption and translocation. This is also demon-
strated by Calluna roots in which the cortex is lacking, and is reflected in
the fact that only the root hairs and the outer epidermal or hypodermal cell
walls form protuberances in response to salt stress and iron deficiency (Sect.
2.2.1.2).

2.2.3 The Endodermis

Very much has been written on the role' of changes of endodermal structure
along the length of the root in relation to transport across the root (e.g. LXUCHLI
1976a). The endodermis is morphogenetically the innermost layer ,of the cortex.
It usually deteriorates during secondary growth of dicotyledonous roots before
it can develop further; but in the monocotyledons the primary endodermis
goes through further developmental stages of secondary and tertiary endodermis
(Fig. 3). In the primary endodermis, usually in the zone of the root hairs, the
anticlinal walls of the endodermal cells are incrusted with suberin-like material,
forming the so-called Casparian strip. The lipid composition of suberin com-
1.6 Import and Export of Mineral Nutrients in Plant Roots 191

pounds of endodermis has recently been analyzed (ESPELIE and KOLATTUKUDY

1979, SCOTT and PETERSON 1979a, b).
In the secondary endodermis, suberin lamellae are layered on the walls all
around the endodermal cells; in the tertiary endodermis, further cellulose layers
are deposited, which may alternate with bands of lignin (SCOTT and PETERSON
1979a, b). Only a few so-called passage cells in the endodermis are exempt
from these secondary and tertiary changes.
Electron microscopical localization studies using electronopaque tracers
clearly show that the apoplastic pathway in the cortex can only be used up
to the primary endodermis (reviewed by LXUCHLI 1976a). Transport across
the endodermis is dependent on metabolic energy; i.e. at least one of the plasma-
lemma barriers depicted in Fig. 7 must be passed by energy-dependent transport.
It is seen on micro-autoradiographs of maize roots that in the presence of
azide, 35S01- accumulates outside the endodermis in the cortex and epidermis
(WEIGL and LUTTGE 1962). In some cases additional cortex cell layers develop
sheaths around the stele with phi thickenings of their radial walls, e.g. in cycads
(PLAUT 1910), Taxus baccata (p. 297 in STRASBURGER 1923) and some angio-
sperms (HAAS et al. 1976, MACKENZIE 1979). Phi thickenings are lignified but
not suberized. In apple (Pyrus malus) and geranium (Pelargonium hortorum)
roots, by contrast to the Casparian strip of the endodermis, the phi thickenings
do not block apoplastic transport of a fluorescent dye (PETERSON et al. 1981).
STELZER and LXUCHLI (1977) observed that the roots of the flooding- and salt-
tolerant grass Puccinellia peisonis develop a second endodermal cell layer with
secondary (suberin lamellae) and tertiary stages (thickened walls). This may
provide additional protection from passive and uncontrolled entry of salt into
the stele under conditions of salinity.
An intriguing question is why passive diffusion into the stele of roots cannot
bypass the apoplast barrier of the Casparian strip by using apoplastic routes
in other root zones. Apoplastic routes exist in the youngest part of the root
below the zone of the primary endodermis, i.e. where the endodermis is not
yet differentiated from the rest of the periblem layers that will ev~ntually form
the cortex, or in the region where the endodermallayer is becoming discernible
but has not yet developed a Casparian strip. Clearly translocation in this zone
will be minimal because vessels have also not yet developed (but see Sect. 2.2.4).
Physiological factors may however, also be involved. Micro-autoradiographs
show that in the very tip region of the roots, Ca2+ and sulphate only penetrate
up to the border between periblem and plerome and not beyond, even though
there is no visible structural barrier (LUrTGE and WEIGL 1962, Fig. 8).
In the zone of a developed primary endodermis URSPRUNG and BLUM (1921)
suggested the existence of an abrupt physiological gradient at the endodermis,
the so-called "Endodermissprung" of water potential. Although th~ finding
of these authors has never been confirmed, various approaches have been used
to assess the possibility of physiological gradients existing at the endodermis
between the cortex and the stele.
Carbohydrate metabolism is very active both in the cortex and the stele.
In the stele there is a somewhat larger activity of pentosephosphate cycle as
compared to glycolysis than in the cortex. Pentosephosphate cycle activity also
192 U. LUTTGE:

Fig. 8. Autoradiograph of a longitudi-

nal section of a maize root tip to
which 20 mM 35S0~- has been
applied for 5 min. Label penetrated
only up to the border between perib-
lem and plerome. (LUTTGE and WEIGL
1962)

increases as the stele matures; presumably it is required to provide NADPH +

H+ for lignin biosynthesis (WONG and Ap REES 1971).
Cytological techniques like micro autoradiography and X-ray microanalysis
have suggested that ion concentrations are larger in the stele than in the cortex
(WEIGL and LUTTGE 1962, 1965, LUTTGE and WEIGL 1964, LAUCHLI 1967, 1972,
LAUCHLI et al. 1971). This conclusion can be criticized, however, because these
methods did not allow resolution between vacuoles and cytoplasm. If the con-
centration of the ions investigated were higher but uniform in the whole root
symplast as compared with the vacuoles, in view of the larger relative content
of cytoplasm in the stelar parenchyma cells these cytological observations would
only demonstrate a larger amount of ions in individual stelar cells but not
larger concentrations, let alone electrochemical potentials (LUTTGE and HIGIN-
BOTHAM 1979). Indeed, BOWLING and coworkers observed no electrochemical
gradients of ions like K+, Ca z +, Mg2+, Na+, CI-, NO;, and SOi - within
the cortex and stele; the Oz gradient is very small; a gradual increase of vacuolar
pH is as yet unexplained (DUNLOP and BOWLING 1971 a, b, c, BOWLING and
ANSARI 1972, BOWLING 1976). Their measurements have been also criticized,
however, because blindly pushing a micro electrode across the root - as done
in the experiments of BOWLING and co-workers - may cause artefacts due to
the wound created by the electrode (ANDERSON and HIGINBOTHAM 1975). Not-
withstanding these uncertainties, it appears that no sudden physiological gra-
dients occur at the endodermis, and that the endodermis has no gland-like
functions pumping ions across the roots, as suggested by older hypotheses (see
SUTCLIFFE 1962). Cells of the primary endodermis are highly vacuolated with
1.6 Import and Export of Mineral Nutrients in Plant Roots 193

a thin layer of cytoplasm along their walls, just as the cortical cells, and their
most conspicuous peculiarity is the apoplast barrier of the Casparian strip.
Recently ROBARDS et al. (1980) have shown that density of protein or glycopro-
tein particles on the freeze-etching fracture faces of the plasmalemma of endo-
dermal cells of Zea mays is about two to three times larger than in cortical
cells. This is discussed as an indication of a special activity of the endodermal
plasmalemma due to the particular necessity to take up ions from the apoplast.
This interpretation, however, is only pertinent if the apoplastic pathway up
to the endodermis is not of minor importance (see Sect. 2.2.2).
By the suberization and lignification in the secondary and tertiary endoder-
mis, the plasmalemma of the endodermal cells becomes inaccessible to the apo-
plastic pathway of the cortex, and thus uptake into the symplast across the
endodermal plasmalemma becomes impossible (see Fig. 7). Working with roots
of Hordeum vulgare, Zea mays, and Cucurbita pepo CLARKSON and coworkers
(CLARKSON et al. 1971, HARRISON-MURRAY and CLARKSON 1973, FERGUSON and
CLARKSON 1975, 1976a) found that this much reduces the radial transport of
Ca2 + and Mg2+ which are not transported symplastically (FERGUSON 1979)
but not of phosphate and K +. Presumably this also explains findings of RICHTER
and MARSCHNER (1974) that in Zea mays and Phaseolus vulgaris Ca2+ levels
are higher in the cortex than in the stele; K + was equally distributed and
Na + levels were larger in the stele.
Obviously the cells of secondary and tertiary endodermis studied by CLARK-
SON and co-workers were still alive and were traversed by plasmodesmata that
could allow symplastic transport across them. The passage cells, which are
exempt from suberisation and lignification of the secondary and tertiary endo-
dermis, must have been of minor importance for the transport of Ca2+ and
Mg2+. In Iris roots the walls of the tertiary endodermis are heavily thickened
by apposition of cellulose layers and the cells eventually die. Radial transport
then becomes restricted exclusively to the passage cells. Since in this zone of
Iris roots a peripheral periderm also prevents ion uptake from the medium,
radial transport through the passage cells is largely directed from the stele to-
wards the cortex (ZIEGLER et al. 1963).
It could be argued that lateral roots breaking through the endodermis and
cortex further up the root may provide pathways for passive movement of
ions across the endodermis. A new endodermis with a Casparian strip enclosing
steles of main root and lateral root is formed, however. In the stages of lateral
root formation where formation of the Casparian strip lags behind division
of endodermal cells, no xylem is developed and the cells of the lateral root
tip are only slightly vacuolated. In this respect the situation much resembles
that of the growing main root shown in Fig. 8 (DUMBROFF and PEIRSON 1971).

2.2.4 The Stele

The stele fulfils mechanical functions in that the central and radial arrangement
of xylem confers considerable tensile strength on the root. It also fulfils physio-
logical functions providing the pathways of long-distance transport, i.e. phloem
and xylem, and mechanisms of loading and unloading of these pathways.
194 U. LUTTGE:

2.2.4.1 Long-Distance Transport in Roots

Export bf ions to the shoots obviously depends greatly on the extent of develop-
ment of the xylem; it is very small from the tips of roots and increases in
root zones further up (e.g. Figs. 1 and 2; BURLEY et al. 1970, MARSCHNER and
RICHTER 1973). Comparative physiological and anatomical studies combined
with radioactive or dye labelling suggest that the metaxylem vessels provide
the major pathway for translocation to the shoots (LOTTGE and WEIGL 1962,
BURLEY et al. 1970, ROBARDS and CLARKSON 1976).
LUNDEGARDH'S (1950) classical experiment, in which a cut cylinder of a
wheat root produced exudate from both ends, seems to suggest that there is
no absolute polarity of translocation in the root. In isolated onion root segments,
Cl- appears to be transported in both directions, K + transport is more strongly
upward than downward to the tip, Ca 2 + is transported solely upward, and
Na+ is largely transported downward (MACKLON 1975a, b). SINGH and JACOB-
SON (1977) observed strictly polar upward transport of Na+, Rb+, Ca 2 + and
Cl- to the cut end of excised barley roots. Polarity of translocation in roots
certainly needs more attention in relation to transport mechanisms (SINGH and
JACOBSON 1977). EVANS and VAUGHAN (1966) think that the apparent polarity
of Ca 2 + transport in isolated root segments of maize is due to wounding re-
sponses. Thus, experiments with intact maize seedlings performed by
MARSCHNER and RICHTER (1973) appear to be crucial. K + and Na + are clearly
transported in both directions, but Ca 2 + exclusively moves upward to the shoot
(Fig. 1). This obviously suggests that phloem plays a considerable role in down-
ward transport of ions, since the phloem-immobile ion Ca 2 + is not translocated
in that direction, whereas phloem-mobile K + is moved.

2.2.4.2 Loading of the Xylem Vessels

The mechanism of loading of ions into the xylem vessels has long been the
most outstanding issue in research on ion transport of roots. Transport of
many major nutrient ions from the soil or culture solution into the cells of
the root and into the xylem fluid is against an electrochemical gradient (DUNLOP
and BOWLING 1971 a, b, c, BOWLING and ANSARI 1972, BOWLING 1976, DAVIS
and HIGINBOTHAM 1976, reviews e.g., LUTTGE 1973, CLARKSON 1974, ANDERSON
1976, LUTTGE and HIGINBOTHAM 1979). Thermodynamically, a single active step
in the system would be sufficient to transport ions against a gradient across
the root. CRAFTS and BROYER (1938) envisaged such a single step being the
uptake from the cell wall apoplast across the plasmalemma of root hair, epider-
mis and cortex cells into the symplast of these cells with subsequent passive
transport through the root symplast and diffusive release of ions into the dead
xylem vessels. This hypothesis has found experimental support by ARISZ (1956).
Conversely, it had been suggested that the xylem parenchyma cells might actively
excrete ions into the xylem vessels (SUTCLIFFE 1962 P 120). WEIGL and LUTTGE
(1965) and LXUCHLI (LXUCHLI 1967, LXUCHLI et al. 1971) supported this by
microautoradiographic and X-ray microprobe evidence showing that the label-
ling of the conducting metaxylem vessels was always much higher than in the
1.6 Import and Export of Mineral Nutrients in Plant Roots 195

surrounding xylem parenchyma.· The drawbacks of these cytological methods

have, however, already been mentioned above. Later the groups of LATIES (cf.
LATIES 1967, 1969, LUTTGE 1969) and EpSTEIN (LXUCHLI and EpSTEIN 1971,
LXUCHLI et al. 1971, LXUCHIJ 1972) debated the existence of active transport
only in the root tissue peripheral to the endodermis versus operation of ion
pumps in the stelar parenchyma on the basis of conflicting interpretations of
kinetic results. More recently, the work of PITMAN and coworkers has made
it convincingly clear, however, that two pumps of distinctly different properties
must be involved, one located at the outer face of the symplast taking up
ions from the outer medium into the root cells, and one located at the inner
face of the symplast actively excreting ions into the xylem vessels (reviewed
by LXUCHLI 1976b, PITMAN 1977, LUTTGE and HIGINBOTHAM 1979).
In the overall process of ion transport from the medium across the root
and into the vessels, differentiations of the root surface (Sect. 2.2.1) and the
endodermis (Sect. 2.2.3) are naturally important. In discussions of the actual
loading mechanism, however, variations in structure along the length of roots
were not considered important. Both the passive loading mechanism of the
"one-pump hypothesis" of CRAFTS and BROYER (1938) and the active loading
mechanism of the" two-pump hypothesis" (PITMAN 1977) consider that loading
occurs from the xylem parenchyma cells into fully mature, i.e. cytoplasm-free
and dead xylem vessels.
The evaluation of another hypothesis first conceived by HYLMO (1953) and
then reconsidered by SCOTT (1965), ANDERSON and HOUSE (1967), HIGINBOTHAM
et al. (1973), DANILOVA and STAMBOLTSYAN (1975) and DAVIS and HIGINBOTHAM
(1976), however, requires us to consider xylem element differentiation along
the length of roots in relation to the loading process. In the tip region of
the root the developing vessel elements still contain cytoplasm with organelles,
plasmalemma and tonoplast. In HYLMO'S hypothesis, ions are loaded by trans-
port across the tonoplast into the vacuoles of the immature vessel cells in this
region of the roots. HIGINBOTHAM et al. (1973) observed that in com roots
cytoplasm lined the early metaxylem cell walls up to 10 em behind the root
tip. In a zone 5-10 cm behind the tip the perforation plates of the cross walls
between vessel elements were already formed and the vacuolar sap appeared
to be continuous with the xylem sap in the dead and mature elements. Thus
symplastic transport into the cytoplasm of young xylem elements followed by
active transport across the tonoplast could constitute the loading mechanism
of vessels. According to this hypothesis, transport across followed by transloca-
tion out of the roots should be small in the very tip region, where xylem elements
are not yet discernible, then should reach a peak, where xylem cells develop
but still have plasmalemma and tonoplast, and should decline again further
behind the tip, where xylem is fully mature. This seems not to be born out
by the experiments of Figs. 1 and 2. If such a phenomenon were observed,
it could also be due to variation of physiological activities along the length
of roots (Sect. 3) and not simply to the structural differentiations of the xylem.
Quantitative aspects also argue against a loading only in the limited region
of a vessel, where its elements are still immature. The development of the xylem,
and hence root growth, would need to be unusually rapid (LATIES 1969), or
196 U. LUTTGE:

fluxes across the tonoplast of living vessel cells must be extraordinarily large
(ANDERSON et al. 1970, ANDERSON 1976), or both.
LXUCHLI et al. (1978) have demonstrated that a catalyzed transport mecha-
nism from living xylem parenchyma into dead xylem elements clearly must
be a key feature of the loading process. They used the amino acid analogue
p-fluorophenylalanine (FPA), whose presence in cells during protein synthesis
may lead to formation of ineffective proteins. FPA appears to be a specific
inhibitor of the xylem loading process, because it is incorporated in a protein
of rapid turnover, which is involved in the loading mechanism (SCHAEFER et al.
1975, PITMAN 1977). This protein might be a transport-ATPase as suggested
by the cytological demonstration of K + -stimulated diethylstilbestrol-inhibited
ATPase activity in xylem parenchyma cells adjacent to the dead vessels (WINTER-
SLUITER et al. 1977). FPA inhibited CI- translocation out of barley roots in
zones where all the xylem vessels were mature and not just in those zones
where some immature vessels were present.
The presence of half-bordered pits with relatively thin cell wall areas sited
between the mature, dead vessels and the xylem parenchyma cells with their
comparatively dense cytoplasm (LXUCHLI et al. 1974b) seems to be a structural
feature that would greatly facilitate metabolism-dependent transport.
The loading mechanisms which actively concentrate ions in the xylem are
usually thought to be the driving force for osmotic water flow across the roots
and into the vessels, thus powering root pressure exudation (see ANDERSON
1976) and guttation from leaves with passive hydathodes (DIEFFENBACH et al.
1980a, b). Organic molecules like malate and glutamine transported in the
vessels, however, may also contribute to the driving force of exudation (BUTZ
and LONG 1979).

2.2.4.3 Reabsorption from the Vessels During Translocation in the Root

It seems to be trivial that the composition of the solution translocated in the
xylem vessels will be modified en route not only by secretion into the vessels,
but also by reabsorption from the vessels. This was already demonstrated by
ZIEGLER et al. (1963; see Sect. 2.2.3). JACOBY (1964, 1965) and JACOBY and
RATNER (1974) found that bean and maize roots accumulate Na + in their upper
regions, which allows a certain control of the supply of Na + to the shoot
under conditions of salt stress. LXUCHLI and coworkers have much extended
this comparing the behaviour of pairs or groups of varieties or related species
with different responses to NaCI stress, e.g. the cereals Zea mays and Hordeum
vulgare, and the soybean varieties Glycine max cv. Jackson, cv. Bragg and cv.
Lee and bean (Phaseolus coccineus).
In Zea mays 50 mM NaCI inhibits growth by 30%-40% (LXUCHLI et al.
1976), and concentrations above 75 mM cause poor yield and mortality (yEO
et al. 1977 a). Nevertheless maize obviously can still be grown or kept for a
limited amount of time on 150 mM NaCI or 75 mM Na 2 S0 4 (YEO et al. 1977b).
As shown by X-ray microanalysis under such conditions of stress, the xylem
parenchyma cells in the upper root zones of Zea mays absorb N a + and CI-
from the transpiration stream; xylem parenchyma cells contained high Na +
1.6 Import and Export of Mineral Nutrients in Plant Roots 197

levels and almost no K +; and Na + levels in the stele relative to those in the
cortex were also high (yEO et al. 1977 a). In the mature parts of the roots,
but not in young roots, the cell walls in the areas of the half-bordered pits
between the vessels and the xylem parenchyma cells are strikingly increased
in thickness but of a very loose fibrillar texture (Fig. 9; YEO et af. 1977a).
Thus maize plants respond to salt stress by the attempt to exclude NaCI from
the shoots. Hordeum vulgare can tolerate much higher NaCllevels than maize.
Barley is a salt includer; its response to salt stress is based on mechanisms
in the leaves being able to cope with higher NaCllevels (see WYN JONES et al.
1979). The xylem parenchyma cells of barley always have high K + contents
(YEO et al. 1977b).
The Glycine max varieties Jackson and Bragg are damaged already by 10 mM
NaCl, whereas growth of the variety Lee and of Phaseo/us coccineus is only
inhibited by NaCI concentrations around 50 mM (KRAMER et al. 1977, LXUCHLI
and WIENEKE 1979). X-ray microanalysis showed that the less sensitive variety
and P. coccineus accumulate Na + in the xylem parenchyma cells of the upper
root zones. These xylem parenchyma cells resemble transfer cells (GUNNING
1977) having cell wall protuberances particularly at their half-bordered pits
towards the vessels (Fig. 9b, c). Unlike Na+, CI~ is not reabsorbed from the
transpiration stream but appears to be accumulated in the cortex of the apical
regions of the root (WIENEKE and LXUCHLI 1979).
It remains to be seen whether real salt tolerance in halophytes can be based
not only on salt inclusion but also on salt exclusion. It is quite clear, however,
from the work of LXUCHLI and coworkers that in response to NaCl stress
some glycophytes exchange Na + for K + at the level of the xylem parenchyma
cells (LXUCHLI 1976b). The capacity of the xylem parenchyma cells to accumu-
late Na + must be rapidly exhausted, however, during longer duration of NaCI
stress. Furthermore, one may have to assume that, as in other metabolically
active cells, high levels of Na + and CI- will be incompatible with the cytoplasm
of the xylem parenchyma cells and must be sequestered in the vacuoles, so
that compatible organic solutes will have to be synthesized and accumulated
in the cytoplasm to provide osmotic balance (WYN JONES et al. 1979). NaCI
also could be compartmented in the ER cisternae, which may serve symplastic
movement (STELZER et al. 1975, KRAMER et al. 1977). Therefore, more peripheral
tissues (e.g. the cortex) must participate in the exhange. In fact LXUCHLI (per-
sonal communication in relation to LXUCHLl1979, WIENEKE and LXUCHLl1980)
has suggested a Na + circulation in the roots. This model assumes Na + uptake
in the root zones closer to the tip and Na + release to the medium in those
root zones where xylem parenchyma cells reabsorb Na + from the vessels. Ques-
tions regarding electrical charge balance have so far been left open. Cl- is
taken up together with Na +, but when Na + is exchanged for K + in the'vessels
more distant from the root tip (LXUCHLI 1976b) and CI- is also retained in
these root zones (LXUCHLI and WIENEKE 1979, WIENEKE and LXUCHLI 1979)
the stoichiometry of exchanges appears problematic. Although the work of
LXUCHLI and WIENEKE (1979) clearly shows smaller NaCI levels in the leaves
of the variety Lee as compared with Jackson, i.e. partial salt exclusion in the
less sensitive variety, there are some results which appear to contradict the
198 U. LUTTGE:

Fig. 9a-c. Cell wall modifications in the area of half-bordered pits between xylem vessels
and xylem parenchyma cells. a Cell wall thickening and loose fibrillar texture, maize
root stressed with 50 mM Na Z S0 4 , 250 mrn behind the tip. (yEO et al. 1977a). b Protuber-
ances, root of the soybean cultivar Lee, 155 mrn behind the tip. (Unpublished electron-
micrograph ; see LXUCHLI et al. 1974a). c Protuberances, Phaseo/us coccineus root. (Un-
published electron-micrograph; see KRAMER et al. 1977)

model ; e.g. the upper root parts of variety Lee grown in 50 mM NaCl contain
larger K + levels than plants grown without NaCl, and the upper root parts
of varieties Lee and Jackson contain similar amounts of Na + when both are
grown on 50 mM NaCl (Fig. 2 in LAUCHLI and WIENEKE 1979).
Transfer cells of the xylem parenchyma and of the pericycle are also involved
in the regulation of transport functions of N 2 -fixing root nodules (PATE 1976,
NEWCOMB and PETERSON 1979). It will continue to be stimulating to follow
up the work described above and in Section 2.2.1.2 and obtain an integrated
picture of the distribution and role of transfer cells in angiosperm roots (LET-
VENUK and PETERSON 1976, NEWCOMB and PETERSON 1979).
The quantitative questions need assessment on the basis of exchange mecha-
nisms discussed for the two pumps involved in transport across the root (Sect.
1.6 Import and Export of Mineral Nutrients in Plant Roots 199

2.2.4.2). JESCHKE (1970, 1972, 1977a, b, 1979, JESCHKE and STELTER 1973) has
demonstrated a Na + IK + exchange mechanism operating between barley roots
and the external medium, which is of similar activity in root tips and differen-
tiated root tissue. COLOMBO et al. (1979) showed that this could be a complex
system of two exchange mechanisms, i.e. an active ATP-consuming and electro-
genic H+ extrusion mechanism capable of exchanging H+ for K +, plus aNa + I
H+ exchange, combining together to give a net release of Na + and uptake
of K +. Since most workers now seem to accept the two-pump hypothesis (Sect.
2.2.4.2), the next step in our research must be to characterize in more detail
both the outer and inner pumps. It has been suggested that proton pumping
or charge separation at the site of both pumps may be the driving force for
a number ofuniport, symport (or co-transport) and antiport (or counter-trans-
port) mechanisms whose differential activities or resistances at both faces of
the symplast may be the basis for regulation of ion uptake, accumulation and
export by roots (HANSON 1978, OKAMOTO et al. 1978, 1979). Together with
observations of variation of pumping activities along the length of roots (Sect.
3.3), this may provide an avenue allowing the quantitative evaluation of cycling
models explaining selectivity.

3 Variations of Physiological Activities Along

the Length of Roots

3.1 Growth, Differentiation and Hormonal Gradients

Different physiological and biochemical activities are associated with cell divi-
sion, extension growth and differentiation. This brings about variations of physi-
ological features along the length of roots. The question here is to what extent
and in which way this can influence transport physiologically, i.e. apart from
the structural interactions discussed in Section 2.
Astonishingly little work has been devoted to this aspect. JESCHKE and
STELTER (1976) have considered the different relative cytoplasm contents of
cells in various root zones. Cells in the tip meristem and below the extension
zone contain much more cytoplasm relatively to the cell volume than cells
further behind the tip. It is known that cytoplasmic reaction systems require
certain K + levels, but that Na + levels must be kept low, and Na + is sequestered
in the vacuoles (WYN JONES et al. 1979). K + INa + ratios in the root tip are
always very high, even in K + -free media. Presumably K + can be transported
in the phloem to the root tips from reserves in other parts of the seedlings
(RICHTER and MARSCHNER 1973). Na + IK + exchange mechanisms are operative
in the more distant root zones (see Sect. 2.2.4.3). On the basis of cytometric
considerations for the various root zones JESCHKE and STELTER (1976) then
arrive at cytoplasmic and vacuolar K + concentrations of 110 and 20 mM, re-
spectively, i.e. K + levels in the cytoplasm high enoug;h for biochemical reaction
systems.
200 U. LUTTGE:

An outstanding physiological factor is the existence of hormonal gradients

in roots, because growth and development of roots is regulated by phytohor-
mones transported from the shoots into the roots and also in the opposite
direction out of the tips towards more distant root zones. lndoleacetic acid
(IAA) is transported from the tops of plants downwards. Other hormones may
be transported in both directions although their synthesis may be restricted
to either roots or shoots. Abscisic acid (ABA) is largely formed in shoots but
can also be synthesized in roots (WALTON et al. 1976), and transport upwards
and downwards can occur both in the phloem and xylem of plants (BELLANDI
and DORFFLING 1974, ZEEVAART 1977). Cytokinins, gibberellic acid, and un-
known inhibitors are exported from the root tips (ATKIN et al. 1973). Cytokinin
synthesis is restricted to the meristematic initials in the very tip of roots (VAN
STADEN and DAVEY 1979). Cytokinins are transported, however, both from
roots to shoots and from shoots to roots. Transport from the shoot is probably
based on cytokinins originally synthesized in the root tips but stored in the
form of glycosides in the leaves. Kinetins can be translocated in the xylem
and phloem, and retranslocated in the xylem and phloem, and retranslocation
and cycling in the whole plant is important (YONK 1979, VAN STADEN and
DAVEY 1979). The ethylene precursor 1-aminocyclopropane-1-carboxylic acid
is synthesized in roots and translocated in the xylem to the shoot where conver-
sion to ethylene takes place (BRADFORD and YANG 1980). Thus all classes of
hormones (auxins, gibberellins, abscisic acid, cytokinins, and ethylene) can be
transported in the xylem (BRADFORD and YANG 1980, KING 1976). Phytohor-
mones and their effects on transport processes in plants are discussed in Volume
2.B.7, this Series, and so this can be only touched here. All the phytohormones
mentioned above are known to be involved in control and regulation of mem-
brane-electric phenomena and of transport in plant cells and tissues (VAN STE-
VENINCK 1976). It seems to be noteworthy though, that naturally occurring
hormonal gradients have hardly been considered yet as a physiological basis
of differences of bioelectric and transport features along the length of roots.

3.2 Bioelectrical Fields Along Roots

B.I.H. SCOTT and coworkers (review: SCOTT 1967) have described bioelectrical
fields along growing roots, which have recently been reinvestigated by WEISEN-
SEEL et al. (1979). Figure 10 shows that a current is directed inwards in the
meristem and elongation zones. Current also enters at the tips of root hairs.
Current leaves in the root hair zone across the epidermal surface beneath the
root hairs. The root cap has either inward or outward current. Experiments
with roots imbedded in bromocresol purple-stained agar reveal a 'Colour-banding
phenomenon corresponding to the current distribution (Fig. 10), which suggests
that much of the current consists of hydrogen ions. Hydrogen ions leak into
growing cells or cell parts and are pumped out of non-growing ones. The biologi-
cal function of these currents may be related to growth (WEISENSEEL et al. 1979),
but they also determine patterns of ion uptake along the length of roots (SCOTT
1967).
1.6 Import and Export of Mineral Nutrients in Plant Roots 201

Fig. 10. Pattern and density of current Rool Hair Zone

traversing growing barley roots (growth Rool Cap Merislem Main Elonga lion Zone .~\\\\
rate 5.1 ± 0.7 ).lm min - 1) in solutions of ,
1 mM NaCI, 0.1 mM KCI and 0.1 mM I

CaCl 2 or in 0.5 mM CaCI 2 , pH 5.2- 5.7. r----~

':'E
Width of the bars represents the spread "-'/" "~I',

I I
u
of the measuring positions; "zero a.
zone " refers to current almost entirely E
• 2 ~
flowing parallel to the surface. Errors "" 1
are SEM. (WEISENSEEL et al. 1979)

..
..,
)
1
~H-
.5 Z.ro Zon. I'm from Ti p
o 1000 2000 3000
T
J

3.3 Differences in Ion Transport Mechanisms Along Roots

Assimilate supply from the shoot to various root zones can vary and is tempera-
ture-dependent. This may be one reason for variations in ion transport along
the roots (ROVIRA and BOWEN 1973). ESHEL and WAISEL (1972, 1973) observed
variations in the uptake of sodium along roots of corn and barley seedlings,
not only in terms of transport rates but also with regard to qualitative differences
in the responses to temperature and ion concentration. Electrical measurements
also show that different parts of roots have different physiological transport
properties (HELMY et al. 1973).
Differences in Fe uptake capacity along roots have already been mentioned
in relation to epidermal transfer cells formed in apical root zones in response
to Fe stress (Sect. 2.2.1.2). In maize Fe translocation along the length of roots
is rather uniform (KASHIRAD et al. 1973); in contrast, the rates of translocation
in barley are very much higher in a zone 1--4 cm from the root tip than elsewhere
in the root, and the differences seem to be related to metabolic factors and
not to root anatomy (CLARKSON and SANDERSON 1978). Since Fe can only be
absorbed in the divalent ferrous state, the release of reductants or in situ reduc-
tion could be involved and might be a phenomenon preferentially located in
the younger root tissues (BROWN 1978).
In different zones of soybean roots, TRAVIS et al. (1979) have found different
activities of the plasmalemma-associated K +-stimulated ATPase, which is
thought to be an integral part of ion uptake systems of higher plant roots
(HODGES 1976). ATPase activity was low in the region of the root cap, incteased
to a maximum in the meristematic region, decreased to a minimum as cell
elongation proceeded, and then increased as lateral root development began.
This ATPase could be an H + extrusion pump exchanging H + for K + and
driving an H +-Na + antiport (COLOMBO et al. 1979, Sect. 2.2.4.3). However,
maximum ATPase activity in soybean roots (TRAVIS et al. 1979) does not occur
in zones similar to those of barley roots where a strong active H + extrusion
202 U. LUTTGE:

was observed (WEISENSEEL et al. 1979, Fig. 10). Almost the inverse relation seems
to apply. It would be very interesting to perform both kinds of study with
the same material and under comparable conditions.

4 Root-Shoot Interactions and Circulation in the Whole Plant

4.1 Some Examples mustrating General Aspects of Circulation

Transport from root to shoot is often investigated by analysis of exudation

from the cut ends of isolated roots that are held either in air, with a capillary
attached, or immersed in solution in a separate compartment of a transport
chamber (e.g. LXUCHLI et al. 1978). It can also be studied by analysis of guttation
fluid, which is driven out of passive hydathodes by root pressure (e.g. DIEFFEN-
BACH et al. 1980a, b), or by shoot analysis when the solutes involved can only
arrive in the shoot via the root. One-way considerations of root-shoot interac-
tions focus on supply of water and inorganic ions from root to shoot in the
xylem and of assimilates from shoot to root in the phloem. In addition the
transport of phytohormones (see Sect. 3.1) and of yet unknown factors is in-
volved. Root factors can regulate photosynthetic activity in leaves (CARMI and
KOLLER 1979). A shoot factor is needed for the development of nitrite uptake
machinery in roots, in which protein synthesis is involved (JACKSON et al. 1974).
The shoot affects transport out of the root directly via passive driving forces
and indirectly via metabolism-dependent driving forces. In addition to root
pressure built up by ion pumping in the root, transpiration is the driving force
for ascent of water and solutes in the plant. Only in smaller plants is root
pressure by itself able to drive the ascent of sap, when the water potential
gradient between the plant and the atmosphere is close to zero. Normally tran-
spiration is the predominant driving force. This passive force is regulated by
active stomatal control in the shoots. Somewhat less directly, the rate of transpi-
ration may also affect ion transport across the root and xylem loading (ANDER-
SON 1976). It has been clearly shown that the supply of metabolic substrates
from shoot to root is essential for the functions of root cells in ion uptake
and translocation, including the maintenance of membrane potentials of root
cells (PITMAN 1972, HATRIcKand BOWLING 1973, GRAHAM and BOWLING 1977).
When relative growth rate of barley seedlings is reduced by limited daily photo-
period (2 h illumination in 24 h) the metabolism-dependent ion transport ma-
chineries in the roots are much more affected than in the leaves (PITMAN 1972,
PITMAN et al. 1974a).
An example of hormonal interaction has been given by selective wilting
of shoots of barley seedlings, whose roots were sealed in a moist chamber
that prevented water flow to the shoots but kept the roots turgid. When brief
wilting of the shoots was reversed by submerging the roots again in solution,
the physiological functions in the shoots recovered within a very short period
but 86Rb + export from the roots was inhibited for up to 3 h after the onset
of recovery (PITMAN et al. 1974b). It seems that some factor regulating ion
1.6 Import and Export of Mineral Nutrients in Plant Roots 203

export by the roots, presumably abscisic acid formed in the wilting leaves,
has been transported from shoot to root. After water stress, abscisic acid is
readily distributed throughout plants via phloem and xylem (HoAD 1978).
This already implies that the one-way considerations are over-simplifications.
Many solutes are mobile both in the xylem and phloem. They can be'retranslo-
cated in the phloem and in the xylem, respectively, arriving in the shoot via
the xylem and in the root via the phloem. Furthermore, the phloem supplies
sinks not only downwards from the mature exporting leaves but also upwards
in the growing parts of shoots. Shoot-root competition for nutrients, carbon
skeletons and metabolites can be excellently documented by the study of utiliza-
tion of seed reserves in germinating seedlings during the first week of develop-
ment (SUTCLIFFE 1976).
In Trifolium alexandrinum, which is a pasture plant used in arid regions
on water- and salt-stressed soils, a retranslocation of Na + seems to control
Na + levels in the leaves. This plant can withstand NaCl-salinity up to 50 mM.
Na + supplied from the roots in the transpiration stream is reabsorbed by special
transfer cells in the phloem of the petioles of the trifoliate leaves and retranslo-
cated to the roots and into the medium (WINTER 1982a, b; WINTER and LXUCHLI
1982; WINTER and PREsTON 1982).
Interesting limitations to this are set by the immobility of some important
solutes in the phloem. It has long been known that Ca2+ is rather immobile
(see ZmGLER 1975). The pH of the phloem sap is always slightly alkaline (PH
7-8,5; ZmGLER 1975). RAVEN (1977) has drawn attention to this in relation
to the cytoplasmic nature of sieve tube contents which never allows large devia-
tion from this pH value. Thus the export of surplus H+ or OH- equivalents,
generated during metabolism in the leaves, is not possible in the phloem. RAVEN
has pointed out that in evolution this may have contributed to specialization
and division of labour between plant organs.

4.2 Nitrogen, Sulphur and Phosphorus

At low NO; concentrations in the root medium, NO; is largely reduced in

the roots, but at higher concentrations (0.8-8 mM) it needs to be transported
to the shoots for reduction (WALLACE and PATE 1967, KIRKBY and KNIGHT
1977, ROBIN et al. 1979). DUKSHOORN (1970) and BEN ZIONI et al. (1971) devel-
oped a cycling model for the cooperation of roots and shoots in this mechanism:
KN0 3 is transported in the xylem to the shoots, where NO; is reduced to
NH3 and OH-; OH- is neutralized by malic acid synthesis by CO 2 fixation
via phosphoenolpyruvate carboxylase; K +-malate is transported in the phloem
to the roots, where malate is decarboxylated; the resulting HCO; then ex-
changes for the NO; taken up from the root medium and the resulting pyruvate
is used as a substrate for root metabolism. For various reasons this model
has been subject to criticism. KIRKBY and ARMSTRONG (1980) suggest that it
does not operate in plants like tomato, sugar cane and tobacco, where the
ratio. of cation and anion uptake by roots is close to unity and little HCO;
is released, but that it seems to work in castor bean ..
204 U. LUTTGE:

Notwithstanding the criticisms, the DIJKSHOORN-BEN ZIONI hypothesis seems

to continue to stimulate research. Sulphur is a case similar to that of nitrogen.
Uptake at the root level is in the oxidized form as sulphate, but SOi- reduction
in photo autotrophic plants is light-enhanced and largely confined to leaves
(KYLIN 1960, SCHWENN and TREBsT 1976, RENNENBERG et al. 1979). As in NO;
reduction, OH - is formed in SOi - reduction. Reduced sulphur is exported
from the leaves of tobacco plants in the phloem in the form of glutathione
(67%-70%), methionine (27%-30%) and cysteine (2%-8%) (RENNENBERG et al.
1979).
The third major nutrient, phosphorus, which is available as an inorganic
anion, is different, because it is used in metabolism in the same oxidized form,
i.e. as phosphate, in which it is taken up by the root. Shoot-root interactions
are also important, however, and phosphate-deficiency experiments suggest that
phosphate transport in the root is under control of the shoot (CLARKSON et al.
1978c).

5 Conclusion

Among the higher plants there is a large morphological variety of roots, depend-
ing on special functions like anchoring, water and nutrient uptake, formation
of symbiotic systems, storage, etc. Roots of a given plant species are, however,
also extraordinarily complex with their three-dimensional pattern of structural
and physiological differentiations. Variations in time, e.g. due to changing envi-
ronmental conditions of nutrient and micronutrient supply, salinity and water
stress, add to the complexity of responses. But only in such a way can roots
be versatile enough to satisfy the many nutritional requirements of plants, as
excellently and concisely reviewed by CLARKSON and HANSON (1980). Physiologi-
cal activities of roots need more attention as possible bases for ecological adapta-
tions.
Acknowledgement. I thank Dr. DETLEF KRAMER for reading the manuscript, critical discus-
sions and for his help in evaluating some of the literature.

References

Anderson WP (1976) Transport through roots. In: Liittge U, Pitman MG (eds) Transport
in plants II, part B, tissues and organs. Encyclopedia of plant physiology New Ser,
vol II. Springer, Berlin Heidelberg New York
Anderson WP, Higinbotham N (1975) A cautionary note on plant root electrophysiology.
J Exp Bot 26: 533-536
Anderson WP, House CR (1967) A correlation between structure and function in the
root of Zea mays. J Exp Bot 18:544-555
Anderson WP, Aikman DP, Meiri A (1970) Excised root exudation: a standing gradient
osmotic flow. Proc R Soc London Ser B 174:445-458
Anderson WP, Robertson RN, Wright BJ (1977) Membrane potentials in carrot root
cells. Aust J Plant PhysioI4:241-252
1.6 Import and Export of Mineral Nutrients in Plant Roots 205

Arisz WH (1956) Significance of the symplasm theory for transport in the root. Proto-
plasma 46:5-62
Atkin RK, Burton GE Robinson DK (1973) Effect of root-growing temperature on
growth substances in xylem exudate of Zea mays. J Exp Bot 24:475-487
Bange GG (1973) Diffusion and absorption of ions in plant tissue. III. The role of
the root cortex cells in ion absorption. Acta Bot N eerl 22: 529-542
Bellandi DM, Dorffling K (1974) Transport of abscisic acid-2-C-14 in intact pea seedlings.
Physiol Plant 32: 365-368
Ben Zioni A, Vaadia Y, Lips SH (1971) Nitrate uptake by roots as regulated by nitrate
reduction products of the shoot. Physiol Plant 24: 288-290
Bowen GD (1969) The uptake of orthophosphate and its incorporation into organic
phosphates along roots of Pinus radiata. Aust J BioI Sci 22: 1125-1135
Bowen GD (1970) Effects of soil temperature on root growth and on phosphate uptake
along Pinus radiata roots. Aust J Soil Res 8: 31-42
Bowling DJF (1976) Ionic gradients in higher plant tissues. Persp Exp Bioi 2:391-399
Bowling DJF, Ansari AQ (1972) Control of sodium transport in sunflower roots. J
Exp Bot 23:241-246
Bradford KJ, Yang SF (1980) Xylem transport of 1-aminocyc1opropane-1-carboxylic
acid, an ethylene precursor, in waterlogged tomato plants. Plant Physiol 65: 322-326
Brown JC (1978) Mechanism of iron uptake by plants. Plant Cell Environ 1 :249-257
Burley JWA, Nwoke FlO, Leister GL, Popham RA (1970) The relationship of xylem
maturation to the absorption and translocation of 32p. Am J Bot 57: 504-511
Butz RG, Long RC (1979) L-Malate as an essential component of the xylem fluid of
corn seedling roots. Plant Physiol 64: 684-689
Capesius I, Barthlott W (1975) Isotopen-Markierungen und rasterelektronenmikrosko-
pische Untersuchungen des Velamen radicum der Orchideen. Z Pflanzenphysiol
75:436-448
Carmi A, Koller D (1979) Regulation of photosynthetic activity in the primary leaves
of bean (Phaseolus vulgaris L.) by materials moving in the water-conducting system.
Plant Physiol 64: 285-288
Clarkson DT (1974) Ion transport and cell structure in plants. McGraw-Hill, London
Clarkson DT, Hanson JB (1980) The mineral nutrition of higher plants. Annu Rev
Plant Physiol 31 :239-298
Clarkson DT, Sanderson J (1969) The uptake of a polyvalent cation and its distribution
in the root apices of Allium cepa: tracer and autoradiographic studies. Planta
89:136-154
Clarkson DT, Sanderson J (1978) Sites of absorption and translocation of iron in barley
roots. Tracer and micro auto radiographic studies. Plant Physiol 61: 731-736
Clarkson DT, Sanderson J, Scott-Russell R (1968) Ion uptake and root age. Nature
220:805-806
Clarkson DT, Robards AW, Sanderson J (1971) The tertiary endodermis in barley roots:
fine structure in relation to radial transport of ions and water. Planta 96: 292- 305
Clarkson DT, Sanderson J, Robards AW (1978a) Permeability studies on the epidermal/
hypodermal layers of the sand sedge (Carex arenaria L.). Agric Res Counc Letcombe
Lab, Annu Rep 1977:25
Clarkson DT, Robards AW, Sanderson J, Peterson CA (1978b) Permeability studies
on epidermal-hypodermal sleeves isolated from roots of Allium cepa (onion). Can
J Bot 56: 1526-1532
Clarkson DT, Sanderson J, Scattergood CB (1978c) Influence of phosphate-stress on
phosphate absorption and translocation by various parts of the root system of
Hordeum vulgare L. (barley). Planta 139:47-53
Colombo R, Bonetti A, Lado P (1979) Promoting effect of fusicoccin on Na + efflux
in barley roots: evidence for a Na +-H+ antiport. Plant Cell Environ 2: 281-285
Crafts AS, Broyer TC (1938) Migration of salts and water into xylem of the roots
of higher plants. Am J Bot 24:415-431
Danilova MF, Stamboltsyan EY (1975) The ultrastructure of differentiating primary
xylem cells of the root in relation to hypothesis of transfer of solutes into xylem
elements. Bot Zh 60:913-926
206 U. LiiTrGE:

Davis RF, Higinbotham N (1976) Electrochemical gradients and K + and Cl- fluxes
in excised corn roots. Plant Physiol 57: 129-136
Dieffenbach H, Kramer D, Liittge U (1980a) Release of guttation fluid from passive
hydathodes of intact barley plants. I. Structural and cytological aspects. Ann Bot
45:397-401
Dieffenbach H, Liittge U, Pitman MG (1980b) Release of guttation fluid from passive
hydathodes of intact barley plants. II. The effects of abscisic acid and cytokinins.
Ann Bot 45: 703-712
Dijkshoorn W (1970) Partition of ionic constituents between organs. Proc 6th Int ColI
Plant Anal Fert Probl, Tel Aviv
Dumbroff EB, Peirson DR (1971) Probable sites for passive movement of ions across
the endodermis. Can J Bot 49: 35-38
Dunlop J, Bowling DJF (1971 a) The movement of ions to the xylem exudate of maize
roots. I. Profiles of membrane potential and vacuolar potassium activity across the
root. J Exp Bot 22: 434-444
Dunlop J, Bowling DJF (1971 b) The movement of ions to the xylem exudate of maize
roots. II. A comparison of the electrical potential and electrochemical potential of
ions in the exudate and in the root cells. J Exp Bot 22: 445-452
Dunlop J, Bowling DJF (1971c) The movement of ions to the xylem exudate of maize
roots. III. The location of the electrical and electrochemical potential differences
between the exudate and the medium. J Exp Bot 22: 453-464
Epstein E (1976) Kinetics of ion transport and the carrier concept. In: Liittge U, Pitman
MG (eds) Transport in plants II, part B, tissues and organs. Encyclopedia of plant
physiology New Ser, vol II. Springer, Berlin Heidelberg New York
Eshel A, Waisel Y (1972) Variations in sodium uptake along primary roots of corn
seedlings. Plant Physiol 49: 585-589
Eshel A, Waisel Y (1973) Variations in uptake of sodium and rubidium along barley
roots. Physiol Plant 28:557-560
Espelie KE, Kolattukudy PE (1979) Composition of the aliphatic components of suberin
of the endodermal fraction from the first internode of etiolated Sorghum seedlings.
Plant PhysioI63:433-435
Evans EC, Vaughan BE (1966) Wounding response in relation to polar transport of
radiocalcium in isolated root segments of Zea mays. Plant Physiol41: 1145-1151
Ferguson IB (1979) The movement of calcium in non-vascular tissue of plants. Commun
Soil Sci Plant Anal 10: 217-224
Ferguson IB, Clarkson DT (1975) Ion transport and endodermal suberization in the
roots of Zea mays. New Phytol 75: 69-79
Ferguson IB, Clarkson DT (1976a) Simultaneous uptake and translocation of magnesium
and calcium in barley (Hordeum vulgare L.) roots. Planta 128:267-269
Ferguson IB, Clarkson DT (1976b) Ion uptake in relation to the development of a
root hypodermis. New Phytol 77:11-14
Glass ADM, Perley JE (1979) Cytoplasmic streaming in the root cortex and its role
in the delivery of potassium to the shoot. Planta 145:399-401
Glinka Z (1977) Effects of abscisic acid and of hydrostatic pressure gradient on water
movement through excised sunflower roots. Plant Physiol 59: 933-935
Graham RD, Bowling DJF (1977) Effect of the shoot on the transmembrane potentials
of root cortical cells of sunflower. J Exp Bot 28: 886-893
Green JR, Northcote DH (1979) Location offucosyl transferases in the membrane system
of maize root cells. J Cell Sci 40: 235-244
Grunwaldt G, Ehwald R, Pietzsch W, Goring H (1979) A special role Qfthe rhizodermis
in nutrient uptake by plant roots. Biochem Physiol Pflanzen 174:831-837
Gunning BES (1977) Transfer cells and their roles in transport 'of solutes in plants.
Sci Prog Oxford 64:539-568
Haas DL, Carothers ZB, Robbins RR (1976) Observations on the phi-thickenings and
casparian strips in Pelargonium roots. Am J Bot 63: 863-867
Hanson JB (1978) Application of the chemiosmotic hypothesis to ion transport across
the root. Plant PhysioI62:402-405
1.6 Import and Export of Mineral Nutrients in Plant Roots 207

Harrison-Murray RS, Clarkson DT (1973) Relationship between structural development

and the absorption of ions by the root system of Cucurbita pepo. Planta 114: 1-16
Hatrick AA, Bowling DJF (1973) A study of the relationship between root and shoot
metabolism. J Exp Bot 24:607-613
Helmy AK, Peinemann N, Ferreiro EA (1973) Effect of root position on measured
root membrane potentials. J Exp Bot 24: 29-32
Higinbotham N, Davis RF, Mertz SM, Shumway LK (1973) Some evidence that radial
transport in maize roots is into living vessels. In: Anderson WP (ed) Ion transport
in plants. Academic Press, London New York
Hoad GV (1978) Effect of water stress on abscisic acid levels in white lupin (Lupinus
albus L.) fruit, leaves and phloem exudate. Planta 142:287-290
Hodges TK (1976) ATP!!Ses associated with membranes of plant cells. In: Liittge U,
Pitman MG (eds) Transport in plants II, part A, Cells. Encyclopedia of plant physiolo-
gy, New Ser, vol II. Springer, Berlin Heidelberg New York
Hylmo B (1953) Transpiration and ion absorption. Physiol Plant 6:333-405
Iren van F, Sluijs van der BP (1980) Symplasmic and apoplasmic radial ion transport
in plant roots. Cortical plasmalemmas lose absorption capacity during differentiation.
Planta 148: 130-137
Iren van F, Spiegel van der A (1975) Subcellular localization of inorganic ions in plant
cells by in vivo precipitation. Science 187:1210-1211
Jackson WA, Johnson RE, Yolk RJ (1974) Nitrite uptake by nitrogen-depleted wheat
seedlings. Physiol Plant 32: 37-42
Jacoby B (1964) Function of bean roots and stems in sodium retention. Plant Physiol
39:445-449
Jacoby B (1965) Sodium retention in excised bean stems. Physiol Plant 18:730-739
Jacoby B, Ratner A (1974) Mechanism of sodium exclusion in bean and corn plants
- a reevaluation. In: Wehrmann J (ed) Plant analysis and fertiliser problems. Ger
Soc Plant Nutr, Hannover
Jeschke WD (1970) Evidence for a K +-stimulated Na + effiux at the plasmalemma of
barley root cells. Planta 94:240-245
Jeschke WD (1972) Wirkung von K+ auf die Fluxe und den Transport von Na+ in
Gerstenwurzeln, K +-stimulierter Na +-Effiux in der Wurzelrinde. Planta 106:73-90
Jeschke WD (1977a) K+-Na+ exchange and selectivity in barley root cells, effects of
K +, Rb+, Cs+, and Li+ on the Na + fluxes. Z PflanzenphysioI84:247-264
Jeschke WD (1977b) K +-Na + exchange and selectivity in barley root cells: effect of
Na+ on the Na+ fluxes. J Exp Bot 28:1289-1305
Jeschke WD (1979) New K +-Na + exchange across the plasmalemma of meristematic
root tissues. Z Pflanzenphysiol 94: 325-330
Jeschke WD, Stelter W (1973) K +-dependent net Na + effiux in roots of barley plants.
Planta 114:251-258
Jeschke WD, Stelter W (1976) Measurement of longitudinal ion profiles in single roots
of Hordeum and Atriplex by use of flameless atomic absorption spectroscopy. Planta
128:107-112
Kashirad A, Marschner H, Richter C (1973) Absorption and translocation of 59Fe from
various parts of the corn root. Z Pflanzererniihr Bodenk 134: 136-147
King RW (1976) Implications for plant growth of the transport of regulatory compounds
in phloem and xylem. In: Wardlaw JF, Passioura JB (eds) Transport and transfer
processes in plants. Academic Press, London New York
Kirkby EA, Armstrong MJ (1980) Nitrate uptake by roots as regulated by nitrate ~ssirni­
lation in the shoot of castor oil plants. Plant Physiol 65: 286-290
Kirkby EA, Knight AH (1977) Influence of the level of nitrate nutrition on ion uptake
and assimilation, organic acid accumulation and cation-anion balance in whole
tomato plants. Plant PhysioI60:349-353
Kramer D, Uiuchli A, Yeo AR, Gullasch J (1977) Transfer cells in roots of Phaseolus
coccineus: ultrastructure and possible function in exclusion of sodium from the shoot.
Ann Bot 41: 1031-1040
Kramer D, Anderson WP, Preston J (1978) Transfer cells in the root epidermis of A triplex
208 U. LiiTTGE:

hastata L. as a response to salinity: a comparative cytological and X-ray microprobe

investigation. Aust J Plant Physiol 5: 739-747
Kramer D, Romheld V, Landsberg E, Marschner H (1980) Induction of transfer-cell
formation by iron deficiency in the root epidermis of Helianthus annuus L. Planta
147:335-339
Kylin A (1960) The incorporation of radio-sulphur from external sulphate into different
sulphur fractions of isolated leaves. Physiol Plant 13: 366--379
Laties GG (1967) Metabolic and physiological development in plant tissues. Aust J Sci
30:193--203
Laties GG (1969) Dual mechanisms of salt uptake in relation to compartmentation and
long-distance transport. Annu Rev Plant PhysioI20:89-116
Liiuchli A (1967) Untersuchungen iiber Verteilung und Transport von Ionen in Pflanzen-
geweben mit der Rontgen-Mikrosonde. I. Versuche an vegetativen Organen von Zea
mays. Planta 75:185-206
Liiuchli A (1972) Translocation of inorganic solutes. Annu Rev Plant Physiol23: 197-218
Liiuchli A (1976a) Apoplasmic transport in tissues. In: Liittge U, Pitman MG (eds)
Transport in plants II, part B. Tissues and' organs. Encyclopedia of plant physiology
New Ser, vol II. Springer, Berlin Heidelberg New York
Liiuchli A (1976b) Symplasmic transport and ion release to the xylem. In: Wardlaw
IF, Passioura JB (eds) Transport and transfer processes in plants. Academic Press,
London New York
Liiuchli A (1979) Regulation des Salztransportes und SalzausschlieBung in Glykophyten
und Halophyten. Ber Dtsch Bot Ges 92: 87-94
Liiuchli A, Epstein E (1971) Lateral transport of ions into the xylem of corn roots.
I. Kinetics lind energetics. Plant Physiol 48: 111-117
Liiuchli A, Wieneke J (1979) Studies on growth and distribution of Na+, K+ and Cl-
in soybean varieties differing in salt tolerance. Z Pflanzenerniihr Bodenkd 142: 3--13
Liiuchli A, Spurr AR, Epstein E (1971) Lateral transport of ions into the xylem of
corn roots. II. Evaluation of a stelar pump. Plant Physiol48: 118-124
Liiuchli A, Kramer D, Stelzer R (1974a) Ultrastructure and ion localization in xylem
parenchyma cells of roots. In: Zimmermann U, Dainty J (eds) Membrane transport
in plants. Springer, Berlin Heidelberg New York
Liiuchli A, Kramer D, Pitman MG, Liittge U (1974 b) Ultrastructure of xylem parenchy-
ma cells of barley roots in relation to ion transport to the xylem. Planta 119:85-99
Liiuchli A, Pitman MG, Liittge U, Kramer D, Ball E (1978) Are developing xylem
vessels the site of ion exudation from root to shoot? Plant Cell Environ 1 :217-223
Liiuchli A, Yeo T, Kramer D (1976) Distribution of soluble inorganic ions in maize
roots; microprobe analysis of deep frozen specimens. 6th Eur Congr Electron Microsc,
Jerusalem
Letvenuk LJ, Peterson RL (1976) Occurrence of transfer cells in vacuolar parenchyma
of Hieraciumflorentinum roots. Can J Bot 54: 1458-1471
Lundegardh H (1950) The translocation of salts and water through wheat roots. Physiol
Plant 3: 103--151
Liittge U (1964) Mikroautoradiographischer Nachweis der Aufnahme von 35SO;- in
die Wurzelhaare von Ahornkeimlingen. Naturwissenschaften 12:296--297
Liittge U (1969) Aktiver Transport (Kurzstreckentransport bei Pflanzen). Protoplasmato-
logia VIII 7b; 1-146
Liittge U (1973) Stofftransport der Pflanzen. Springer, Berlin Heidelberg New York
Liittge U, Higinbotham N (1979) Transport in plants. Springer, Berli~ Heidelberg New
York
Liittge U, Weigl J (1962) Mikroautoradiographische Untersuchungen der Aufnahme und
des Transportes von 35SO; - und 45Ca + + in Keimwurzeln von Zea mays L. und
Pisum sativum L. Planta 58: 113--116
Liittge U, Weigl J (1964) Der Ionentransport in intakten und entrindeten Wurzeln. Ber
Dtsch Bot Ges 77: 63--70
MacKenzie KAD (1979) The development of the endodermis and phi layer of apple
roots. Protoplasma 100:21-32
1.6 Import and Export of Mineral Nutrients in Plant Roots 209

Macklon AES (1975 a) Cortical cell fluxes and transport to the stele in excised root
segments of Allium cepa L. I. Potassium, sodium and chloride. Planta 122: 109-113
Macklon AES (1975b) Cortical cell fluxes and transport to the stele in excised root
segments of Allium cepa L. II. Calcium. Planta 122: 131-141
Malone CP, Barke JJ, Hanson JB (1977) Histochemical evidence for the 9ccurrence
of oligomycin-sensitive ATPase in corn roots. Plant Physiol 60: 916-922
Marre E (1979) Fusicoccin: a tool in plant physiology. Ann Rev Plant PhysioI30:273-288
Marschner H, Richter C (1973) Akkumulation und Translokation von K +, Na + und
Ca 2 + bei Angebot zu einzelnen Wurzelzonen von Maiskeimpflanzen. Z Pflanzen-
ernaehr Bodenkd 135: 1-15
Marschner H, Kalisch A, Romheld V (1974) Mechanism of iron uptake in different
plant species. Proc 7th Int Colloq Plant Anal Fert Probl, Hannover 2: 274-281
Newcomb W, Peterson RL (1979) The occurrence and ontogeny of transfer cells asso-
ciated with lateral roots and root nodules in Leguminosae. Can J Bot 57: 2583-2602
Okamoto H, Ichino K, Katon K (1978) Radial electrogenic activity in the stem of Vigna
sesquipedalis: involvement of spatially separate pumps. Plant Cell Environ 1 : 279-284
Okamoto H, Katon K, Ichino K (1979) Distribution of electric potential and ion transport
in the hypocotyl of Vigna sesquipedalis. VI. The dual structure of radial electrogenic
activity. Plant Cell PhysioI20:103-114
Pate JS (1976) Transport in symbiotic systems fixing nitrogen. In: Liittge U, Pitman
MG (eds) Transport in plants II, part B. Tissues and organs. Encyclopedia of plant
physiology New Ser, vol II. Springer, Berlin Heidelberg New York
Peterson CA, Emanuel ME, Weerdenburg CA (1981) The permeability of phi thickenings
in apple (Pyrus malus) and geranium (Pelargonium FlOrtorum) roots to an apoplastic
fluorescent dye tracer. Can J Bot 59: 11 07-111 0
Pitman MG (1965) Sodium and potassium uptake by seedlings of Hordeum vulgare.
Aust J BioI Sci 18:10-24
Pitman MG (1972) Uptake and transport of ions in barley seedlings. III. Correlation
between transport to the shoot and relative growth rate. Aust J BioI Sci 25: 905-919
Pitman MG (1976) Ion uptake by plant roots. In: Liittge U, Pitman MG (eds) Transport
in plants II, part B, Tissues and organs. Encyclopedia of plant physiology New Ser,
vol II. Springer, Berlin Heidelberg New York
Pitman MG (1977) Ion transport into the xylem. Annu Rev Plant Physiol28 :71-88
Pitman MG, Liittge U, Uiuchli A, Ball E (1974a) Ion uptake to slices of barley leaves,
and regulation ofK content in cells of the leaves. Z Pflanzenphysiol72:75-88
Pitman MG, Liittge U, Lauchli A, Ball E (1974b) Effect of previous water stress on
ion uptake and transport in barley seedlings. Aust J Plant Physiol1 :377-385
Pitman MG, Schaefer N, Wildes RA (1975) Relation between permeability to potassium
and sodium ions and fusicoccin-stimulated hydrogen-ion efflux in barley roots. Planta
126:61-73
Pitman MG, Anderson WP, Liittge U (1976) Transport processes in roots. In: Liittge
U, Pitman MG (eds) Transport in plants II, part B. Tissues and organs. Encyclopedia
of plant physiology New Ser, vol II. Springer, Berlin Heidelberg New York
Plaut M (1910) Untersuchungen zur Kenntnis der physiologischen Scheiden bei Gym-
nospermen, Equiseten und Bryophyten. Jahrb Wiss Bot XLVII: 121-185
Raven JA (1977) H+ and Ca 2 + in phloem and symplast. Relation of relative immobility
of the ions to the cytoplasmic nature of the transport paths. New Phytol 75: 465-480
Rennenberg H, Schmitz K, Bergmann L (1979) Long-distance transport of sulfur in
Nicotiana tabacum. Planta 147: 57-62
Richter Ch, Marschner H (1973) Umtausch von Kalium in verschiedenen Wurzelzonen
von Maiskeimpflanzen. Z Pflanzenphysiol 70: 211-221
Richter Ch, Marschner H (1974) Verteilung von K +, Na + und Ca + + zwischen Wurzel-
rinde und Zentralzylinder. Z Pflanzenphysiol 71: 95-100
Robards AW, Clarkson DT (1976) The role of plasmodesmata in the transport of water
and nutrients across roots. In: Gunning BES, Robards AW (eds) Intracellular commu-
nication in plants: studies on plasmodesmata. Springer, Berlin Heidelberg New York
Robards AW, Newman TM, Clarkson DT (1980) The distinctive nature of the plasma-
210 U. LUTTGE:

membrane of the endodermis in roots as shown by freeze-fracture electron microscopy.

Letcombe Lab Annu Rep 1979, Agric Res Counc, Wantage
Robin P, Blayac D, Salsac L (1979) Influence de l'alimentation nitrique sur la teneur
en nitrate et l'activite nitrate reductase des racines et des feuilles de plantules de
mais. Physiol Veg 17: 55-56
Rovira AD, Bowen GD (1970) Translocation and loss of phosphate along roots of
wheat seedlings. Planta 93: 15-25
Rovira AD, Bowen GD (1973) The influence of root temperature on 14C assimilate
profiles in wheat roots. Planta 114:101-107
Russell RS, Sanderson J (1967) Nutrient uptake by different parts of the intact roots
of plants. J Exp Bot 18:491-508
Schaefer N, Wildes RA, Pitman MG (1975) Inhibition by p-fluorophenylalanine of
protein synthesis and of ion transport across the roots in barley seedlings. Aust J
Plant Physiol 2: 61-73
Schwenn JD, Trebst A (1976) Photosynthetic sulfate reduction by chloroplasts. In: Barber
J (ed) The intact chloroplast. Elsevier, Amsterdam
Scott BIH (1967) Electric fields in plants. Annu Rev Plant Physiol18 :409-418
Scott FM (1965) The fine structure of xylem vessels. In: Baker KF, Snyder WC (eds)
Ecology of soilborne pathogens. U niv Cal Press, Berkeley
Scott MG, Peterson RL (1979a) The root endodermis in Ranunculus acris. I. Structure
and ontogeny. Can J Bot 57:1040-1062
Scott MG, Peterson RL (1979b) Histochemistry of the endodermis and the synthesis
of phenolic compounds in roots. Can J Bot 57: 1063-1077
Shone MGT, Clarkson DT, Sanderson J (1969) The absorption and translocation of
sodium by maize seedlings. Planta 86:301-314
Singh C, Jacobson L (1977) Polar movement of ions in barley roots. Physiol Plant
39:37-78
Smith FA, Raven JA (1979) Intracellular pH and its regulation. Annu Rev Plant Physiol
30:289-311
Smith RC (1970) Time course of exudation from excised corn root segments of different
stages of development. Plant Physiol 45: 571-575
Staden van J, Davey JE (1979) The synthesis, transport and metabolism of endogenous
cytokinins. Plant Cell Environ 2: 93-106
Stelzer R, Liiuchli A (1977) Salz- und Uberflutungstoleranz von Puccinellia peisonis.
II. Strukturelle Differenzierung der Wurzel in Beziehung zur Funktion. Z Pflanzen-
physiol 84: 95-1 08
Stelzer R, Liiuchli A, Kramer D (1975) IntrazelluliireTransportwege in Wurzeln intakter
Gerstenpflanzen. Cyto biologie 10: 449-457 .
Steveninck van RFM (1976) Effect ofhorrnones and related substances on ion transport.
In: Liittge U, Pitman MG, (eds) Transport in plants II, part B. Tissues and organs.
Encyclopedia of plant physiology New Ser, vol II. Springer, Berlin Heidelberg New
York
Strasburger E (1923) Das botanische Praktikum. Fischer, Jena
Sutcliffe JF (1962) Mineral absorption in plants. Pergamon, New York London
Sutcliffe JF (1976) Regulation in the whole plant. In: Liittge U, Pitman MG (eds) Trans-
port in plants II, part B. Tissues and organs. Encyclopedia of plant physiology New
Ser, yol II. Springer, Berlin Heidelberg New York
Travis RL, Geng S,. Berkowitz RL (1979) Analysis of the distribution of potassium-
stimulated adenosine triphosphatase activity in soybean root. Plant Physiol
63:1187-1190
Troll W, Hahn K (1973) Allgemeine Botanik. Enke, Stuttgart
Ursprung A, Blum G (1921) Zur Kenntnis der Saugkraft. IV. Die Absorptionszone
der Wurzel. Der Endodermissprung. Ber Dtsch Bot Ges 39: 70-79
Vakhmistrov DB (1967) On the function of apparent free space in plant roots. A study
of the absorbing power of epidermal and cortical cells in barley roots. Sov Plant
Physiol14: 103-107
1.6 Import and Export of Mineral Nutrients in Plant Roots 211

Vonk CR (1979) Origin of cytokinins transported in the phloem. Physiol Plant

46:235-240
Wallace W, Pate JS (1967) Nitrate assimilation in higher plants with special reference
to the cocklebur (Xanthium pennsylvanicum Wallr.). Ann Bot 31 :213-228
Walton DC, Harrison MA, Cote P (1976) The effects of water stress on abscisic-acid
levels and metabolism in roots of Phaseolus vulgaris L. and other plants. Planta
131: 141-144
Weigl J, Liittge U (1962) Mikroautoradiographische Untersuchungen iiber die Aufnahme
von 35SO';:- - durch Wurzeln von Zea mays L. Die Funktion der primiiren Endodermis.
Planta 59: 15-28
Weigl J, Liittge U (1965) Die Ionenaufnahme durch die Luftwurzeln von Epidendrum.
Protoplasma 60: 1-6
Weisenseel MH, Dorn A; Jaffe LF (1979) Natural H+ currents traverse growing roots
and root hairs of barley (Hordeum vulgare L.). Plant Physiol 64: 512-518
Wiebe HH, Kramer PJ (1954) Translocation of radioactive isotopes from various regions
of roots of barley seedlings. Plant Physiol 29: 342-348
Wieneke J, Liiuchli A (1979) Short-term studies on the uptake and transport of Cl-
by soybean cultivars differing in salt tolerance. Z Pflanzenernaehr Bodenkd 142:799-
814
Wieneke J, Liiuchli A (1980) Effects of salt stress on distribution of Na + and some
other cations in two soybean varieties differing in salt tolerance. Z Pflanzenernaehr
Bodenkd 143: 55-67
Winter E (1982a) Salt tolerance of Trifolium alexandrinum L. II. Ion balance in relation
to its salt tolerance. Aust J Plant PhysioI9:227-237 -
Winter E (1982b) Salt tolerance of Trifolium alexandrinum L. III. Effects of salt on
ultrastructure of phloem and xylem transfer cells in petioles and leaves. Aust J Plant
PhysioI9:239-250 .
Winter E, Liiuchli A (1982) Salt tolerance of Trifolium alexandrinum L. I. Comparison
of the salt response of T. alexandrinum and T. pratense. Aust J Plant Physiol9: 221-226
Winter E, Preston J (1982) Salt tolerance of Trifolium alexandrinum L. IV. Ion measure-
ments by X-ray microanalysis in unfIxed, frozen hydrated leaf cells at various stages
of salt treatment. Aust J Plant Physiol 9: 251-259
Winter-Sluiter E, Liiuchli A, Kramer D (1977) Cytochemical localization of K +-stimu-
lated adenosine triphosphatase activity in xylem parenchyma cells of barley roots.
Plant Physiol 60: 923-927
Wong WL, Ap Rees T (1971) Carbohydrate oxidation in the stele and cortex isolated
from roots of Pisum sativum. Biochim Biophys Acta 252: 296-304
Wyn Jones RG, Brady CJ, Speis J (1979) Ionic and osmotic relations in plant cells.
In: Laidman DL, Wyn Jones RG (eds) Recent advances in the biochemistry of cereals.
Academic Press, London New York
Yeo AR, Kramer D, Liiuchli A, Gullasch J (1977a) Ion distribution in salt-stressed
mature Zea mays roots in relation to ultrastructure and retention of sodium. J Exp
Bot 28:17-29
Yeo AR, Liiuchli A, Kramer D, Gullasch J (1977b) Ion measurements by x-ray micro-
analysis in unfIxed, frozen, hydrated plant cells of species differing in salt tolerance.
Planta 134: 35-38
Zeevaart JAD (1977) Sites of abscisic acid synthesis and metabolism in Ricinus communis
L. Plant Physiol 59: 78&-791
Ziegler H (1975) Nature of transported substances. In: Zimmermann MH, Milburn JA
(eds) Transport in plants I, phloem transport. Encyclopedia of plant physiology New
Ser vol I. Springer, Berlin Heidelberg New York
Ziegler H, Wei~ J, Liittge U (1963) Mikroautoradiographischer Nachweis der Wande-
rung von 5S0';:- - durch die Tertiiirendodermis der Iris-Wurzel. Protoplasma
56:362-370
1.7 Cycling of Elements in the Biosphere
C.C. DELWICHE

1 The Sources of Plant Constituents

1.1 Soil and Atmospheric Sources

Of the 18 or more elements required by plants or beneficial to their development,

only carbon has its immediate source in the atmosphere. Except in a few special
cases such as lichens and some aerial plants, other required elements are obtained
from the soil. One exception to this is the fixation of N 2 by some symbiotic
associations (Chapt. 11.2). Under most circumstances the availability of elements
is limited, however, and so the death and decomposition of plant material is
essential to the maintenance of the supply of necessary elements for new growth
(Goszet al. 1973).
The system is not a completely closed one, and there is a continual loss
from the soil by volatilization, erosion and leaching. The replenishment of ele-
ments lost in this manner is required to maintain productivity. The atmosphere
is the principal vehicle by which some of these elements are returned to the
soil from the sea or other reservoirs where they have been concentrated. This
process is not a complete one either. Some elements, such as phosphorus, which
is present in seawater only at very low concentrations, must be provided by
other means. Because phosphorus forms some compounds of exceedingly low
solubility (explaining its low concentration in seawater), its mobility in soil
is low and so the rate of its replenishment need not be as great as that of
a more mobile element such as sulfur, even though the requirements of the
two elements by plants are of the same order of magnitude (VAN WAZER 1961).
Thus the weathering of soil materials although it takes place slowly under most
circumstances suffices to resupply phosphorus as well as some other elements
to the system.

1.2 The Weathering Process

Any effort to classify the mineral elements required by plants introduces some
inconsistencies, but in considering the weathering process as a contributor to
the nutrient requirements of plants it is helpful to distinguish between those
elements which are volatile or form volatile compounds such as carbon, oxygen,
hydrogen, nitrogen, sulfur and the halides. (and which are of relatively low
concentration in igneous rocks), as contrasted with the other elements which
are igneous rock constituents. The sedimentary rocks that provide the parent
material for many soils can have higher concentrations of the volatile elements.
1. 7 Cycling of Elements in the Biosphere 213
2
+ +
Fig. 1. Relative quantities of various
elements in plants, animals, igneous H t
C +
H
C +
0
CI
No *
rocks and seawater on a dry weight ot Si +
basis. Note that the scale is logarith-
mic o + AI +
Ma+
N + Na+ C
S +
Fe*M~
K + Co=!:
0 Cat K
Nt
P +
C +
K+ Kts Ti + N +
-I Na+ Br +
Cat
.. +CI P + F +
C Mg::: P AI +
.. Mg+ Mn+
~ S +
S + Si +
11. Fe+ B +
e0 -2
<i
...
0
Znt
.J CI+ Cut
Fe+
-3
Mn+

Zn+ P +
-4 Mnt
Cut
Cut Mo+

-5 Zn +
Mo+
I + I +
Cut

Plants Animals Igneous Sea

Rocks Water

Carbon and sulfur are major constituents of some sedimentary materials

(GARRELS and MACKENZm 1971).
Figure 1 gives a comparison of the concentration of certain elements in
biological materials, igneous rocks and seawater. It is not surprising that biologi-
cal materials are made up of those elements readily available in the environment
in which they evolved and roughly in proportion to their availability. What
is of interest is the manner in which some elements commonly available in
abundance have been shunned by the biological world and, if present in plants
or animals, are there only as passive or apparently inadvertent constituents.
An example of these is aluminum, which is quite ubiquitous but which serves
no known biological role.
"Weathering" is a broad term including both the physical processes resulting
from alternate heating and cooling, wind and water erosion, and other mechani-
cal processes and the chemical processes of solution and reprecipitation. Ele-
ments liberated in the weathering process become available to plants, and plants
are active participants in the weathering process. The organic constituents of
plants and the products of microbial degradation of plant material participate
214 c.c. DELWICHE:

in and accelerate the chemical weathering process. As a consequence, the weath-

ering process is itself a function of the vegetation present. The vegetation is
determined by the nature of the weathering products as well as other climatic
and biotic factors. Some of these intricate interrelationships will be discussed
later.

2 The Nature of Cycles

2.1 The Hydrologic Cycle

The concept of cyclic natural processes probably has always been a part of
philosophical thought. As our understanding of natural processes has expanded,
and as we have gained an appreciation for the significance of the approximately
5 x 109 years that the Earth has existed, we continue to refine our concepts
of these cycles. Depending upon our interests or the degree of our myopia
we can define any number of cycles based on elements, organisms or processes.
Our primary interest here is to deal in det~il with some of the elemental cycles
of significance to plants. But before we can treat these effectively it is necessary
to give brief consideration to the hydrologic cycle, the sedimentary cycle and
the magmatic cycle. All of these cycles are related and each influences the
others. The biological world (and therefore the biological cycles) are strongly
dependent upon the hydrologic, sedimentary and magmatic cycles and in tum
have strongly influenced these. Corollary to this concept is the realization that
all of the cycles are now quite different than they were in early pre-Cambrian
times prior to the evolution of life on the planet. Moreover, as the characteristics
of life have changed through evolution so has the nature of these cycles been
altered.
The hydrologic cycle, perhaps the first to have been recognized as such,
not only dominates the biological world, but is one of the most important
determinants of the character of the globe.
Most of the water of the Earth that we can recognize (about 1.4 x 10 18 m 3
or 80% of the total) is in the oceans. Pore water of sedimentary rocks, the
next largest fraction, constitutes about 3.2 x 10 17 m 3 (18.8%). Another 1.2%
is tied up in ice. Fresh-water lakes and rivers make up about 0.002% and
the atmosphere contains a small 0.006% of the total. About 0.026% of the
total falls in precipitation every year (and an equivalent amount is evaporated
annually). Although annual evaporation (and rainfall) equals about 1 part in
30,000 of the water of the hydrosphere, it would be misleading to use this
figure to estimate the time constant of the hydrologic cycle. Othef'Superimposed
cycles, such as the circulation of the oceans and the storage of water in glaciers,
have periods of thousands of years, as does the storage of water underground
in some places. The cycle involving pore water of sediments lasts millions of
years. Thus, although the quantity of water which is in the atmosphere and
that which falls as precipitation on land are small compared with the total,
they are vital to plants not only because of the water that is supplied, but
I. 7 Cycling of Elements in the Biosphere 215

because of the weathering and erosional processes that result and the elements
which are transported from the atmosphere to the soil by rainfall (PENMAN
1973).
The processes of evaporation and transport of water provide the driving
force for circulation and mixing of the atmosphere; and the effect of clouds
and precipitation on the reflectivity of the Earth are major moderating factors
in determining the temperature of the continents. Water vapor in the atmosphere
serves to limit radiative loss of heat, narrowing the temperature differentials
between seasons and from day to night. Although one could visualize the adap-
tation of life to extremes of temperature, the unique characteristics of water
as a solvent make it particularly suited as a medium for life, and by inference
suggest an ambient temperature within its liquid range. Higher atmospheric
pressures could expand this range but not indefinitely.
Thus, the hydrologic cycle can be looked upon as a steady state with minor
fluctuation (on a geologic time-scale as well as the short-term fluctuation which
we refer to as "weather").

2.2 The Sedimentary Cycle

Because of the length of time over which it has been operating, the sedimentary
cycle also can be viewed as something near a steady-state condition with the
weathering of old sediments and the formation of new sediments taking place
at approximately equal rates. As new igneous rocks are exposed by weathering
or brought to the surface by vulcanism, they are also subject to weathering
and contribute to the sedimentary burden. Sedimentary rocks constitute about
75% of the total rock exposed on the continents and so their contribution
to soil and plants is a major one (GARRELS and MACKENZIE 1971). The difference
in composition of sedimentary and igneous rocks, as given in Table 1, would
suggest a considerable difference in the composition of soils coming from these
two sources. Differences are recognizable in soils and in the vegetation they
support, but these are not as great as the composition figures alone would
suggest. Climatic and biotic factors are also important and the plant population
tends to regulate the characteristics of the soil and the mineral elements available
in it. In some cases differences in elemental composition caused by processes
of sedimentation and metamorphosis can result in deficiencies of some essential
elements or toxic concentrations of others which in tum limit vegetation. Under
these conditions the vegetation mosaic will reflect the differences in tolerance,
deficiency or toxicity involved. The most familiar examples of this sort are
those resulting from extremes of water availability or total salt concentration,
but other more subtle manifestations are common. ,
The processes of weathering and sediment formation include the mechanical
and chemical weathering of the parent material, the transfer of the products
of weathering either laterally or downward, and the formation of secondary
precipitates including secondary clay minerals. These small particulates are
transferred to other locations, eventually precipitate and result in the formation
of new sediments. The new sediments usually contain additional materials of
216 C.c. DELWICHE:

Table 1. A comparison of the elemental composition of various rock types. (Derived

from the data of HORN and ADAMS 1966, TAYLOR 1964 and POLDERVAART 1955)

Element I!g atoms g-l in

Igneous rocks Shales Sandstones Limestones

0 29,000 30,200 30,700 31,000

Si 10,000 2,600 13,000 850
Al 3,040 2,960 920 150
H 1,400 5,600 1,800 860
Ca 1,030 550 975 7,530
Na 1,030 420 140 17
Fe 1,000 850 180 70
Mg 960 600 290 1,900
K 540 680 270 70
Ti 120 96 31 8.3
P 34 25 5.5 13
F 33 40 14 17
Mn 17 15 9.1 20
C 17 1,275 32 19
S 8.1 75 75 37
CI 3.6 5.1 0.28 4.2
N 1.4 41 6.4
Zn 11 1.5 0.25 0.31
B 0.91 9.1 3.2 1.9
Co 0.42 0.32
Mo 0.015 0.027

biological ongm such as the skeletal remains of micro flora and amorphous
organic material.
The burial of carbon in sediments is an important part of the geobiological
cycle, contributing to the accretion of oxygen in the atmosphere. This will be
discussed in more detail later.
Compared with our estimates of the time constant of the hydrologic cycle,
the sedimentary cycle is a long one with the mean life of sediments being in
the neighborhood of 3.5 x 10 8 years. This is still less than 1/ 10 the age of the
Earth but it is long in comparison with the rate of biological evolution. For
this reason and because biological processes strongly influence the sedimentary
cycle, we would expect to find in the sedimentary record some evidence of
this biological change. The most striking biological effect should be in the rate
of weathering but there should also be recognizable differences in the chemical
composition of weathering products. The first great biological invention, of
course, was life itself. Presumably this supplied a new source of chelating agents
which would hasten mobilization and transport of metallic ions, particularly
the less soluble ones such as iron. It also resulted in the increased rate of
burial of reduced carbon resulting from photosynthesis. This made possible
the accumulation elsewhere of more oxidized products of photosynthesis, first
sulfur and then oxygen. These will be discussed in more detail when the cycling
of these particular elements is considered.
1. 7 Cycling of Elements in the Biosphere 217

2.3 The Magmatic Cycle

Superimposed upon the sedimentary and hydrologic cycles, influencing both

of them and in turn being influenced by both is the magmatic cycle, whereby
through the processes of ocean floor spreading and subduction, magmatic mate-
rial is transported across the ocean floors and buried under the continental
plates. This cycle has an approximate time constant of 200 million years, of
the same order of magnitude as that of the sedimentary cycle. Along with
the burial of the basaltic ocean floor some of the sedimentary burden which
it had picked up on its trek across the ocean is also buried. This results in
the deep burial of reduced carbon and other materials of the ocean floor mud,
further contributing to the possibility of accretion of oxygen in the atmosphere
and indirectly altering the composition of sedimentary materials. Interstitial
water and dissolved salts included in the sediments are also buried, altering
the composition of the ocean and the sediments.
At some later time, and presumably with a shorter time constant, the volatile
constituents of this buried material, particularly H 2 0, CO 2 , H 2 S, and assorted
volatile nitrogenous compounds, are returned to the surface by continental vul-
canism and again are able to participate in the hydrologic, sedimentary, and
geobiological cycles. The altered composition of the sediments from the charac-
teristics of the original igneous or sedimentary materials from which they were
derived would be expected to be evident in a differing composition of lava
from these continental volcanos as contrasted with that of oceanic lava flows.
Such differences in lava composition are discernible but confused by other pro-
cesses of fractionation in the magma and the differences in depth from which
various lava flows originate.
The above description is a highly simplified synthesis of a complex process.
It is presented here only as backdrop for a discussion of the geobiological
processes and cannot be dealt with in detail.

2.4 The Geobiological Cycles

The appearance of life on the planet influenced the hydrologic, sedimentary

and magmatic cycles, at first only modestly but to an increasing extent as succes-
sive biological innovations evolved (BRODA 1975b). The ancient sedimentary
record is vague but evidence is strong that some simple life forms existed at
least 2.7 x 109 years ago. From virtually the beginning this life had to be powered
by solar energy. In other words, photosynthesis of one sort or another has
always been a part of life though it need not have been the three basic classifica-
tions of photosynthesis we now recognize, the photolytic splitting of 9rganic
compounds, hydrogen sulfide, or water. More probably, these three common
classifications of photosynthesis are simply manifestations of electron sources
for the photosynthetic process.
Photosynthesis made possible the dismutation of available compounds into
some which were more oxidized and some which were more reduced than ther-
modynamic equilibration would dictate. Thus, the combination of carbon
218 c.c. DELWICHE:

dioxide and water could be converted into more reduced carbon (carbohydrate
or hydrocarbon) and more oxidized water (0 2 ). Carbon dioxide and hydrogen
sulfide could be converted to more reduced carbon (carbohydrate or hydrocar-
bon) and more oxidized hydrogen sulfide (sulfur). The existence of such a ther-
modynamic disequilibrium in tum made possible the existence of heterotrophic
organisms which gain their energy by carrying on the reverse process, the oxida-
tion of the more reduced carbohydrate or hydrocarbon with the more oxidized
sulfur or oxygen or other electron acceptors.
Other elements from the weathering of igneous rocks such as reduced iron
provided additional electron sources for reduction of the available electron ac-
ceptors. Before any appreciable accretion of oxygen in the atmosphere could
take place, ferrous iron in the oceans or in the soil would first have to have
been oxidized.
The biological cycle of photosynthesis producing oxygen and the hetero-
trophic oxidation of the products of photosynthesis is a rapid one compared
with the sedimentary cycle. For this reason the accretion of oxygen in the
atmosphere was dependent more on the rate at which reduced compounds of
biological origin could be buried than it was upon the rate of photosynthesis
itself. Initially photosynthesis was largely in the oceans and other bodies of
water or on soils which were regularly kept moist. In either case the rate of
photosynthesis was limited by the availability of other elements necessary to
support life and so the total photosynthetic rate could not have been much
greater than the present marine rate which presumably is phosphorus-limited.
The one great biological innovation which changed this, shifted the center
of gravity of photosynthesis to the continents and resulted in a much greater
overall photosynthetic rate, was the evolution of vascular tissue. Once plants
with a vascular system evolved it became possible to mine the soil more deeply
for mineral elements necessary to support life and thereby made possible large
plants with a higher photosynthetic capability. This in tum greatly slowed physi-
cal weathering, made possible the accumulation of a significant soil layer on
the continents and slowed the rate at which the more reduced material from
the weathering of igneous rocks was delivered to the oceans as a sink for atmo-
spheric oxygen. This remarkable development took place during Silurian and
Devonian times and resulted in an Earth similar to the one we see today.
It is at this point that we can undertake to piece together available informa-
tion on the cycling of elements of biological interest with reasonable confidence
that although these cycles have been altered by the course of evolution they
have been operating with their essential features as we now know them for
nearly 500 million years.
It is not possible to assign any hierarchy of importance to the various ele-
ments essential for plant growth. Each element is utilized, transported, precipi-
tated or otherwise transformed in a manner unique to its chemistry and biologi-
cal function (DEEVEY 1970). The cycles of some have received more attention
from the scientist than others, either because of their complexity, their intrinsic
interest, or their susceptibility to modification by human activity. The cycles
of hydrogen, oxygen, and carbon have not been subject to much study from
an agronomic stand-point aside from the obvious problems of water manage-
1.7 Cycling of Elements in the Biosphere 219

ment and a few studies of carbon dioxide enrichment of atmospheres over

plants. Recently the cycle of carbon has become of interest, not as a plant
physiological problem but because of the potential implications which the
burning of fossil fuels and release of carbon dioxide may have to the atmospheric
level of CO 2 and the Earth's heat budget.
Many of the elements have comparatively simple cycles in that they are
taken up by the plant from the soil, perform their function in the plant and
are returned to the soil upon death and decay of the plant or animal material.
Nitrogen and sulfur have more complex cycles for several reasons. In addition
to their functional role in the plant they are participants in the energy metabo-
lism of microorganisms in the soil and are alternately oxidized or reduced de-
pending upon soil conditions. The nitrogen cycle is made even more complex
by the release of nitrogen gas to the atmosphere where it is no longer available
to most plants and must be "fixed" before it can again participate in biological
processes.
Phosphorus is of interest because in many environments, particularly the
ocean, the availability of phosphorus is a primary limitation on productivity.

3 The Nitrogen Cycle

3.1 Overall Cycle Features

In addition to the characteristics which the nitrogen cycle shares with most
other elements of the biosphere, the chemistry of nitrogen is such as to make
the cycle considerably more complex than most (SODERLUND and SVENSSON
1976, DELWICHE 1970). The nitrogen cycle is outlined in Figs. 2,3.
Nitrogen forms several gaseous compounds including several oxides, ammo-
nia, some volatile amines and nitrogen gas (N2). Consequently, the atmosphere
plays an important role in the nitrogen cycle with gaseous nitrogen being by
far the major nitrogen pool excluding nitrogen which is buried in sediments.
Secondly, the oxidation-reduction range of nitrogen from a valence of + 5 in
nitrate ion to one of - 3 in ammonia and amino nitrogen, is well within the
oxidation range of all plants and most microorganisms (Table 2). As a conse-
quence nitrogen undergoes alternate oxidation and reduction in a number of
energy reactions of plants and microorganisms in addition to its structure and
catalytic roles in all organisms.
The nitrogen cycle can be looked upon as two superimposed cycles, one
involving the soil-plant-soil cycle already mentioned which is a compa~atively
rapid one with the mean turnover time being a few decades or less depending
upon the ecosystem involved, and the other involving the loss of gaseous nitro-
gen to the atmosphere. This nitrogen is no longer available to most plants,
with only a few microorganisms or in some cases microorganisms associated
with plants (or some animals) being capable of fixing atmospheric nitrogen
and making it available to the biosphere. Because the atmospheric reservoir
220 c.c. DELWICHE:
ATMOSPHERE
N2 2.BXIOB

.,,,,, ~

)~~~~~r\.c
~I.nd
fif!
N2 Fi ••
oce.n

(§
Oenit ri fication

OCEAN

\'"'
ANIMALS 14

o
IGNEOUS ROCK
--_.-
".
Fig. 2. Detailed nitrogen cycle. Reservoirs (pools) are shown in rectangles and transfer
rates in circles. Units used are Gigagram atoms of N (10 12 gram atoms). Many values
for the nitrogen cycle are uncertain, as for example, ocean processes, the land and ocean
reservoirs of organic N and others. 01 alues given here are obtained from various sources:
SODERLUND and SVENSSON 1976, HARDY and HAVELKA 1975, DELWICHE 1970)

.,.,.,...----
/,,"-'"
Atmospheric ,,- ,...'--'=------'
N- fixation /
/ /
/ /
I
I I
I I
I I
I i'
~
/
Denitrification
shunt
/
/
/
/
,.../

Fig. 3. Schematic outline of the nitrogen cycle. Processes shown by solid lines are typical
of many elements. Those shown by broken lines are unique to the nitrogen cycle
I. 7 Cycling of Elements in the Biosphere 221

Table 2. Oxidation states of nitrogen and typi-

cal compounds of interest to the biologist

Valence Typical compound

+5 NO;- Nitrate ion

+3 NO; Nitrite ion
+1 [RNO] Nitroxyl
o N2 Nitrogen Gas
-1 HONH 2 Hydroxylamine
-3 NH3 Ammonia

is so large this second portion of the cycle is a ponderous one requiring 2 x 10 7

years or more before a given molecule is returned to soil or the sea. Actually,
because this cycle is superimposed upon the other sedimentary magmatic cycles
mentioned earlier, the residence time of a molecule in the atmosphere is even
greater.
Overall elements of the nitrogen cycle are outlined ih Fig. 2.

3.2 Nitrification

Most of the nitrogen of plants and animals is at the reduced or - 3 valence

level of ammonium or amino nitrogen. When it is released to the soil by the
decomposition of plant and animal tissues, and if oxygen is present, there is
energy available in the oxidation of ammonium ion to nitrate ion and a number
of "nitrifying" organisms use this as their energy source for growth (ALEEM
1977, ALEXANDER 1965). These organisms are classically looked upon as "au-
totrophs" because they do not require organic compounds for growth (ALEEM
1970). There are some "heterotrophic" organisms (ones which require an
organic substrate) which are also capable of nitrification (ALEXANDER et al.
1960). As a result of this nitrifying process most of the inorganic nitrogen
in soil is as nitrate ion (NO;) which is the soil nitrogen species most likely
to be taken up by plants. Once taken up by the plant this nitrate must again
be reduced requiring the expenditure of energy either from the oxidation of
organic materials or utilizing directly the energy captured in photosynthesis.
There are other consequences of the nitrification reaction. Ammonium ion
as a cation is more likely to be held on the exchange complex of the soil
near its surface, particularly if the soil has significant quantities of organic
matter or layered clay minerals. Once oxidized to nitrate, however, the nitrogen
is more mobile and more readily moved through the soil by rainfall where
it is delivered into the root zone. If plants are absent, as under fallow coriditions,
the nitrate ion is more likely to be leached through the root zone and appear
in groundwaters unless some other fate befalls the ion. When nitrogen fertilizers
containing ammonium ion are used, efforts are taken to minimize this possibility
of loss. These include the timing and placement of nitrogenous fertilizers, the
222 c.c. DELWICHE:

use of "slow release" fertilizers which liberate the ammonium ion to the soil
slowly, or the use of agents to inhibit the activity of nitrifying organisms.
Since ammonium ion, whether from the decomposition of organic materials
or from applied nitrogen fertilizers, is in partial equilibrium with gaseous ammo-
nia, there is also the possibility of a volatilization of ammonia to the atmosphere.
Because ammonia is highly water-soluble, its residence time in the atmosphere
is short and it is soon returned to the soil in rainfall or in combination with
sulfate ion or other anions as dry fallout.
The nitrification reaction is usually thought of as taking place in two steps
because of the nature of the metabolic processes involved. The first of these
is the oxidation of ammonium ion to nitrite ion (N02). This reaction is carried
on by several species of Nitrosomonas and several other genera including some
heterotrophs as mentioned above (BARTHOLOMEW and CLARK 1965).

NHt +1.5 O 2 -N0 2 +H 20+2 H+

LlG' (at pH 7)~ -65 kcal.
The second step in the reaction, the oxidation of nitrite ion to nitrate is
carried on by organisms of the genus Nitrobacter or other organisms including
some heterotrophs.

N0 2 + 0.5 O 2 - NO;
LlG' (at pH 7)~ -18 kcal.
Nitrous oxide (N 20) is also sometimes produced in the nitrification reaction
(BREMNER and BLACKMER 1978), under some circumstances in considerable
quantities. Since the nitrogen of nitrous oxide is intermediate in valence level
between that of nitrate and ammonia, the formation of gaseous nitrous oxide
short circuits the nitrification sequence, not only depriving Nitrobacter of a
substrate but also losing nitrogen from the system. This may escape to the
atmosphere or, as will be discussed below, may participate in other microbial
reactions. Nitrous oxide production in nitrification may be more prevalent in
the ocean than it is on land (HAHN 1974).

3.3 Denitrification
In addition to the possibility of being lost to groundwaters by leaching, nitrate
ion may undergo other reactions which exclude it from plants. If there is organic
material present and the oxygen supply is low, as for example in waterlogged
soils or highly eutrophied waters, the nitrate can play a quite different role.
Some microorganisms, in the absence of sufficient oxygen, can utilize nitrate
ion instead of oxygen as "electron acceptor" for the oxidation of organic materi-
als in order to obtain their energy. In this process the nitrate ion is reduced
first to nitrite and eventually to nitrogen gas or nitrous oxide (DELWICHE and
BRYAN 1976). Both ofthese commonly escape to the atmosphere, the N 2 entering
the large atmospheric pool and the nitrous oxide a much smaller and much
shorter-lived atmospheric pool. The atmosphere contains about 0.35 parts per
I. 7 Cycling of Elements in the Biosphere 223

million N 2 0 on a volume basis. Much of this, perhaps most, is of biological

origin although there are other processes in the atmosphere which can result
in N 2 0 production (ROBINSON and ROBBINS 1970). Its mean residence time
in the atmosphere is comparatively short, between 10 and 100 years (JUNGE
1972,1974). Among the atmospheric reactions in which nitrous oxide can parti-
cipate are those involving the formation and destruction of ozone (0 3 ), The
function of ozone as an ultraviolet shield makes of interest the effect which
any human influence on the rate of production of N 2 0 might have. The use
of nitrogenous fertilizers and the concentration of organic wastes in urban areas
and other management practices presumably could increase the rate of N 20
released to the atmosphere. Some practices such as the draining of wetlands
could result in a lessened rate ofN 2 0 production. Some of these will be discussed
in more detail below.
There are other manifestations of denitrification including its role in the
problem of "bloat" in ruminants and gas production in foods and other stored
materials.
Nitric oxide (NO) is also a possible product of denitrification, particularly
under acid conditions. When plant materials which are high in nitrate are sub-
jected to fermentation as in the ensiling process, nitric oxide, which is highly
toxic, may be formed in concentrations sufficient to be hazardous to humans.
Silo fillers' disease, an example of this category, occurs frequently enough to
be of clinical interest and is sometimes fatal.

3.4 Nitrogen Fixation

Nitrogen fixation (or dinitrogen fixation) is discussed in detail in Chaps. II.1
and II.2, this volume and is dealt with only briefly here. To replace the nitrogen
which is lost from the soil or from waters by denitrification, sedimentation
or other processes, new nitrogen from the atmosphere must be fixed and brought
into the biological system (DELWICHE 1970, BRILL 1977). Ionizing events in
the atmosphere such as electrical discharge and cosmic radiation result in the
formation of some compounds of nitrogen (largely oxides) which are brought
to the earth by rainfall. This replenishes some of the nitrogen supply. Most
of the requirement is met by biological reactions. Organisms such as Azotobacter
free living or in association with grasses such as Paspalum are capable of fixing
nitrogen under aerobic conditions and these make some contribution to soil
nitrogen. With some tropical grasses this contribution may be a major one.
In environments low in oxygen such as rice paddies and swamps, anaerobic
or microaerophilic microorganisms, also free-living, can make sizable contribu-
tions. Examples here are various clostridia and other bacteria. Blue-green algae,
growing either independently or in lichen associations also fix nitrogen.
Perhaps the greatest contribution to the biological pool, however; is the
nitrogen fixed by microbial association with higher plants. Although there are
a number of such associations, the best-known are those of the legumes which
bear root nodules supporting Rhizobium. There are an estimated 600 genera
and 13,000 species of legumes, most of them presumably capable of nitrogen
fixation in this sort of symbiotic arrangement. Much of the tropical rain forest
224 C.c. DELWICHE:

is populated with legumes and the widespread use of agricultural legumes has
long been a contributor to the human food supply.
A considerable number of other higher plants representing at least 10 plant
families also bear root nodules and are capable of nitrogen fixation in associa-
tion with microorganisms (BoND 1976, TORREY 1978). In most cases the microor-
ganism is an actinomycete, but this group has been studied in much less detail
than have the legumes and our knowledge of them is limited (TORREY 1976).
They are, however, significant contributors to the nitrogen budget of many
plant associations, particularly the chaparral and deciduous forest lands of tem-
perate zones. Attention is now being given this group, and their nitrogen fixation
potential for both food and fiber production undoubtedly will be exploited.
The transfer of the nitrogen fixation capability to other plants, particularly
cereals, by genetic manipulation is a possibility now being given some attention
(see Chap. II.1).
The intensification of agricultural management, particularly in Europe and
the Americas, but in other countries as well, has resulted in a great expansion
of the use of industrially fixed nitrogen for fertilizer. Approximately 50 million
metric tons of nitrogen are now fixed industrially and an additional approxima-
tely 20 million metric tons are fixed inadvertently by combustion processes
annually. This fixation plus that contributed by leguminous agricultural crops
probably is about equal to the nitrogen fixed annually by "natural" processes
before human intervention. Thus, at least the fixation portion of the nitrogen
cycle has been doubled. Denitrification probably has not kept pace with this
new nitrogen input, although other management practices such as the concentra-
tion of wastes in urban areas and the delivery of soil nitrogen to coastal waters
by erosion probably also have increased denitrification (BROADBENT and CARL-
TON 1978).
The nitrogen fixation reaction, whether it is carried on industrially or by
biological systems, requires considerable energy, particularly under aerobic con-
ditions for the biological system. Biological fixers accomplish this by the use
of photosynthetic energy (their net dry matter yield is probably reduced accord-
ingly) and so they have the advantage that they can be used at locations where
logistic, economic or other factors prohibit the use of industrially fixed nitrogen.
The energy limitation on the fixation of nitrogen apparently is imposed by
the kinetic barrier of the high activation energy of the dinitrogen molecule,
a barrier which neither the biological system nor the industrial process has
overcome. (The high temperature reaction of nitrogen with oxygen presumably
could circumvent this problem but because the resultant product would be
a nitrate weighing at least four times as much as anhydrous ammonia per
unit nitrogen, the energy cost of its transport and other problems in its use
would outweigh the energy advantages over ammonia synthesis.)

3.5 Human Influences

A number of human activities, including the drainage of wetlands, the use
of nitrogenous fertilizers, the fixation of nitrogen in combustion reactions such
as automobiles, the concentration of wastes in urban areas and others, have
greatly altered the nitrogen cycle. What the significance of these changes might
I. 7 Cycling of Elements in the Biosphere 225

be is difficult to evaluate. Each of these changes can be viewed as alarming

or insignificant depending upon the bias of the analyst, but in most cases avail-
able information is not sufficient to yield reliable objective analysis. Probably
in most cases the negative consequences outweigh the positive ones and therefore
these perturbations should be minimized. Some of the consequences of nitrogen
mismanagement, such as the eutrophication of streams and estuaries, are suffi-
ciently obvious as to result in corrective action, whereas others are more subtle
and some perhaps have not been recognized (DELWICHE 1981).
The use of nitrogenous fertilizers, particularly in heavy applications, appears
to result in an increased rate of denitrification. Under most field conditions
about 90% of the nitrogen lost in denitrification is released as N 2 and about
10% is released as N 20. (The figures range from 0% to 20% in most cases
studied.) With heavy applications of nitrogen fertilizers the ratio of N 2 0 to
nitrogen may be higher. This added increment ofN 2 0 to the atmospheric budget
can result in some reduction of concentration of ozone and a slight lowering
of the ozone "layer" in the stratosphere (CRUTZEN and EHALT 1977). The
ozone layer and the atmospheric N 20 content have been under study in this
context for a comparatively short time and so we are uncertain of trends at
present, but there probably has been some increase in the concentration of
atmospheric N 2 0. Where organic wastes are concentrated, such as in feedlots
and sewage disposal systems, there is also opportunity for increased denitrifica-
tion and increased N 20 production, as well as for the volatilization of ammonia
to the atmosphere.
Excessive use of nitrogen can also cause higher nitrate concentration in
foods and feeds with potentially deleterious effects. Nitrates have always been
a part of the human diet, but this does not mean that higher levels are permissi-
ble. Some nitrites are formed by the reduction of nitrate in the intestine and
these can lead to the formation of nitrosamines which are potentially carcino-
genic. Data to date dealing with the problem are limited. Similarly, domestic
water supplies have always contained some nitrate ion. In some locations the
nitrate ion concentration of groundwaters has increased significantly as a result
of fertilizer use or from sewage disposal systems or other sources. Legumes
also can contribute to groundwater nitrate. Acceptable levels of nitrate in
groundwaters are likewise difficult to establish, but at least for some individuals
high nitrate waters can be hazardous.
The intensive management of agricultural lands probably will continue and
increase so the problem appears to be one of gathering the information necessary
to minimize the negative consequences of this intensive management and then
implementing the policies which the information dictates.

4 The Sulfur Cycle

4.1 Comparison with the Nitrogen Cycle
An outline of the sulfur cycle is given in Fig. 4. Sulfur experiences the same
cycling from soil to plant and back again, as do other elements essential to
plants. Like nitrogen, because it has various oxidation levels (Table 3), sulfur
 226 c.c. DELWICHE:

ATMOSPHERE
0.1 ILar"I, a. alide. ~
/ or al,-aeidel

~_..... '1.2 ~ GaHo)~OIical

___

~
Soure.. and fallout Soure"I_lntl Precipitation. fallout
@ \
Go_. Biolotical 0
and ,a.tau. ablGrptlon

b
Sources (land I I

L.~_:. : ~= L:' :I : : : T:~I ~:I-:_C. .;: :. A~;: I~: :~=NT=-:~ ~

....

~
--,...... ESTURINE a
TIDAL ZONES
srI a
,

"
11
... II ., .
... IGNEOUS ROCK
o o

Fig. 4. The sulfur cycle. Reservoirs (pools) are shown in rectangles, transfer rates in
circles. Units are gigagram atoms (10 12 g at). (Sources: BOLIN and CHARLESON 1976,
GARRELS et al. 1975)

Table 3. Oxidation states of sulfur and some typi-

cal compounds

Valence Typical compound

+6 H 2 S0 4 Sulfuric acid
+4 H 2 S0 3 Sulfurous acid
+2 H 2 S0 2 Sulfoxylic acid
o S Sulfur
-2 H 2S Hydrogen sulfide

participates in the energy metabolism of a number of organisms. It also forms

volatile compounds such as oxides, [sulfur dioxide (S02), sulfur trioxide (S03)'
and others], hydrogen sulfide (H 2 S), and various methylated derivatives. The
atmosphere, therefore, is an important medium for the cycling of sulfur, but
unlike nitrogen, the primary reservoir of sulfur is not the atmosphere. Most
compounds of sulfur are metabolized by at least some organisms and so the
system is more tightly coupled than is that of nitrogen (BOLIN 1976, BORMANN
and LIKENS 1979). Although sulfur is not required in as large quantities by
most organisms as is nitrogen, it is still a major plant constituent and there
are many examples of sulfur deficiency. The most common form of inorganic
sulfur in the soil is sulfate ion (SO~-) which is highly mobile. It is therefore
somewhat surprising that since sulfate is readily transported to the oceans,
I. 7 Cycling of Elements in the Biosphere 227

sulfur deficiency is not more common than it is. The answer must be attributed
to the many processes which there are for the volatilization of sulfur and its
return through the atmosphere to the land. The oxidation-reduction levels of
sulfur range from its most oxidized form as sulfate ion in which sulfur has
a valence of + 6 to that of hydrogen sulfide (H 2S) in which sulfur has a valence
of - 2. This range like that of nitrogen also represents eight electrons' (Table 3).

4.2 Microbial Oxidation

As with nitrogen most of sulfur in plants is reduced, as SH groups at the

oxidation-reduction level of hydrogen sulfide. Upon decomposition of plant
and animal material, as sulfide ion is returned to the soil, it also can serve
as energy substrate for autotrophic organisms. Various organisms such as Thio-
bacillus thiooxidans can oxidize sulfide ion either to sulfur (S2) or all the way
to sulfate ion (SO~ -). Other organisms can carryon the full oxidation sequence,
or some of them some steps of it. The energy release is considerable as shown
by the following reactions:

H 2S + 0.5 O 2 --+ H 20 + S
A GO = - 52.1 kcal mol- 1
S+1.50 2 +H 20--+H zS04
at pH 7 AG' = -145.5 kcal mol- 1 .

As is the case with nitrogen, available sulfur in the soil is usually in its most
oxidized form as sulfate ion. The energy released in the overall oxidation of
sulfide to sulfate is even greater than that released in the oxidation of ammonium
ion to nitrate. Conversely, when sulfate ion is taken up by plants an equivalent
amount of energy and more must be expended to again reduce it to the level
of sulfide, its form in plant protein.

4.3 Sulfate Reduction

These energy arguments apply to the conditions under which most plants live
where molecular oxygen is available at a pressure of approximately 0.2 atmo-
spheres. If oxygen is absent, sulfate ion can serve as "electron acceptor" by
microorganisms for the oxidation of whatever soluble organic materials may
be available. This is a situation comparable with that which results in denitrifica-
tion of nitrate ion. Sulfate ion and various reduced compounds between it
and sulfide ion (SZ -) are formed, with the organic substrate being converted
to carbon dioxide and water. Energy yielded in the reaction goes to do the
work of the organisms involved, including the synthesis of new cell material.
The sulfides of a number of elements, notably those of iron, are quite insoluble
and common constituents of anoxic soils and muds. Deposits of sulfur usually
are the consequences of sulfide ion oxidation or are Juvenile in origin. Hydrogen
228 c.c. DELWICHE:

sulfide (H2S) commonly is volatilized in the reaction and escapes to the atmo-
sphere as gas. In the atmosphere it is auto oxidized to oxides of sulfur or sulfate
ion. This is one of the major avenues of return of sulfur from tidal flats or
estuarial waters to the atmosphere and eventually to the land.

4.4 Patterns of Sulfur Movement

Since most sulfates are cpmparatively soluble and sulfate ion therefore relatively
mobile in the soil, its general pattern of movement is into groundwaters and
eventually to the sea. Processes of sulfate reduction with the yield of gaseous
hydrogen sulfide result in the return of sulfur to the atmosphere and its redistri-
bution to the soil in precipitation. Although sulfur, like other elements, is rather
tightly cycled from the soil back to the plant, this process of sulfur return
is particularly important in areas of abundant rainfall where the leaching of
mobile elements can be significant. In more arid climates, where leaching takes
place less rapidly, sulfates may accumulate in the soil and under conditions
of extreme aridity, as in desert lands, sulfates may be present in large quantities
along with nitrates and chlorides. In the extreme, precipitates of gypsum
(CaS0 4 ) are formed.
The significance of the volatilization and transport of sulfur resulting from
sulfate reducing processes frequently is overlooked. The abundance of sulfur
and chlorine in the earth's crust is the same order of magnitude, yet the concen-
tration of sulfur in the ocean is only about 1/20 that of chlorine. Either the
sulfur is precipitated as sulfides or volatilized and returned to the land. On
the other hand, the requirement of land plants for sulfur is perhaps a thousand
times the requirement for chlorine and were it not for this mechanism of return
of sulfur to the land, sulfur deficiency would be much more prevalent than
it is.

4.5 Human Influences

The effects of human activity on the sulfur cycle have been at least as striking
as those on the nitrogen cycle. Sulfur has not been extensively mined and trans-
ported to the soil as a fertilizer, nor indeed have deficiencies of sulfur on agricul-
tural soils been recognized extensively. The intensification of agricultural mana-
gement with demands for exogenous supplies of mineral elements has occurred
more or less simultaneously with the industrial revolution. The fossil fuels which
powered the industrial revolution (coal, natural gas and petroleum) were formed
under reducing conditions and all of them contain significant quantities of sulfur
as sulfides. Their combustion releases to the atmosphere oxides of sulfur which
eventually are returned to the Earth in precipitation. This hUman influence
can perhaps be looked upon as a positive one, helping to meet what might
otherwise have been a more common sulfur deficiency. In many cases this proba-
bly is true but there are also discernible negative consequences. In the atmos-
phere the oxides of sulfur are eventually oxidized all the way to SO~- which
in combination with water yields sulfuric acid (GRANAT et al. 1976). Sulfuric
I. 7 Cycling of Elements in the Biosphere 229

acid is highly dissociated so that the resultant precipitation is acidic (LIKENS

et al. 1979). Downwind from industrial centers acid precipitation has influenced
weakly buffered soils and water bodies. Examples are known from the Appala-
chian Mountains and from Scan'dinavian lakes where the pH has been lowered
to 4 or less resulting in the killing of fish and other life. Weakly buffered
soils in these environments have been correspondingly affected. Not all of the
acidity is attributable to sulfate ion, some of it resulting from nitrate ion also
from combustion sources as well as other processes of nitrogen volatilization.
Recently alarming damage, in particular in coniferous forests, has been con-
firmed in different regions of Central Europe, popularly described as acid rain
causing "Waldsterben" (forest decimation). In this case also S02 and H 2S0 4
arising directly or indirectly from power plants working with coal and from
other technical sources are very probably the prevailing pollutants. Considerable
work has still to be invested to analyze the whole complex of injurious factors.
For more details the reader is referred to Chapter 14, Volume 12D, this Series.

5 The Phosphorus Cycle

5.1 Oxidation and Reduction

Phosphorus, belonging to the same subgroup as nitrogen in the periodic table

and possessing many of the general characteristics of nitrogen, nevertheless
has a strikingly different biological role and geochemical or geobiological beha-
vior (PmRRou 1976). Among the less abundant elements of the lithosphere (ab-
out 0.1 % of igneous rocks) it commonly occurs in the + 5 valence state except
under most strongly reducing conditions such as the metal and sulfide phases
of meteorites and possibly some strongly reducing biological systems. Phospho-
rus tends to be a geochemical companion of titanium, an element which is
chemically quite different. This coherence is largely the consequence of a coinci-
dental similarity of some of their respective minerals.
Phosphorus like nitrogen exists at a number of valence levels from the re-
duced ( - 3) level of the gas phosphine to the oxidized ( + 5) level of phosphate
ion (Table 4). Unlike nitrogen, however, phosphorus is normally in its most

Table 4. Oxidation states of phosphorus and some typi-

cal compounds

Valence Typical compound

+5 H 3 P0 4 Phosphoric acid
+3 H 3 P0 3 Phosphorous acid
+1 H 3 P0 2 Hypophosphorous acid
o P Phosphorus
-1 [PH] Unstable gas
-2 P2H4 Hydrogen diphosphide
-3 PH 3 Phosphine
230 C.C. DELWICHE:

ATMOSPHERE
< .001

Combustion

C""
b
LAND OCEAN
PLANTS 70 PLANTS 4.0
LAND OCEAN
ANIMALS .2 ANIMALS .25
.-::-::-:,--------,

"
IGNEOUS ROCK • 0

Fig. 5. The phosphorus cycle. Reservoirs (pools) are shown in rectangles, transfer rates
in circles. Units are in gigagram atoms (10 12 g at). (Sources: GARRELS et al. 1975, STUMM
1972, LIKENS and BORMAN 1972, DELWICHE and LIKENS 1977

oxidized state (PO~ -) and is comparatively nonmobile in the soil. Whereas

almost all salts of nitrates and many of the sulfates are quite soluble and hence
mobile in the soil-water system, most phosphates are sparingly soluble, with
some, such as apatite, being particularly resistant to weathering. This chemical
stability of the phosphate ion and the relative immobility of phosphate have
resulted in the development of some interesting biological characteristics geared
to survival in an environment where a critical life support material was fre-
quently in short supply. The phosphorus cycle is shown in Fig. 5.

5.2 Movement and Transport in the Biosphere

Because of its triple negative charge as well as the low solubility of most of
its salts, phosphorus is tenaciously held in the soil, and its transport through
the soil and into waters is slow compared with most of the elements essential
to plants. For this reason and because normally it does not fonn volatile com-
pounds, it is closely cycled in the biosphere and its concentration in seawater
is exceedingly low (about 0.1 ppm or 3 x 10- 6 moll- 1). The low seawater con-
centration is commonly considered to be the principal limiting factor on seawa-
ter productivity although other elements, particularly iron, may also be limiting.
In soil, although its concentration is always comparatively low in the soil
solution, phosphorus is most available between pH's 6 and 7. Under more alka-
line conditions it tends to be bound as insoluble phosphates or held on the
I. 7 Cycling of Elements in the Biosphere 231

exchange complex tightly; and at lower pH values, as the concentrations of

iron and aluminum become greater, these elements tend to bind the H 2P04"
or HPO~- ions as insoluble complexes. Consequently, in both more acid and
more alkaline soils, there tends to be some selection for those plant species
which are more able to extract phosphorus from this system of limited availabil-
ity. Frequently, this appears to include symbiotic associations of microorga-
nisms. Mycorrhizae, as their name implies, are fungi closely associated with
the roots of some plant species and appear to provide such a function. Such
associations are most common under conditions oflow nutrient supply, particu-
larly where phosphorus, zinc, or other elements are limiting. The fungus appears
to assist in the solution and transport of this limiting ion supply.
Plants require about twice as much phosphorus as sulfur on a molar basis,
so that it is fortunate that phosphorus is not transported to the sea more rapidly
than it is. There is no mechanism whereby it would be returned to the land
from the sea, and the supply in soil must be maintained by the weathering
of rocks. Igneous rocks contain about 0.3% phosphorus on a weight basis
so that about 330 g of rock would be required to yield 1 g ofP. Of that phospho-
rus which does reach the ocean, most is concentrated in deposits on the continen-
tal shelf or continent margins, near zones of upwelling in subtropical climates.
The peninsula of Florida and neighboring regions contain large quantities of
high grade phosphatic ores. This rock phosphate consists of fluorapatite,
Cas (P0 4 h F and the corresponding chlorine derivative, mixed with carbonate
precipitates.
Although under most circumstances inorganic phosphorus is in its oxidized
form, under highly reducing conditions such as in marshes and anoxic muds,
phosphorus can be reduced biologically to the level of phosphine gas (PH3)
analogous with ammonia. There is evidence that "marsh gas" (mostly methane)
contains some phosphine gas or its methylated analog, the autooxidation of
which when it reaches the atmosphere being responsible for the glow frequently
associated with marsh gas, at least in legend (the will-o-the-wisp). It is unlikely
that this phenomenon is a significant mechanism for the transport of phosphorus
in the biosphere.

5.3 Human Influences

A small quantity of phosphorus is released to the atmosphere by the burning

of fossil fuels (about 5 x 1011 g annually) and more by the combustion of plant
material (about 2.5 x 10 12 g annually) (DELWICHE and LIKENS 1977). The
quantity from fossil fuel combustion is less than 1% on a molar basis of the
nitrogen and sulfur emissions so that its significance in acid precipit~tion is
negligible.
The annual use of approximately 4 x 10 13 grams of phosphorus as fertilizer
does not in itself constitute a major perturbation of the phosphorus cycle, largely
because of the comparative insolubility and immobility of phosphorus in the
soil. The misuse of such fertilizers can result in the loss of phosphates to streams
with resultant eutrophication, but the more likely source of phosphorus contami-
232 c.c. DELWICHE:

nation is erosional clay particles whether or not phosphorus fertilizers have

been used. On the other hand, because phosphorus is less mobile, streams and
other water bodies are usually low in phosphorus so that even a slight addition
of phosphorus to the system can result in a considerable change from natural
conditions.
One possible consequence of the use of phosphatic fertilizers may deserve
some attention. The rock phosphate deposits from which they are derived are
all of secondary origin and because of the close similarity between phosphorus
and uranium, there is a tendency for uranium compounds to coprecipitate with
phosphates in these rocks. Depending on the processes used for the manufacture
of fertilizers, it is possible to transport uranium to agricultural soils by this
means. Some fields have received an increment of uranium by the use of rock
phosphate fertilizers, but present methods for the preparation of more highly
processed" superphosphate" fertilizers tend to avoid this difficulty.
For information on the phosphorus load in the waste waters of industry
and households see Chapter 14, Volume 12D, this series.

6 Other Elements

6.1 Biological Cycling

Most of the other mineral elements required by plants also have complex and
in many cases poorly understood cycles. These have received less attention
than have the cycles of nitrogen, phosphorus, sulfur and carbon, largely because
examples of their deficiency are less prevalent or because perturbations of their
cycles by human activity have been less obvious. Each of these elements has
its own unique chemical and biological characteristics, and each of their cycles
has its own unique behavior (BOWEN 1966, AHRENS 1979).

6.2 The Special Significance of Iron and Aluminum

Although, as indicated above, aluminum is present in comparatively high con-

centrations in igneous rocks, and therefore an abundant product of weathering,
it is not an essential element for any organism as far as is now known. In
retrospect, it is possible to rationalize this inessentiality of aluminum. It is com-
monly found in its + 3 valence state and in aqueous media cannot be further
reduced. Consequently, it would not be expected to participate in any biological
oxidation-reduction reaction involving energy metabolism. Neither would it be
expected to be a constituent of any electron transport coenzyme and so its
only remaining role would be that of some structural unit. This is a role which
the element appears not to have played, even though it is reasonably ubiquitous
in its distribution.
I. 7 Cycling of Elements in the Biosphere 233

Indirectly, however, aluminum has had an influence on the biological world.

Both in the aluminosilicates of layered clay minerals and as derivatives of its
oxides, aluminum is a common soil constituent. The layered clay minerals have
a comparatively high cation exchange capacity. The sesquioxide clays which
will be discussed below are made up of oxides of aluminum and iron and
do not have a strong cation adsorption capability.
Iron can exist in either the ferrous (Fe 2 +) level, or ferric (Fe3+) level within
the normal range of biological oxidation-reduction potentials. The oxidized
ferric ion forms highly insoluble ferric hydrate precipitates, particularly above
pH 7, and so its availability in more alkaline soils frequently may be limiting.
Some soils (for example oxisols) may be made up largely of colloidal ferric
hydrate particles and these present special problems of management for agricul-
tural purposes. These sesquioxide clays which may also contain large quantities
of the oxides or hydrated oxides of aluminum are widespread in their distribu-
tion and particularly common in tropical and subtropical climates. Their
comparatively low cation exchange capacity and their common association with
the soils of higher rainfall regions in the tropics is of some interest (MALAVOLTA
et al. 1974). Ecosystems existing on these sesquioxide clays usually carry most
of the mineral element pool in the vegetation itself or in the small amount
of litter remaining on the forest floor. The soils are not "fertile" in the sense
of their function as mineral element reservoirs, but they nevertheless support
abundant vegetation in many locations. The conversion of native ecosystems
on these soils to agricultural production which is seasonal in nature can have
disappointing results because of the absence of exchange capacity and an atten-
dant mineral element pool.
The low solubility of ferric hydroxide at pH 8 provides a special mechanism
for the transfer of oxidizing capability to depths in an anoxic medium. Oxidized
ferric clays may be colloidally dispersed in river waters flowing to the sea.
In higher concentrations of salt as in seawater the colloid is flocculated and
the particles resulting can sink. As they settle to anoxic zones the ferric oxide
can serve as electron acceptor for biological oxidations and be reduced to ferrous
ion which then will combine with sulfide ion and be precipitated as pyrite
or other reduced sulfides. This provides a more rapid mechanism for transferring
electron acceptors to depth than does the slower diffusion of oxygen. The large
amount of reduced iron which must have been present in the ocean in early
pre-Cambrian times had to be removed either by precipitation or oxidation
before any significant accumulation of oxygen could have taken place in the
atmosphere. Consequently, this process may have had an important role in
these early phases of atmosphere evolution.

6.3 Hydrogen Ion

Many climatic and vegetation patterns influence the pH of a soil. Cool climates
with adequate rainfall where the decomposition of soil organic matter is slow
or incomplete, result in a somewhat acid leaching of the upper layers of the
soil leaving behind the less soluble silica and silicates. These soils frequently
234 C.C. DELWICHE:

have a low cation exchange capacity, and like the highly leached sesquioxide
clays of tropical and subtropical climates, maintain much of the mineral budget
in the vegetation and litter. Conversion of these soils to seasonal cultivation
may have the same consequences of low productivity. As a result, "swidden"
agriculture involving the burning of vegetation to prepare an agricultural field
which is then abandoned after 2 or 3 years, has been practiced in widely differing
areas such as the jungles of New Guinea and the Boreal Forests of Scandinavia.
In more arid climates, alkaline soil pH's are common, and these can result
in problems of a different sort. Both the accumulation of salts and toxicities
or deficiencies resulting from the high pH are common. Over the comparatively
narrow range of pH from 4.5 to 8.0, the relative availability of various ions
can vary greatly. In general, neutral to slightly acidic pH's are favorable for
most plants, although some are more tolerant to alkaline pH's and others to
more acid conditions.
If aluminum is present in the soil, as it frequently is, it can cause toxic
effects at pH's below 5.5. Aluminum toxicity is rather nonspecific, much like
other "heavy metal" toxicities encountered when concentrations of a number
of metals such as copper, zinc, mercury, and others are high. Aluminum is
also more soluble at very high pH's (in the neighborhood of 10) but many
other problems would be encountered in attempting to grow plants under such
alkaline conditions.

6.4 Characteristics of Sediments

The figures given in Table 1 for the elemental composition of various rocks
and other materials must be interpreted with caution. They represent mean
values useful for geochemical calculations, but values for a particular rock might
vary widely from these. Processes of metamorphism can greatly change the
composition of a parent material. Such rocks as serpentine which have been
highly altered by water and heat frequently show concentrations of some metals
and depletions of others. Soils derived from serpentine rocks commonly are
relatively high in magnesium (when compared with calcium) and frequently
are deficient in molybdenum. Other elements, some of them toxic to plants,
can be concentrated by metamorphic processes (MACKENZIE et al. 1979). Exam-
ples of these are selenium, mercury, boron, and others. The figures given here
are useful when interpreted as part of the soil-forming process. The weathering
of a limestone, for example, yields largely calcium ion, which is mobile, and
carbon dioxide, which is lost to the atmosphere. The resulting soil is made
up of the "impurities" in the limestone, secondary clay minerals formed at
the site, and the organic residues of plants and animals. As a result, soils formed
from limestone frequently are shallow even though they resulted from the weath-
ering of many meters of rock.
These are merely examples of the complexities which element cycles might
exhibit. Space does not permit dealing with them in detail here but the cycles
of these "other" elements, although they have received little attention, cannot
be looked upon as simple. BOWEN (1966) provides detailed treatment of some
of these.
1.7 Cycling of Elements in the Biosphere 235

6.S Passive Cycling

All of the elements which play an essential role in the structure or metabolism
of plants are actively cycled in the sense that their biological role is responsible
for their transfer from one compartment to another. But the plant is not a
completely selective organism and many ions which play no known role in
either its structure or metabolism are nevertheless cycled passively by being
taken up by the plant and returned by whatever route to the soil. Sometimes
the cycles of these elements are of importance because of their toxicity to plants
and the consumers of plants or because of their interaction with other elements
which are essential to plants. Even essential elements frequently are passively
cycled in the sense that they may be taken up in luxury or even toxic quantities
because of the inability of the plant to exclude them. Other passively cycled
elements are some of the heavy metals such as mercury, cadmium and silver.
Sometimes the passive cycling of these elements serves as an indicator of their
presence in the soil. Known accumulators of gold, uranium and other metals
are used as "indicator" plants, the analysis of their tissues providing informa-
tion regarding concentration in the soil.
In some cases the similarity of the chemistry of an element to that of an
essential element lends it a particular toxicity. Selenium is an example of this.
It frequently accompanies sulfur and is substituted for sulfur in some of the
amino acid and other sulfur compounds of plants and animals resulting in
toxic effects. Some plants are more tolerant of selenium than others and can
be accumulator plants for this element.
The process of passive ion cycling (as well as the cycling of required elements)
commonly results in the concentration of these materials near the soil surface.
The heavy metals in particular are comparatively immobile in soil and older
soils formed on sedimentary materials frequently show high surface concentra-
tions of them.

6.6 Possibilities of Deficiency

Soil formation is a complex process involving factors of parent material and

climate particularly and all soils change in characteristics with time. The more
mobile elements - nitrogen, sulfur, potassium, calcium, and magnesium - tend
to move downward if the water supplied in rainfall exceeds that lost in runoff
or evaporative loss. Other elements such as iron, which are less soluble, can
still be moved downward through the root zone as chelates of organic com-
pounds. As these elements are moved downward organic material frequently
is slowly oxidized and the elements again abandoned as insoluble in0rganic
compounds some distance down in the soil. The consequence is formation of
a hardpan layer of iron oxide or iron oxide-cemented secondary clay minerals.
This layer can be quite impervious to water, changing the drainage characteris-
tics of the system, lessening leaching and causing greater evaporative loss. The
hardpan can prevent the penetration of roots depending upon the depth at
which it is located, resulting in less plant growth and a further deterioration
236 C.C. DELWICHE:

of the quality of the site. The most important element in this process is time,
placing certain limitations on the productive age of a soil.
Some generalizations can be made regarding the likelihood of essential ele-
ment deficiencies. Old soils are more likely to show mineral deficiencies than
are young soils. Highly leached soils, either because of a light texture or heavy
precipitation, are more likely to show mineral deficiencies than others. Soils
with a high sesquioxide clay content or other soils with a low cation exchange
capacity are more likely to show deficiencies than others. Soils in climates of
abundant rainfall are more likely to show deficiencies in soluble monovalent
ions such as potassium than are those in more arid environments. Soils evolved
on parent material that has been subject to high-temperature solution and meta-
morphosis such as those formed on serpentine rock, are more likely to exhibit
either deficiency or toxicity problems because of the tendency of the serpentini-
zation process to concentrate some elements and remove others. Unfortunately,
these generalizations do not always hold true and the nutrient status of the
soil can be determined only by analysis and study.

7 "Open" Versus "Closed" Agricultural Systems

One of the major effects on the cycles of elements required by plants has been
that resulting from the shift in patterns of food production and utilization
(DELWICHE 1981). Urban centers have always been a part of civilization but
the combination of population growth and the industrialization of agriculture
have made this population concentration particularly significant during the past
half-century. Most population centers are bordering the seas or on large rivers.
Products of agriculture which are exported to urban centers eventually appear
as organic wastes, either from domestic consumption or from industrial pro-
cesses. These organic wastes are processed with varying degrees of effectiveness
or sometimes not processed. In either case degradation of the organic wastes
results in the liberation of the inorganic elements which they contain and these
eventually are transferred to groundwaters or to rivers, estuaries, and the ocean.
In addition to the eutrophication resulting from increased levels of inorganic
elements in these waters, the net result of the process is a continuous transfer
of elements required by plants from the soil on which the plants were grown
to waters and eventually the ocean. Soil-forming processes are slow compared
with the rate of ion utilization by plants and the concentration of these ions
in soils is steadily diminished. The loss of fertility is particularly evident for
the elements required in largest amounts such as nitrogen, phosphorus, sulfur,
and potassium, but all nutrients are affected. To make up for this loss, fertilizers
are used requiring energy for the fixation of nitrogen, the mining and transport
of minerals and the delivery of the fertilizer to the soil.
The waste organic materials, in addition to the nutrient elements they
contain, are potential sources of energy if they can be gathered economically
and provided the technology can be developed for the efficient utilization of
r.7 Cycling of Elements in the Biosphere 237

this energy. Ideally, energy should be extracted for fuel and the mineral elements
recovered so that they can be returned to the agricultural system. Transport
of the organic materials back to the soil directly is impractical in most cases
because of the energy requirement as well as the labor cost.
This opening of the production-utilization chain is unlikely to be reversed
by any process of de-urbanization and so a technological process which is energy
efficient appears to be the only solution.
Even the dispersion of populations to smaller communities would only
appear to be a solution to the problem by diluting large points of concentration
to a larger number of smaller points of concentration (villages) and these would
not assure the return of the essential elements to the soil.

References
Ahrens LH (1979) Origin and distribution of the elements. Pergamon, New York London
Aleem MIH (1970) Oxidation of inorganic nitrogen compounds. Annu Rev Plant Physiol
21 :67-90
Aleem MIH (1977) Energy coupling in chemolithotrophic bacteria. Symp Soc Gen Micro-
bioI 27: 351-381 -
Alexander M (1965) Nitrification. In: Bartholomew WV, Clark FE (eds) Soil nitrogen.
Am Soc Agron, Madison
Alexander MK, Marshall KC, Hirsch P (1960) Autotrophy and heterotrophy in nitrifica-
tion. Trans 7th Int Congr Soil Sci, Madison 2:586-591
Bartholomew WV, Clark FE (eds) (1965) Soil nitrogen. Am Soc Agron, Madison
Berner RA (1979) Kinetics of nutrient regeneration in anoxic marine sediments. In:
Ahrens LH (ed) Origin and distribution of the elements. Pergamon, New York
London
Bolin B (1976) Transfer process and time scales in biogeochemical cycles. In: Svensson
BH, Soderland R (eds) Nitrogen, phosphorus and sulfur - global cycles. SCOPE
Rep 7. Ecol. Bull (Stockholm) 22: 17-22
Bolin B, Charlson RJ (1976) On the role of the tropospheric sulfur cycle in the shortwave
radiation climate of the earth. Ambio 5:47-54
Bond G (1976) The results of the IBP survey of root-nodule formation in non-leguminous
angiosperms. In: Nutman PS (ed) Symbiotic nitrogen fixation in plants, vol. VII.
Int BioI Progr 443-474, Cambridge Univ Press
Bormann FH, Likens GE (1979) Pattern and process in a forested ecosystem. Springer,
Berlin Heidelberg New York
Bowen HMJ (1966) Trace elements in biochemistry. Academic Press, London New York
Bremner JM, Blackmer AM (1978) Nitrous oxide: Emission from soils during nitrification
of fertilizer nitrogen. Science 199: 295-296
Brill WJ (1977) Biological nitrogen fixation. Sci Am 236(3):68-81
Broadbent FE, Carlton AB (1978) Field trials with isotopically labeled nitrogen fertilizer.
In: Nielsen DR, MacDonald JG (eds) Nitrogen in the environment, vol I. Academic
Press, London New York .
Broda E (1975a) The history of inorganic nitrogen in the biosphere. J Mol Evol7: 87-100
Broda E (1975b) The evolution of the bioenergetic process. Pergamon, Oxford
Crutzen PJ, Ehalt DH (1977) Effects of nitrogen fertilizers and combustion on the stra-
tospheric ozone layer. Ambio 6(2-3): 112-117
Deevey ES Jr (1970) Mineral cycles. Sci Am. 223: 149-158
Delwiche CC (1970) The nitrogen cycle. Sci Am 223(3): 137-146
Delwiche CC (1981) Nitrogen fertilizer. In: Bower FA, Ward RB (eds) Stratospheric
ozone and man. CRC Press, Boca Raton .
238 C.C. DELWICHE: 1.7 Cycling of Elements in the Biosphere

Delwiche CC, Bryan BA (1976) Denitrification. Annu Rev Microbiol 30:241-262

Delwiche CC, Likens GE (1977) Biological response to fossil fuel combustion products.
In: Stumm W (ed) Global chemical cycles and their alterations by man. Dahlem
Konferenzen, Berlin
Garrels RM, Mackenzie FT (1971) Evolution of sedimentary rocks. Norton, New York
Garrels RM, Mackenzie FT, Hunt C (1975) Chemical cycles and the global environment.
Kaufman, Los Altos, Cal
Gosz JR, Likens GE, Bormann FH (1973) Nutrient release from decomposing leaf and
branch litter in the Hubbard Brook Forest, New Hampshire. Ecol Monogr
43(2): 173--191
Granat L, Rodhe H, Hallberg RO (1976) The global sulfur cycle. In: Svensson BH,
SOderlund R (eds) Nitrogen, phosphorus and sulfur - global cycles. SCOPE Rep
7. Ecol Bull (Stockholm) 22:89-134
Hahn J (1974) The North Atlantic Ocean as a source of atmospheric N 2 0. Tellus
26:160-168
Hardy RWF, Havelka UD (1975) Nitrogen fixation research: A key to world food?
Science 188: 633--643
Hom MK, Adams JAS (1966) Computer-derived geochemical balances and element abun-
dances. Geochim Cosmochim Acta 30: 279-297
Junge CE (1972) The cycle of atmospheric gases-natural and man made. 1972. QJR
Meteorol Soc 98:711-729 .
Junge CE (1974) Residence time and variability of tropospheric trace gases. Tellus
26:477-488
Likens GE, Bormann FH (1972) Nutrient cycling in ecosystems. In: Wiens J (ed) Ecosys-
tem structure and function. Oregon State Univ Press, Corvallis
Likens GE, Wright RF, Galloway IN, Butler TJ (1979) Acid rain. Sci Am 241(4):43--51
Mackenzie FT, Lantzy RJ, Paterson V (1979) Global trace metal cycles and predictions.
Math Geol 11(2):99-142
Malavolta E, Haag HP, Mells FA, Brasil MOC (1974) Nutricao Minerale Adubacao
De Plantas Cultivadas. Quazzelli, Sao Paulo
Penman HL (1973) The water cycle. In: Hamilton CL (ed) Chemistry in the environment.
Freeman, San Francisco
Pierrou V (1976) The global phosphorus cycle. In: Svennson BH, Soderlund R (eds)
Nitrogen phosphorus and sulfur- global cycles. SCOPE Rep 7. Ecol Bull (Stockholm)
22:89-134
Poldervaart A (1955) Chemistry of the earth's crust. Geol Soc Am Spec Pap 62: 119
Robinson E, Robbins RC (1970) Gaseous nitrogen compound pollutants from urban
and natural sources. J Air Pollut Contrib Assoc 20: 303-306
Soderlund R, Svensson BH (1976) The global nitrogen cycle. In: Svensson BH, Soderlund
R (eds) Nitrogen, phosphorus and sulphur - global cycles. SCOPE Rep 7. Ecol Bull
(Stockholm) 22: 23--72
Stumm W (1972) The acceleration of the hydrogeochemical cycling of phosphorus. In:
Dyrssen D, Jagner D (eds) The changing chemistry of oceans. Wiley and Sons, New
York
Taylor SR (1964) Abundance of chemical elements in the continental crust: A new table.
Geochim Cosmochim Acta 28:1273-1285
Torrey JG (1976) Initiation and development of root nodules of Casuarina (Casuarina-
ceae). Am J Bot 63:335-344
Torrey JG (1978) Nitrogen fixation by actinomycete-nodulated angiosperms. BioSci
28:586-592
Wazer van F (ed) (1961) Phosphorus and its compounds, vol II. Wiley-Interscience,
New York
II. Inorganic Nitrogen Nutrition
11.1 Physiology, Biochemistry and Genetics
of Dinitrogen Fixation
H. BOTHE, M.G. YATES, and F.C. CANNON

1 The Nitrogen-Fixing Organisms and the

Nitrogenase Reactions

1.1 Introduction

(Di)nitrogen fIxation is the terminus technicus for the conversion of the dinitro-
gen molecule to ammonia according to the following equation: 6 H+ +6 e- +
N 2 = 2 NH 3 . This reduction can proceed either chemically or biologically in
a reaction catalyzed by the enzyme complex nitrogenase. The direct comparison
of the data for chemical and biological nitrogen fIxation reveals the importance
of the latter reaction. The chemical reaction, known as the Haber-Bosch process,
is used to supply nitrogen fertilizers for agriculture. Estimates of the global
consumption of fertilizers in 1974 were 30 x 106 (BURNS and HARDY 1975) or
40 x 106 t year- 1 (Rothamsted Subject Day 1975), and this latter compendium
stated that the world-wide consumption had increased tenfold since 1950. The
Haber-Bosch process is highly energy-consuming since it proceeds at a pressure
of some 200 atm and temperatures between 400°-600°C where molecular hydro-
gen and nitrogen are converted to ammonia in the presence of a metal catalyst
and with a yield of approximately 13%. It is said (SPRENT 1979) that about
1.5 kg of fuel oil is needed to deliver 1 kg of fertilizer N to the farm. In compari-
son, the rate of biological nitrogen fIxation (in 106 t yr- 1 ) was calculated to
be 175 (BURNS and HARDY 1975), 170-270 (SODERLUND and SVENSSON 1976)
and 150 (U.S. Council of Agricultural Science and Technology, Cast, Anon,
1976). These values are probably underestimates, since the contributions of
the nitrogen-fIxing organisms to the nitrogen balance in the sea are virtually
unknown. Though the biological reaction requires the expenditure of cell energy
as discussed in Section 2, it proceeds at 1 atm of pressure and at ambient
temperatures. Nitrogen fIxation by living cells is thus a fIne example of the
potential of an enzyme-catalyzed reaction. It is often said that biological nitro-
gen fIxation, next to photosynthesis, is the most important reaction in nature.
This may be the reason that nitrogen fIxation has become a popular scientifIc
subject in the past few years, resulting in a vast number of publications annually
including a diversity of books (see e.g. POSTGATE 1971,1978; HARDY 1977-1979,
QUISPEL 1974, AYANABA and DART 1977, BURNS and HARDY 1975, SPRENT
1979) and reviews (WINTER and BURRIS 1976; ZUMFT and MORTENSON 1975,
MORTENSON and THORNELEY 1979, ZUMFT 1976 etc.). Nitrogen fIxation and
its relationship to photosynthesis has been described by STEWART and QUEBE-
DEAUX in Chaps. 6 and 7, respectively, Vol. 6, this Series.
242 H. BOTHE et al.:

Relatively few groups of microorganisms contribute to the overall global

nitrogen fixation rate (Table 1). Since WILSON'S (1958) impressive review on
nitrogen fixation in Volume 8 of the first Encyclopedia of Plant Physiology,
the names of several organisms in the list have changed. Some microorganisms,
previously considered to fix nitrogen, proved to be excellent scavengers of com-
bined nitrogen (e.g. from dust) and thrive on this nitrogen source. It was not
until the introduction of more sensitive assay methods eSN-analysis, C2 H 2 -
reduction) that such organisms could be distinguished from true nitrogen-fixers.
Even now the list is probably incomplete. New nitrogen-fixing organisms have
been found rt;!cently and will presumably be detected in future. Two points
immediately emerge from looking at Table 1: Firstly, there is no example of
a nitrogen-fixing eukaryotic organism. All such candidates (yeasts) did not with-
stand critical examination by the more sensitive assay methods (MILLBANK
1969). Secondly, the capability for nitrogen fixation is relatively scarce but
spread widely among the different groups of prokaryotes which have little else
in common. Since there are no generally accepted systematic classifications
for bacteria at present, nitrogen-fixing microorganisms are often listed accord-
ing to the physiological characteristics of the nitrogen-fixing process (as in
Table 1).

Table 1. List of selected genera of nitrogen-fixing organisms

Genus or type Species

A. Free-living nitrogen-fixers
1. Obligate aerobes
Azotobacter vinelandii, chroococcum, paspali,
beijerinckii
Beijerinckia indica
Derxia gummosa
Azotococcus agi/is

2. Obligate aerobes that fix nitrogen only at low oxygen tensions

Azospirillum brasilense, lipoferum
Xanthobacter autotrophicus, flavus
Thiobacillus ferro-oxidans
Rhizobium cowpea group
Methylosinus sporium
Methylococcus capsulatus

3. Facultative anaerobic bacteria that fix nitrogen only under anaerobi<; conditions
Klebsiella pneumoniae
Bacillus polymyxa, macerans
Propionibacterium shermanii, petersonii
Escherichia intermedia
Citrobacter freundii
Enterobacter cloacae, agglomerans
Erwinia herbicola
11.1 Physiology, Biochemistry and Genetics of Dinitrogen Fixation 243

Table 1 (continued)

Genus or type Species

4. Obligate anaerobes
Clostridium pasteurianum, butyricum
Desulfovibrio gigas, desulfuricans, vulgaris
Desulfotomaculum ruminis

5. Phototrophic bacteria
a) Rhodospirillaceae
Rhodospirillum rubrum, tenue,fulvum, molischianum,
photometricum
Rhodopseudomonas palustris, viridis, capsulata, spheroides

b) Chromatiaceae
Chromatium vinosum, gracile, minus, violacea
Thiocystis violacea
Thiocapsa roseopersicina, pfennigii
Amoebobacter roseus
Ectothiorhodospira spaposhnikovii

c) Chlorobiaceae
Chlorobium limicola, vibrioforme
Pelodiction luteulum

6. Blue-green algae or cyanobacteria

a) Unicellular, aerobically fixing strains
Gloeothece alpicola (5 strains)
(formerly Gloeocapsa)
Aphanothece

b) Filamentous, heterocystous forms which fix aerobically and anaerobi-

cally
Nostoc muscorum, commune
Anabaena cylindrica, variabilis
Aphanizomenon fios- aquae
Cylindrospermum Various strains
Calothrix Various strains

c) Filamentous, non-heterocystous forms which fix micro aerobically

Plectonema boryanum
Oscillatoria Various strains
Pseudanabaena Various strains
Lyngbya Various strains
Phormidium Various strains

B. Symbiotic nitrogen-fIXing systems and associations

Host family Host genus N 2-fixing microorganism

a) Rhizobium symbioses
Leguminosae Most .species Rhizobium species
Ulmaceae Parasponia Rhizobium
244 H. BOTHE et al.:

Table 1 (continued)

Host family Host genus N 2-fixing microorganism

b) Non-Rhizobium symbioses
Betulaceae Alnus
Myricaceae Myrica, Comptonia
Eleagnaceae Eleagnus, Hippophae
Shepherdia In all cases
Rhamnaceae Ceanothus, Trevoa, Actinomycetes
Discaria " Frankia "
Rosaceae -Dryas, Cercocarpus,
Purshia
Coriariaceae Coriaria, Colleta
Casuarinaceae Casuarina

c) Lichens Collema, Peltigera Nostoc

Dendriscocaulon Scytonema

d) Liverworts Anthoceros, Blasia Nostoc sphaericum

Cavicularia

e) Waterfern Azolla Anabaena azollae

t) Cycads Cycas, Ceratozamia, Nostoc, Anabaena

Encephalartos, Macro-
zamia, Dioon etc.

g) Higher plants
Haloragaceae Gunnera Nostoc punctiforme

h) Associative symbioses and casual associations

Phyllosphere Leaves Azotobacter spp.
Rhizosphere Roots of grasses
Paspalum notatum Azotobacter paspa/i
Zea mays Azospirillum

Compiled mainly from data by BURNS and HARDY 1975, SPRENT 1979, POSTGATE 1978,
SIEFERT 1976, RIPPKA et al. 1979

1.2 Nitrogen Fixation by Free-Living Organisms

Among the aerobic free-living forms, Azotobacter is the most investigated

example. Azotobacter is present in almost every soil though never abundant.
It rarely meets optimal conditions for growth and nitrogen fixation which
include a sufficient supply of carbohydrates and oxygen. Azotobacter is said
to have the highest respiratory rate among bacteria. On the other hand, it
must protect its oxygen-sensitive nitrogenase from damage by the gas (see
Sect. 4).
The facultative anaerobic bacteria can grow in air when provided with com-
bined nitrogen. Nitrogen fixation, however, requires oxygen exclusion (e.g. Kleb-
siella pneumoniae, Bacillus polymyxa). In K. pneumoniae, nitrogenase has been
11.1 Physiology, Biochemistry and Genetics of Dinitrogen Fixation 245

studied extensively, both biochemically and genetically. Many of the techniques

developed for Escherichia coli can be directly transferred to study the genetics
of the nitrogen-fixing system in Klebsiella (see Sect. 7).
Among the strict anaerobic bacteria, Clostridium pasteurianum is the classical
organism in studies on nitrogen fixation. It was used in investigations by the
Du Pont group in the early 1960's which resulted in the description of the
first nitrogenase preparation with reproducibly high activities (CARNAHAN et al.
1960). Moreover, it was in the course of these investigations that the electron
carriers ferredoxin (MORTENSON et al. 1962) and flavodoxin (KNIGHT et al. 1966)
were discovered rather than in studies on photosynthesis or nitrite reduction.
The capability for nitrogen fixation is widespread among photosynthetic
bacteria. An extensive survey by SIEFERT (1976) showed 24 nitrogen-fixing strains
out of 26 Rhodospirillaceae, 16 out of 19 Chromatiaceae and 13 out of 17
Chlorobiaceae.
Many of the blue-green algae, or cyanobacteria (according to bacteriologists;
STANIER and COHEN-BAZIRE 1977) fix nitrogen. Under aerobic growth condi-
tions, nitrogenase in the filamentous forms is restricted to heterocysts which
are specialized cells. Whereas all heterocystous forms so far investigated are
capable of fixing nitrogen, the reverse is not true. STEWART and LEX (1970)
showed that the filamentous, non-heterocystous Plectonema boryanum fixed ni-
trogen when grown anaerobically or, more correctly, microaerobically, since
blue-green algae evolve oxygen by photosynthesis. Several non-heterocystous
strains of Oscillatoria, Pseudanabaena and the LPP (= Lyngbya, Phormidium,
Plectonema) group (RIPPKA and WATERBURY 1977, RIPPKA et al. 1979) are
now known to fix nitrogen microaerobically. The remarkable exceptions among
the unicellular blue-green algae are Gloeocapsa (WYATT and SILVEY 1969, re-
named Gloeothece, RIPPKA et al. 1979) and Aphanothece (SINGH 1973) which
fix nitrogen under air. There is no other example of an unicellular nitrogen-fixing
blue-green alga. The blue-green algal-like inclusions (termed cyanelles) which
are found in the eukaryotic Cyanophora paradoxa, Glaucocystis and Glauco-
sphaera do not fix N2 (BOTHE and FLOENER 1978).
Non-photosynthetic microaerophilic bacteria include Xanthobacter jZavus
(formerly Mycobacterium jZavum), Xanthobacter autotrophus (formerly Coryne-
bacterium autotrophicum) and the Azospirillum species, see also Chapter 11.3,
this Volume. In addition, some strains of Rhizobium which were reported to
fix nitrogen in the free-living state need microaerobic conditions.

1.3 Symbiotic Nitrogen Fixation

The overwhelming majority of the 12,000 species of the Leguminosfle form

nodules containing Rhizobium which fix nitrogen. Nodules infected by Rhizo-
bium have also been found in the non-legume Parasponia (TRINICK 1973, AK-
KERMANS et al. 1978). Nodules containing nitrogen-fixing actinomycetes, called
Frankia, occur at the roots of several plants which bear no systematic relation~
ship .. Blue-green algae are symbiotic partners oJ a diversity of eukaryotic hosts
(see Table 1 and the contribution by QUISPEL Chap. n.2, this Vol.). The associa-
246 H. BOTHE et al.:

Table 2. Estimates of the contributions of nitrogen-fixing microorganisms in the field

Plant Associative organisms Fixation rate

1. Rhizobium symbioses
Soybean: Glycine max R. japonicum 40-200
Peanut: Arachis hypogaea Cowpea Rhizobium 70-240
Field bean: Vicia faba R. leguminosarum 45--300
Red clovers: Trifolium sp. R. trifolii 45--350
Lucerne: Medicago sativa R. meliloti 50-460

2. Actinomycetes symbioses
Alder grove: Alnus sp. Frankia 50-150
Dunes with: Casuarina equisetifolia Frankia 58
Bog myrtle: Myrica gale Frankia 27
Sea buckthorn: Hippophae rhamnoides Frankia 180

3. Blue-green algal symbioses

Fern: Azolla Anabaena azollae 50-80
Lichen: Lobaria Nostoc (+ fungus) 5--10
Cycad: Macrozamia riedlei Nostoc 18
Higher plant: Gunnera Nostoc 58

4. Associative symbioses and casual associations

Grasses: Paspalum nota tum Azotobacter paspali 10
Zea mays (maize) or Klebsiella, Bacillus,
Saccharum officinarum Azospirillum, and Beijerinckia <3
(sugar cane)

5. Free living organisms

Blue-green algae 80-120
Azotobacter, Clostridium, Klebsiella and other <1
free-living Nrfixers

Rates are given in kgN fixed ha- 1 yr- 1 . The data are grossly estimated from field
studies and are therefore subject to large variations (climate, soil, coverage by the plants).
Table 2 is compiled principally from data cited by SPRENT 1979, POSTGATE 1978, VINCENT
1974, MISHUSTIN and SHIL'NIKOVA 1971 and the Rothamsted Experimental Station (Rot-
hamsted Subject Day 1975)

tive symbiosis between free-living nitrogen fixing bacteria and plants are dis-
cussed by DOBEREINER (Chap. II.3, this Vol.).
Table 2 contains gross estimates of the maximum nitrogen fixation rates
by different organisms in the field. It is generally accepted that legumes contrib-
ute most to global nitrogen fixation. The waterfern Azalla containing the blue-
green alga Anabaena azallae forms a green manure which increases the nitrogen
supply substantially in rice fields of southeast Asia (see SPRENT 1979). The
contributions of the free-living organisms to global nitrogen fixation are mar-
ginal. Probably the only organisms of serious ecological importance are the
blue-green algae which populate many different environments and can improve
the nitrogen balance significantly.
II.1 Physiology, Biochemistry and Genetics of Dinitrogen Fixation 247

1.4 Substrates of Nitrogenase

Nitrogenase catalyzes the reduction of a variety of substrates (Table 3), most

being small molecules which, like N 2, contain triple bonds. The apparent Km
values depend on the organisms assayed and whether intact cells or purified
nitrogenase is tested. Nevertheless, dinitrogen clearly has the highest affinity
for nitrogenase (Table 3). The Km value for proton reduction may be as low
but this cannot be determined. The reduction of C 2H 2 to C 2H 4 , discovered
by DILWORlH (1966) and SCHOLLHORN and BURRIS (1967), has become a stan-
dard technique for determining nitrogen fixation activities. These two gases
are easily separated· and determined quantitatively by gas chromatography
where less than 10 pmol of C 2H 4 in the assays can be detected accurately.
This method is quick and inexpensive compared with other nitrogenase assays
(e.g. 15N incorporation by mass spectrometry). For detailed descriptions of
the techniques the reader should consult articles by BURRIS (1972, 1975) and
HARDY and HOLSTEN (1977). All nitrogen-fixing bacteria except the methane
oxidizer, Methylosinus, when grown on methane, will reduce acetylene to ethyl-
ene. In methane-oxidizing organisms acetylene inhibits the oxidation of CH 4
to CH 3 0H (DEBoNT and MULDER 1974), thus preventing the supply of energy
and reducing equivalents to nitrogenase.
The acetylene reduction test is therefore a useful qualitative indication of
nitrogen-fixing ability. However, it is often unwise to make quantitative assess-
ments of nitrogen-fixing activity in the field on the basis of a 6- and 2-electron
requirement for N2 and C 2H 2 reduction respectively. Quantitative field tests

Table 3. The different substrates of nitrogenase

Substrate Product Electrons Apparent Km Relative

utilized formed transferred (M) rate

1. N2 2NH3 6 2 x 10- 5-1 X 10- 4 1

2. N; NH 3, N2 2 2 x 10- 4-1 X 10- 3 3
3. N 20 N 2, H 2O 2 1 X 10- 3 3
4. HCN CH 4 , NH3 6 2 x 10- 4-4 X 10- 3 0.6
(CH 3NH 2) 4
5. CH 3CN C 2H 6 , NH3 6 5 X 10- 1 0.004
6. CH 3NC CH 3NH 2, CH 4 6 2 x 10- 4-2 X 10- 3 0.8
7. C 2H 2 C 2H 4 2 4 x 10- 4-1 X 10- 3 3-5
8.2H+ H2 2 Not determined 4
H2 H2
C C
9. /\
HC=CH
/\
H 2C-CH 2
,CH 3CH=CH 2

These data are adopted from tables given by DALTON and MORTENSON 1972, ZUMFT
and MORTENSON 1975, ZUMFT 1976, BURNS and HARDY 1975, McKENNA and HUANG
1979 where the original references are given.
248 H. BOTHE et al. :

have shown that the C 2H 2/N 2 conversion factor varies from 2 to 25 (HARDY
et al. 1973). Possible reasons for such variations include: (1) a greater proportion
of electrons are diverted to H2 production when N 2 rather than C 2H 2 is the
substrate. (2) C 2H 2 inhibits several metabolic processes, including nitrification,
denitrification, methane oxidation, methanogenesis and bacterial hydrogenase
activity. It also inhibits the processes which oxidize the ethylene which is formed
naturally in soils. In addition, repeated exposures to acetylene may induce con-
formational changes in nitrogenase which enhance the rate of acetylene reduc-
tion (see YATES 1980). Such effects must distort the simple C 2H 2:N 2 conversion
factor.
It has been suggested that nitrogenase may have evolved to detoxify acetylene
or cyanides present during the anaerobic phase of the earth's development.
After this period combined nitrogen became limiting and the enzyme developed
the ability to reduce N2 (SILVER and POSTGATE 1973). An alternative theory
(POSTGATE 1974) is that nitrogenase is a relatively young enzyme evolved on
plasmids which were incorporated into the genetic material of various organ-
isms. Either possibility would explain the rather haphazard distribution of nitro-
genase amongst prokaryotic organisms, but both theories lack substantial evi-
dence for their support.

2 Biochemistry of Nitrogen Fixation

2.1 Introduction

Nitrogenase consists of two iron-sulphur proteins (see Fig. 1), which, individual-
ly, have no detectable enzymic activity but together catalyze ATP hydrolysis
and all substrate reductions. One of these is called the MoFe protein; it contains
up to 36 Fe atoms, an almost equivalent number of acid labile sulphide ions,
and 2 Mo atoms per 220,000 M.W. The other, the Fe protein, contains one
[4 Fe-4 S] cluster per ",60,000 M.W. Both proteins are oxygen-sensitive, the
Fe protein extremely so. Thus an anaerobic environment is necessary to assay
nitrogenase activity. A quick and simple assay involves sodium dithionite as

electron enzyme electron nitrogenase proteins

donor carrier MgATP

~~~;~tel
NADH 1 - - - - e' --+--- e' --+----=::.-.
e' ---4-----1
or oxidoreductase ferredoxin or Fe protein
H2 ~ ______ flavodoxin
~ ______- L______ ______
~ ~

Fig. 1. Scheme: The nitrogenase complex

11.1 Physiology, Biochemistry and Genetics of Dinitrogen Fixation 249

the reductant, an ATP-generating system consisting of creatine phosphokinase,

ADP and phosphocreatine, nitrogenase proteins in an anaerobic environment
at neutral pH and acetylene as the reducible substrate. The A TP-generating
system is included because ADP inhibits nitrogenase activity and must be kept
at a negligible concentration to ensure a linear reduction rate.
Nitrogenase proteins have to be purified under strict anaerobic conditions.
These include sparging all buffers with 02-free N 2, using low concentrations
of sodium dithionite to react with the remaining traces of O 2 and dithiothreitol
to protect sulfhydryl groups. When these precautions are taken, active nitrogen-
ase proteins can be isolated by conventional techniques of enzyme purification,
including ion exchange and molecular sieve chromatography. Nitrogenase pro-
teins are strongly coloured (brown), stable at room temperature and are one
of the major protein constituents in N 2-fixing bacteria. These three properties
ease purification procedures considerably. Review articles covering the early
developments of the biochemistry of nitrogen fixation include those by HARDY
and BURNS (1968), BURRIS (1971), ZUMFT and MORTENSON (1975) and EADY
and SMITH (1979). Purification procedures for nitrogenase have been reviewed
by EADY (1980).

2.2 Nomenclature of Nitrogenase Proteins

The MoFe protein of nitrogenase is called component 1, molybdoferredoxin,

azofermo or nitrogen reductase; the Fe protein is called component 2, azoferre-
doxin, azofer or nitrogenase reductase. The names nitrogen- and nitrogenase-
reductase were given in an attempt to assign biochemical roles (HAGEMAN and
BURRIS 1978a). The nomenclature used in this review is mechanistically noncom-
mittal: involving the initials of the parent bacterium and the order of elution
from DEAE cellulose. Thus Av1, Kp1 and Cp1 represent the MoFe proteins
and Av2, Kp2 and Cp2 the Fe proteins of Azotobacter vinelandii, Klebsiella
pneumoniae and Clostridium pasteurianum respectively (EADY et al. 1972).

2.3 Physicochemical Properties of Nitrogenase Proteins

Nitrogenase proteins have been isolated from more than 20 organisms; ten
MoFe proteins and seven Fe proteins have been partly characterized (Table
4). Important basic criteria include molecular weights, subunit contents and
numbers of metal atoms and acid labile sulphur groups per protein. The molecu-
lar weights of the MoFe proteins range from 180,000 to 270,000, they contain
18 to 36 g atoms of Fe and 1 to 2 g atoms of Mo mol- 1 ; acid-labile sulphides,
when determined, are usually 2 to 3 g ions mol- 1 below the Fe values. It is
not known whether the range of molecular weights and metal contents reported
reflect species differences, impure preparations or the inadequacy of analytical
methods. The most active preparations, obtained either from A. vinelandii, K.
pneumoniae or C. pasteurianum contain > 30 g atoms Fe and 2 g atoms Mo/mol.
It is now generally accepted that the MoFe proteins have four subunits of
250 H. BOTHE et al. :

Table 4. Properties of highly purified nitrogenase proteins C

MoFe proteins

Cpl Bpi Kpl Rrl Ac1 Anc1 b

Molecular weights
By gel filtration 210,000 215,000 220,000 230,000 216,000 230,000
(215,000)
By ultracentrifugation 201,000 200,400
By disc electrophoresis 240,000 217,000 234,000 230,000 216,000
From amino acid 221,800 229,000 227,000
composition

Sum of subunit
Mol.wts. 220,000 221,000 240,000
Subunit Two Not Two Not Probably Two
Composition types known types known 2 types types
Mol wts of 50,700 51,300 58,500 60,000 52,800
the subunits 59,500 59,600 (56,000) Approx. 55,000

Mo
g atom/mol 2 2 2.01 1.7 (2) 1.9±0.3 2

Fe
g atom/mol Approx.24 33 32.5±3 20 22±2 20
(25-30)
Acid-labile sulfide
g ion/mol Approx. 24 21 16.7±1" (19-22) 20±2 20

O 2 sensitivity
t1/2 min 10 10

a This value refers to a Kpl preparation containing 17.5 g atoms Fe/mol

b HALLENBECK et al. (1979)
C Adapted from Yates 1980

two types (rx 2P2). Analytical evidence, involving sodiumdodecylsulphate-poly-

acrylamide gel electrophoresis (SDS-PAGE) followed by peptide and amino
acid analysis of the separated subunits, as well as genetical evidence, indicating
the presence of separate genes for the two subunit types, support this conclusion.
The Fe proteins of nitrogenase from most sources contain two identical
subunits when analyzed by SDS-PAGE. This has been confirmed by amino
acid sequence analysis of Cp2 (TANAKA et al. 1978). The one exception is the
Fe protein from R. rubrum which contains two non-identical subunits ofmolecu-
lar weights 31,500 and 30,000 respectively (LUDDEN and BURRIS 1976). Molecu-
lar weights of Fe proteins range from 55,000 to 75,000 (Table 4).
All nitrogenase proteins are acidic, containing a preponderance of glutamic
and aspartic acid residues. An arithmetical analysis of their relatedness shows
that they are quite similar (EADY and SMIlH 1979), although Cp 1 is distinguished
II.1 Physiology, Biochemistry and Genetics of Dinitrogen Fixation 251

Fe proteins

Xal Rjl Rl1 Cp2 Bp2 Kp2 Rr2 Ac2, Xa2

235,000 180,000 184,000 56,000 55,000 62,000 (65,000) 63,000

36,000 68,000
237,000 62,000 61,500
223,500 67,800 57,000 82,900

237,400 228,000 228,000 58,000 69,200 61,600 73,000

Two One One One Two One One
? ?
types type type type types type type
61,200 55,000 57,000 27,000 34,600 30,000 30,800 36.500
57,500 31,500

2.2 1.3

23.1 28.2 4 3.2 4 3.5 4 3.8

(4.15)

20.1 26.2 4 3.6 3.85 3.9 3.9 2.4

>0.5 0.75

Cp Clostridium pasteurianum; Bp Bacillus polymyxa; Kp Klebsiella pneumoniae; Rr Rho-

dospirillum rubrum; Ac Azotobacter chroococcum; Ane. Anabaena cylindrica; Xa Xantho-
bacter autotrophicus; Rj Rhizobium japonicum; Rl Rhizobium leguminosarum bacteroids

by its relatively low tryptophan content. These similarities may account in part
for the ability of MoFe protein from one species to combine with the Fe protein
from another to produce active nitrogenases. Sequence analysis shows that Cp 2
protein has 273 amino acid residues. The basic amino acids occur randomly
while the acidic residues occur in clusters. It has no sequence homology with
other types of iron-sulphur proteins (e.g. ferredoxins). A partial sequence of
the oc and P subunit of Av1 has also been made (LUNDELL and HbwARD
1978).

2.4 Metal Ousters in Nitrogenase Proteins

Nitrogenase proteins can be studied by several spectroscopic techniques. These
include electron paramagnetic resonance (EPR), Mossbauer, nuclear magnetic
252 H. BOTHE ct al.:

resonance (NMR), extended X-ray absorption edge and fine structure spectro-
scopies and ultraviolet (UV)-visible spectrophotometry.
Metal clusters in iron-sulphur proteins are, usually, of two types: 2- and
4-iron although some proteins apparently contain 3-iron clusters (see Sect. 3).
The Fe atoms are linked to each other by acid-labile sulphide bridges and
to the protein by cysteine residues. The [2 Fe-2 S] clusters are planar and the
[4 Fe-4 S] type are cubane. The former undergoes oxidation and reduction be-
tween the fully oxidized, ferric-ferric state and the partly reduced ferric-ferrous
state. [4 Fe-4 S] clusters can, theoretically, be in anyone of five oxidation levels,
designated 0 to 4 (latest nomenclature, Eur J Biochem 93 : 427-430 1978). States
1 and 3 are paramagnetic and have epr signals; states 0, 2 and 4 are diamagnetic
and EPR-silent. According to the most recent publications there are two major
types of iron centres in the MoFe protein of nitrogenase: the iron-molybdenum
cofactor (FeMoco), and cubane type [4 Fe-4 S] clusters. The number of clusters
of each type within the molecule is uncertain but tentative conclusions are
possible on the basis of evidence which will be discussed in the following sections.

2.5 EPR and Mossbauer Spectroscopy on the MoFe Protein

The MoFe protein of nitrogenase possesses a unique EPR spectrum with g

values close to 4.3,3.7 and 2.01. It is observed only below 30 K and is probably
a single rhombic EPR signal associated with one type of Fe, probably that
in the FeMo cofactor. Early Mossbauer studies by SMITH and LANG (1974)
on Kp 1 protein containing 18 g atoms Fe mol- 1 showed the presence of 3 Fe
species containing 2, 8 and 8 Fe atoms respectively. MUNCK et al. (1975) made
similar findings with Av 1 containing 18 g atoms Fe mol-I.
Recent studies (for references see HUYHN et al. 1979) have reinterpreted
the Mossbauer data using Cpl, Kpl and Av1. The general conclusions are
that dithionite-reduced MoFe proteins should contain 30 ± 2 g atoms Fe mol- 1
in four states.
These include (1) 12 Fe atoms associated with the EPR signal (called M EPR ),
(2) a component, prominent at 4.2 K, containing 41 % of the total Fe, (3) a
Fe2 + component (13% total Fe) and (4) an unidentified state (S)containing
2 Fe mol- 1. Components (2) and (3) together form 3 to 4 [4 Fe-4 S] clusters,
each containing 3 Fe from component (2) and 1 Fe from component (3). These
P clusters, as they are called, are new structures since they are EPR-silent when
oxidized or reduced, but their Mossbauer spectra indicate that they are closely
related to conventional [4 Fe-4 S] clusters.
Cluster displacement techniques, using organic solvents to extract metal clus-
ters from proteins and UV -visible spectrophotometry or EPR spectroscopy to
identify them, have confirmed the Mossbauer data. ORME-JOHNSON et al. (1977)
extracted Cpl with o-xylyl-a, a'-dithiol in the presence of DMSO (80%) in
tris buffer, pH 8.0, and associated the extruded Fe-S clusters with the apopro-
teins of B. polymyxa ferredoxin [4 Fe-4 S] or adrenodoxin [2 Fe-2 S]. From
the resulting EPR spectra they calculated 3 [4 Fe-4 S] clusters in Cp 1 containing
24 g atoms Fe mol- 1 . KURTZ et al. (1979) analyzed extruded cores ofCpl spec-
11.1 Physiology, Biochemistry and Genetics of Dinitrogen Fixation 253

trophotometrically and found 3.4 to 4 (interpreted as 4) 4 Fe-4 S centres/(x'2fJ2

subunit. Because the clusters were not so readily extruded as [4 Fe-4 S] clusters
in ferredoxins, the authors speculated that the P-cluster of MoFe proteins may
not be bound to the protein by cysteine residues.

2.6 The FeMo Cofactor and the Fe Protein

Many molybdo-enzymes (e.g. xanthine oxidase, nitrate reductase and nitrogen-

ase) contain Mo cofactors; the nitrogenase cofactor also contains Fe. Mutants
of A. vinelandii and K. pneumoniae containing inactive MoFe proteins which
are activated by acid-treated Avl or Kpl have been used to isolate the FeMo
cofactors (called FeMoco) from these organisms (SHAH and BRILL 1977, SMITH
1980). The cofactor contains the Mossbauer spectral features of M EPR and its
EPR spectrum is similar to that of the holoprotein (RAWLINGS et al. 1978,
SMITH 1980).
FeMoco is unstable and 02-sensitive. It contains Mo, Fe and acid-labile
sulphur, but their proportions are disputed: that from Av1 has a Mo:Fe:S 2-
ratio of 1: 8 : 6 according to SHAH and BRILL (1977) or 1: 6: 6 to 8 according
to HUYHN et al. (1979). FeMoco from Kpl has a ratio of 1 :8:4 (B.E. SMITH
personal communication). In some preparations it has an associated peptide.
The Mo atom in FeMoco or nitrogenase cannot be studied by Mossbauer,
nor, because it is either diamagnetic or masked by Fe signals, by EPR. Extended
X-ray absorption spectroscopy (EXAFS), however, has yielded interesting data
about the environment of Mo in both FeMoco and the MoFe protein. The
absorption edge and fine structure spectra of Cpl, Avl and FeMoco are very
similar and show the presence of Mo-S and Mo-Fe ligands but not Mo =
unless the protein is damaged by 02' Fourier transform analysis of EXAFS
°
data indicate that Mo is bound to 3 or 4 S atoms at 2.36 A and 2 or 3 Fe
atoms at 2.7 A spacings. There may also be one or two S atoms bound with
longer Mo-S bonds (CRAMER et al. 1978). Chemical complexes containing struc-
tures analogous to [4 Fe-4 S] clusters with Mo replacing one Fe have been
synthesized (e.g. CHRISTOU et al. 1978). Their EXAFS spectra are very similar
to that of nitrogenase but their chemical compositions do not correspond with
FeMoco.
To summarize, present knowledge of the structure of the MoFe protein
indicates that it is an (X,2fJ2 iron-sulphur protein containing 2 Mo atoms and
30±2 Fe atoms/M.W. 220,000. The Mo atoms and 12-16 Fe atoms form the
FeMo cofactor(s) which is EPR-active and is thought to contain the active
site of nitrogenase. Sixteen Fe atoms occur in [4 Fe-4 S] clusters which are
EPR-silent and which may not be bound to the protein by cysteine 'residues.
Two other Fe atoms are spectroscopically inert and may be involved in the
quaternary structure (SMITH and LANG 1974). This analysis conforms to a sym-
metrical structure. It is not known whether FeMoco contains 1 or 2 Mo mol-i.
All the Fe in the Fe protein of nitrogenase is present as a single [4 Fe-4 S]
cluster. Evidence from four different spectroscopic techniques indicate that this
is so: (1) Mossbauer spectroscopy; SMITH and LANG (1974) showed that the
254 H. BOTI-IE et al.:

Mossbauer spectrum of reduced Kp2 was similar to that of reduced [4 Fe-4 S]

ferredoxin; (2) cluster displacement techniques: ORME-JOHNSON et al. (1977) ex-
tracted Cp2 Fe and showed that it was similar to that of C. pasteurianum
ferredoxin which has [4 Fe-4 S] clusters; (3) Linear electric field effect (LEFE),
which measures the shift in the EPR spectrum under the influence of a strong
electric field. This shift is large when the charge around the centre is non-
symmetrical as in a [4 Fe-4 S] cluster. Such was the case with Cp2 (ORME-
JOHNSON et al. 1977). (4) Magnetic circular dichroism (MCD): the MCD of
reduced Av2 or Kp2 was similar to that of a [4 Fe-4 S] ferredoxin (STEPHENS
et al. 1979).
So much evidence in favour of a single [4 Fe-4 S] cluster raised two interest-
ing questions. Firstly, how is a single cluster distributed between two identical
subunits? Secondly, the oxidized Fe protein will take up two electrons mol- 1
upon reduction, one rapidly and the other slowly. Only the rapid phase corre-
lates with the appearance of the EPR spectrum (rhombic with g values close
to 2.05, 1.94 and 1.87) associated with the [4 Fe-4 S] cluster. Neither the nature
of this second electron-accepting site nor its relevance to enzyme activity is
known. Further evidence indicating its presence is that the EPR spectrum of
the Fe protein integrates to less than 1 electron mol- 1 which suggests the pres-
ence of two interacting paramagnetic centres (LOWE 1979). It is difficult to
envisage the nature of the second centre since all the Fe atoms appear to be
associated with the [4 Fe-4 S] centre.

2.7 Nitrogenase Proteins in Photosynthetic Organisms

Some nitrogenase proteins have been purified from symbiotic and photosyn-
thetic sources (Table 4). MoFe proteins from symbiotic systems and from A.
cylindrica conform with the general pattern. Homogeneous preparations of Fe
proteins have not yet been obtained from these organisms.
The Fe protein from the photosynthetic bacterium R. rubrum is unusual
in that it has non-identical subunits, is associated with equimolar concentrations
of adenine, ribose and phosphate ions and requires a protein-activating factor
which may function to remove these ligands (LUDDEN and BURRIS 1978).

2.8 The Mechanism of Nitrogenase Activity

As mentioned earlier, nitrogenase requires two component proteins, a source

of electrons, ATP, Mg2+ plus an ATP-generating system, and an anaerobic
environment. In the absence of any added reducible substrate, such as N2 or
C 2H 2, nitrogenase will convert protons to H 2. This ATP-dependent H2 evolu-
tion also occurs, though to a lesser degree, in the presence of added reducible
substrates, thus complicating kinetic studies of their reduction. The questions
to be resolved are obvious: they include (a) the roles of the two component
proteins; (b) the roles of ATP and ADP; (c) the number and nature of the
reducing sites and (d) the mechanism of N 2 reduction.
11.1 Physiology, Biochemistry and Genetics of Dinitrogen Fixation 255

2.8.1 The Roles of the Two Proteins

The present opinion of the function of the two proteins is as follows: the Fe
protein accepts electrons from the electron carrier (see Sect. 3) and then binds
Mg ATP which causes a conformational change to lower the redox potential;
the Fe protein· Mg ATP then complexes with the MoFe protein and transfers
one electron to the latter with the concomitant hydrolysis of Mg A TP to form
Mg ADP. The electrons are then transferred to the reducible substrate by the
MoFe protein. The series of steps are depicted by the following equations:

Feox~Fered (1)
Fered+MgATP~Fered'MgATP (2)
Fe red ' Mg ATP+MoFe'N2~Fered'Mg ATP' MoFe·N 2 (3)
Fered' Mg ATP' MoFe' N2~Feox' Mg ADP' MoFesuper_red'N2+Pi (4)
Feox ' Mg ADP'MoFesuper_red'N2~Feox' +Mg ADP+MoFe+NH 3 (5)
There are many unresolved questions in this scheme: why are 2 to 2.5 ATP's
hydrolyzed per electron transferred? Does the complex between the two proteins
dissociate after each electron transfer or is it longer-lived than the enzyme turn-
over? What is the rate-limiting step responsible for a relatively slow enzyme
turnover? What is the stoichiometry of complex formation? Other questions
will emerge later.
There are several lines of evidence for the role of the Fe protein. (1) The
oxidized Fe-protein is rapidly reduced by sodium dithionite (yATES et al. 1975),
a reduction which is inhibited by ADP (Ki = 9 J.1M; THORNELEY and YATES
unpublished data), whereas the oxidized MoFe protein is reduced slowly and
no further than the half-reduced state. Using stopped-flow spectrophotometry
THORNELEY (1975) obtained direct evidence that Kp2 transferred electrons to
Kp 1 in an ATP-dependent reaction. This substantiated earlier deductions from
EPR measurements (see ZUMFT and MORTENSON 1975). EADY et al. (1979)
showed that ATP was hydrolyzed concomitantly with this electron transfer
(2.3 Mg ATP hydrolyzed per electron transferred). (2) Evidence that Mg ATP
and Mg ADP bind to the Fe protein is summarized as follows: Mg ATP in-
creases the sensitivity of the Fe protein to O 2, towards the sulfhydryl group
reagent 5,5'-dithiobis-(2-nitrobenzoic acid) (DTNB) and towards Fe-chelating
reagents. Mg ADP, on the other hand, has little effect on these sensitivities.
Either Mg ATP or Mg ADP change the EPR spectrum of the Fe protein from
a rhombic to an axial form and also change the mid-point potential from - 280
to '" -400 mV (see ZUMFT and MORTENSON 1975). Tso and BURRIS (1973)
measured the level of 14C_ATP or 14C_ADP in Cp2 after equilibration between
an aqueous phase and Sephadex G 50. They showed that Mg2+ was necessary
to bind both nucleotides and calculated two binding sites for Mg ATP with
a dissociation constant (K D) of 16.7 J.1M and one for Mg ADP (KD 5.2 J.1M).
They concluded that Mg ATP inhibited Mg ADP binding, whereas Mg ADP
inhibited ATP binding at one site but enhanced binding at the other site. WYETH
and ORME-JOHNSON (personal communication) have repeated these experiments
256 H. BOTHE et al.:

but obtain three to five binding sites for Mg ATP. BISHOP et al. (1977) studied
the binding of Mn 2 + or Mg2+ to Kp2 by enhancement of the proton NMR
relaxation rate. They showed four sites of Mn or Mg-binding and evidence
for ternary complexes involving metal, protein and ATP or ADP. However,
they could not distinguish the number of nucleotide binding sites. To summarize,
although the evidence is not conclusive, it is reasonable to assume that the
Fe protein possesses at least two binding sites for Mg ATP and one or more
for MgADP.
The MoFe protein is capable of existing in three oxidation states, two of
which are associated with enzyme activity. These are the EPR signal state and
a "super-reduced" state obtained during enzyme activity by A TP-mediated elec-
tron transfer from the Fe protein. An oxidized state, obtained by treating the
EPR-signal-state with thionine, is not considered to be part of the active process
because the mid-point potential of this reaction is not conserved: i.e. it varies
between MoFe proteins from different sources (O'DONNELL and SMITH 1978).
Addition of acetylene or a change in pH can effect the EPR spectrum of
the MoFe protein which indicates that either protons or acetylene bind to the
MoFe protein but not to the Fe protein. This is evidence that the MoFe protein
may contain the substrate-reducing site. CN- binds to both nitrogenase pro-
teins, but this may be a non-specific effeCt (BIGGINS and KELLY 1973). There
is no evidence that N 2 binds to either of the isolated proteins.
Electron transfer between the nitrogenase proteins cannot occur without
complex formation beforehand. The stoichiometry and lifetime of this complex
are questions which have received much attention since they bear on the location
and nature of the active site. Ultracentrifuge experiments showed that the nitro-
genase proteins from K. pneumoniae or A. chroococcum aggregated in equimolar
complexes (EADY 1973). Titration data, measuring nitrogenase activity when
the Fe protein concentration was constant and the MoFe protein concentration
increased, indicated an equimolar ratio for maximum activity. Studies of the
dilution effect, which is a disproportionately low activity at low enzyme concen-
tration, attributed to dissociation of the component proteins, also indicated
a 1: 1 molar complex. However, titration data also exist supporting the presence
of an active 2:1 Fe protein: MoFe protein ratio (BERGERSEN and TURNER 1973,
HAGEMAN and BURRIS 1978b). The latter authors implied that the nature of
the reducible substrates may influence the stoichiometry of the complex. All
these approaches can be criticized on the grounds that these apparent molar
ratios may be influenced by the percentages of inactive proteins in preparations;
such experiments should be performed with proteins of the highest specific
activity.
The occurrence, of the dilution effect and the fact that reductant-free Kp 1
and Kp2 form a complex, which suggests that the two nitrogenase proteins
are complexed at the end of the catalytic cycle, tentatively support the concept
of a long-lived complex. However, HAGEMAN and BURRIS (1978a), using a high
ratio of Av1:Av2 to slow down electron-flow, found that ATP hydrolysis was
linear whereas a time lag occurred before· H+ was reduced to H 2 • They con-
cluded that electrons were transferred singly and that dissociation of Avl and
Av2 occurred after each transfer. otherwise the second electron would transfer
II.1 Physiology, Biochemistry and Genetics of Dinitrogen Fixation 257

rapidly following the first and H2 would be evolved immediately. From these
observations they proposed that the MoFe protein should be called nitrogen
reductase and the Fe protein nitrogenase reductase.

2.8.2 Evidence for Interaction of Mg ATP and Mg ADP

with the MoFe Protein
HAGEMAN and BURRIS (1978 a) asserted that A TP is associated only with electron
transfer from the Fe- to the MoFe-protein. However, nitrogenase possesses
a reductant-independent ATPase which requires both proteins (JENG et al. 1970).
Secondly, ATP hydrolysis is almost independent of electron transfer at high
ratios of MoFe protein to Fe protein or at low temperatures (see MORTENSON
and THORNELEY 1979). Thirdly, the ratio of moles of ATP hydrolyzed per elec-
tron transferred varied with reducible substrate in the presence of antibody
to the MoFe protein (RENNIE et al. 1978). Finally, chemical analogues of ATP
alter the electron distribution between H2 production and C 2H 2 reduction by
nitrogenase (see MORTENSON and THORNELEY 1979). These observations suggest,
but do not unequivocally prove, a role for ATP associated with the MoFe
protein.
THORNELEY and CORNISH-BoWDEN (1977) found that the kinetics of inhibi-
tion by Mg ADP of the electron transfer between the proteins were different
from those ofH 2 production suggesting that Mg ADP inhibited the latter activi-
ty at a different site, possibly on the MoFe protein.

2.8.3 The Nature of the Active Site(s)

Nitrogenase is capable of reducing several substrates (Table 3) but kinetic studies
suggest that the active site for each substrate may not be identical. The results
of these studies indicate, according to RIVERA-ORTIZ and BURRIS (1975), that
there are four different substrate-reducing sites: (1) N2 and, possibly, N 20;
(2) azide, HCN and methylisocyanide; (3) C 2H 2; (4) H+. Whether the recently
discovered substrate cyclopropene (McKENNA and HUANG 1979) is reduced
at yet a fifth site is not known. Finally, kinetic data are available to suggest
that, ,at high concentrations, acetylene induces a second reduction site in vitro
(DAVIS etal. 1979) and in vivo (WALKER and YATES 1978a). Alternatively,
substrates may be reduced by different (substrate-induced) forms of the enzyme
(SMITH BE et al. 1976).
The chemical nature of the reducing sites for nitrogenase substrates has
not been resolved, but there is considerable evidence that metal atoms, probably
Fe and Mo, are involved. Mossbauer and EPR spectroscopy and stopped-flow
spectrophotometry all indicate that Fe atoms are oxidized and reduced during
the catalytic cycle. LOWE et al. (1978), from EPR and Mossbauer data, have
devised an elegant scheme explaining the oxidation/reduction cycles of FeS
clusters during H+ and C 2H 2 reduction. However, it is likely that Mo is the
metal involved in the active substrate reducing site, whereas FeS clusters are
involved in electron transfer. However, evidence for the location and function
of Mo is circumstantial: demolybdo-Cp 1 is inactive (see ZUMFT and MORTENSON
258 H. BOTHE et al. :

1975) and addition of FeMoco (the FeMo cofactor) activates extracts from
mutants of A. vinelandii or K. pneumoniae defective in one of the nitrogen
fixation genes (nifB) (SHAH and BRILL 1977a, SMITH 1980). Organometallic
complexes where N2 is bound to Mo can be reduced to yield NH 3.

2.8.4 Pathways ofN 2-Reduction

NHt or NH3 was the only known intermediate ofN 2-fixation until THORNELEY
et al. (1978) devised conditions for accumulating an intermediate which yielded
hydrazine upon treatment of the enzyme with acid or alkali. The fact that
nitrogenase catalyzed N rdependent hydrogen-deuterium (HD) exchange led
several investigators to suggest that an unstable di-imide-type enzyme-bound
intermediate occurred which reacted with D2 according to the equation:

E-N=N + D2 ~ E-N 2 + 2HD

I I
H H
(NEWTON et al. 1977). However, THORNELEY et al. (1978) thought it more likely
that the intermediate which gave rise to hydrazine was an enzyme-bound com-
plex of type E-Mo;;;N:.:.:NH 2. A chemical model to explain how N2 bound
to Mo can be reduced to ammonia has been proposed by CHATT et al. (1978):
H+ H+
Mo=N=N ) Mo=N····NH ) Mo::::::::N=NH 2
e- e-
(I) (II) (III)

N2 i 3 H+
1 H+

Mo + NH3 (
Mo=N + NH3 ( Mo::::::::N-NH3
3 e- e-
(VI) (V) (IV)

The scheme envisages the reduction of the Mo in six one-electron steps, the
electrons being provided via the iron-sulphur clusters in the enzyme. In this
way, the Mo need not change its oxidation by as many as six redox states
during the catalytic cycle.
The model must explain ATP-dependent H2 evolution by nitrogenase. CHATT
(1980) envisaged the active site in a cleft in the molecule which limited proton
access. Mo at this site is bound to hydrogen atoms which are replaced by N 2.
This means an obligatory evolution of 1 mol Hz per N 2 reduced according
to the equation:

N 2+8 H+ +8e--t2 NH 3 +H z

In the absence of N z, or other reducible substrates, protons combine with H-

to yield H 2. CO, H2 or D2 inhibited N2 reduction by displacing N2 from
11.1 Physiology, Biochemistry and Genetics of Dinitrogen Fixation 259

this site. CHATT suggested that this fully reduced (3 H-) state of the enzyme
is the only state in which N2 can be reduced and in which D2 can bind and
form deuteride ions which, in tum, are displaced by N 2 to give HD exchange.
When CO is bound the protons react with the third H- to discharge H 2.
The scheme does not satisfactorily explain why very little or no HD exchange
occurs in the absence of N 2. Nevertheless, it is an attractive working model
based on the observed properties of the enzyme and on characterized organome-
tallic complexes involving Mo and N in various oxidation states.

3 Electron Transport to Nitrogenase

3.1 Introduction

This subject has been reviewed in detail by several authors (YATES 1977, EVANS
and PmLLIPS 1975, JOCH and VALENTINE 1972). In experiments with isolated
nitrogenases, the most commonly used reductant is sodium dithionite which
does not require additional electron transport components in the assays. Natural
carriers which transfer the electrons to nitrogenase are ferredoxin and flavo-
dQxin. These proteins can be substituted by the low redox potential electron
carriers methyl- or benzylviologen. Natural electron donors which reduce ferre-
doxin or flavodoxin and thus couple cell metabolism to nitrogen fixation are
pyruvate, NAD(P)H and, possibly, formate or hydrogen (see Table 5). The
reactions between electron donors and carriers are enzyme-catalyzed.

Table 5. The electron transport to nitrogenase in some representative microorganisms

Organism Electron Electron Enzyme
carrier donor

1. Clostridium Ferredoxin Pyruvate Pyruvate: fd oxidoreductase

pasteurianum (flavodoxin) NADH NADH: fd oxidoreductase
2. Bacillus 2 ferredoxins Pyruvate Pyruvate: fd oxidoreductase
polymyxa Formate: CO 2 oxidoreductase (?)
3. Azotobacter sp. Flavodoxin NADH (?) Possibly NADH: flavodoxin
2 ferredoxins oxidoreductase
4. Rhizobium sp. Ferredoxin NADH (?) NADH: ferredoxin oxidoreductase
(flavodoxin)
5. Mycobacterium 2 ferredoxins ? ?
flavum
6. Klebsiella Flavodoxin? ? ?
pneumoniae
7. Blue-green Ferredoxin NADPH NADPH: ferredoxin oxidoreductase
algae (flavodoxin) Pyruvate Pyruvate: ferredoxin oxidoreductase
Photo- Hydrogenase via photosystem I
system I, H2
260 H. BOTHE et al.:

3.2 Ferredoxins (for a review see YOCH and CARITHERS 1979)

These iron-sulphur proteins are active not only in nitrogen fixation but also
in a variety of redox reactions. Examples are the photosynthetic NADP+ -reduc-
tion and the cyclic photophosphorylation of chloroplasts, the assimilatory thio-
sulphonate and nitrite reductions in plants, the conversion of nitrate to nitrite
in some organisms (blue-green algae, Azotobacter) as well as the hydrogenase
and the pyruvate phosphoroclastic reactions (see Sect. 3.4) of Clostridium pas-
teurianum. Such reactivity requires an electronegative redox potential of about
-400 mV, which is a characteristic feature of ferredoxins. Electron transfer
is mediated through [FeS] clusters in the protein (see Sect. 2). Ferredoxins
contain either one [2 Fe-2 S], one [4 Fe-4 S] or two [4 Fe-4 S] clusters. In the
redox reactions, they usually operate between the + 2 and + 1 oxidation levels
(see Sect. 2). They are paramagnetic when reduced, have a typical EPR-spectrum
with a Lande-Factor g<2 and are diamagnetic when oxidized. Other types
of ferredoxins have been isolated from nitrogen-fixing A. vinelandii (SWEENEY
et al. 1975), X. flavus (BERNDT et al. 1978) and Rhodospirillum rubrum (YOCH
et al. 1977). These are paramagnetic in the oxidized state and EPR-silent in
the reduced form. They operate between the + 2 and + 3 oxidation states.
Some organisms (B. polymyxa, X.flavus, R. rubrum, Azotobacter) form constitu-
tively more than one soluble ferredoxin which can act independently as electron
carriers to nitrogenase when assayed in vitro. The physiological roles of these
different proteins have not been established in many cases. However, in R.
rubrum ferredoxin I is formed only in the light, whereas ferredoxin II is present
both in the light and during heterotrophic growth in the dark. In addition
to these soluble ferredoxins, R. rubrum contains two iron-sulphur proteins (ferre-
doxin III and IV) which are integral parts of the chromatophore membranes
(SHANMUGAM et al. 1972, YOCH et al. 1977). Other nitrogen-fixing microorgan-
isms also have membrane-bound iron-sulphur centres with unresolved biological
and chemical characteristics.

3.3 Flavodoxins (for a review see MAYHEW and LUDWIG 1975)

These are small proteins of about 20,000 m.w. having one molecule of FMN
in the prosthetic group. They are distinguished from other flavoproteins by
the formation of a stable radical or semiquinone form in high yields. Conse-
quently, flavodoxins have the two different redox couples: the oxidized/semi-
quinone and semiquinone/hydroquinone forms. Only the latter redox couple,
having a potential of - 350 to - 500 mV, is sufficiently electronegative to act
in lieu of ferredoxin. In all reactions tested (including nitrogen fixation) flavo-
doxins substitute for ferredoxins, and, in most cases, with similar specific activi-
ties. The physiological role of flavodoxin has not been established in all cases.
In some organisms (Clostridium, blue-green algae, Rhodospirillum) flavodoxin
is only formed under iron deficiency when the biosynthesis of ferredoxin is
limited. In Azotobacter and the non-nitrogen-fIxing Escherichia coli both ferre-
doxin and flavodoxin are formed constitutively. Surprisingly, Azotobacter
ILl Physiology, Biochemistry and Genetics of Dinitrogen Fixation 261

appears to be the only organism which may utilize flavodoxin exclusively as

the electron carrier to nitrogenase (for the details see YATES 1980). Xanthobacter
flavus, on the other hand, forms two ferredoxins and no flavodoxin (BoTHE
and YATES 1976). Also in Rhizobium, ferredoxin is predominantly formed where-
as the flavoprotein is synthesized in small amounts and only under some culture
conditions (CARTER et al. 1980).

3.4 Electron Donors (for a review see YATES 1977, 1980, BOTHE et al. 1980)

The best-known electron donor is pyruvate (see Table 5) which, in the presence
of coenzyme A, is split to acety1coenzyme A and CO 2; the remaining two
electrons reduce ferredoxin. This pyruvate phosphoroclastic reaction requires
the participation of the enzyme pyruvate: ferredoxin oxidoreductase which was
purified and characterized in some detail from Clostridium acidiurici (UYEDA
and RABINOWITZ 1971). Acety1coenzyme A can be converted to acetylphosphate
and further to ATP catalyzed by phosphotransacetylase and acetate kinase.
Thus the pyruvate cleavage provides reduced ferredoxin and ATP for nitrogen-
ase and other reactions. Although ideal for nitrogen fixation, the pyruvate
phosphoroclastic reaction occurs only in Clostridium species, in Bacillus poly-
myxa and in blue-green algae. Pyruvate is also the best electron donor to nitro-
genase in K. pneumoniae, though the manner of the pyruvate cleavage has not
been established. Even in C. pasteurianum pyruvate is not the only electron
donor. Fermentation balances indicated that approximately 25% of the elec-
trons to reduce ferredoxin came from NADH (TRADER et al. 1969). The enzyme
involved, NADH: ferredoxin oxidoreductase, requires acety1coenzyme A as an
allosteric effector. Ferredoxin was also reduced by H2 plus hydrogenase or
formate plus formate-dehydrogenase by Clostridial extracts. However, this is
believed to be an artefact of cell-free systems, since hydrogenase favours H 2-
evolution and formate dehydrogenase the formate cleavage in vivo (TRADER
et al. 1969).
In the aerobically growing filamentous blue-green algae Nostoc and Ana-
baena nitrogen fixation takes place in the heterocysts. Neighbouring vegetative
cells supply these specialized cells with carbohydrates which are degraded via
the hexosemonophosphate shunt to provide NADPH. Heterocysts were shown
to contain high levels of the two key enzymes of this pathway, glucose-6 phos-
phate dehydrogenase and 6-phosphogluconate dehydrogenase. Since NADPH:
ferredoxin oxidoreductase is also present in heterocysts, a major route of elec-
tron donation is likely to comprise glucose-6-phosphate, the dehydrogenase,
NADPH, NADPH: ferredoxin· oxidoreductase, ferredoxin, nitrogenase (see
BOTHE 1970 and several publications thereafter). An additional or alternative
pathway may involve pyruvate as an electron donor to nitrogenase. Pyruvate:
ferredoxin oxidoreductase was observed in blue-green algae (LEACH and CARR
1971) and was found to be particularly active under nitrogen-fixing conditions
(BOTHE et al. 1974). The presence of this enzyme in heterocysts, however,
remains to be established. Nitrogen fixation is light-stimulated in isolated hetero-
cysts. Light is presumably necessary to meet the A TP requirements for nitrogen
262 H. BOTHE et al. :

fixation by cyclic photophosphorylation, whereas the reducing equivalents may

be generated in the dark. However, the involvement of photosystem I in the
transfer of electrons to nitrogenase cannot be ruled out. Photo system I is defini-
tely involved in electron donation to nitrogenase when H2 and hydrogenase
are electron donors in heterocysts (see Sect. 6). In these cells, the electrons
cannot directly come from the photosynthetic water-splitting reaction, since
they lack photosystem II. In the unicellular, non-heterocystous alga, Gloeocapsa,
nitrogen fixation is more dependent on the operation of the two photosynthetic
reactions than in heterocystous forms (GALLON 1980).
The pathways of electron transport to nitrogenase in Azotobacter, Rhizobium
or X. flavus and surprisingly Klebsellia pneumoniae are less well understood.
There are two conflicting theories of electron transfer to nitrogenase in Azoto-
bacter. According to BENEMANN et al. (1971), the main electron donor is
NADPH. These authors observed low, but positive C 2H 2-reduction activities
with the system NADPH, NADPH:ferredoxin oxidoreductase (from spinach),
Azotobacter ferredoxin and flavodoxin, unidentified factor(s) and nitrogenase.
The rates which they reported were approximately 1% of the nitrogen-fixing
activity in vivo, and the spinach reductase could not be substituted by an enzyme
native to Azotobacter. NADH also supports nitrogenase activity: YATES (1971)
observed significant C 2H 2-reduction activities using NADH, a homogeneous
dehydrogenase from Azotobacter chroococcum, benzylviologen and nitrogenase.
However, benzylviologen could not be replaced by ferredoxin, flavodoxin or
any other component from Azotobacter. Since the redox potential ofNAD(P)H/
NAD(P)+ is -320 mY, the reduction of ferredoxin or flavodoxin requires a
high level of NAD(P)H in the cells. However, HAAKER et al. (1974) did not
find a correlation between the NAD(P)H content and the nitrogenase activity
in intact cells. Moreover, high NAD(P)H levels inhibit key enzymes of Azoto-
bacter (YATES and JONES 1974). To circumvent such difficulties, HAAKER et al.
(1974) postulated that an energized membrane was the driving force to generate
reductants for nitrogenase in aerobes. According to the Mitchell hypothesis,
the proton motive force consists of the A pH and A If/ terms. Since LAANE
et al. (1979a, b) did not find a correlation between proton gradients and nitro-
genase activity, they assumed that the membrane potential was the driving force
for generating reductant to nitrogenase in Azotobacter and Rhizobium legumino-
sarum. This latter assumption was substantiated by the observation that the
membrane de-energizing agent 4: 5: 6: 7 tetrachloro-2-trifluoromethyl-benzimid-
azole (TTFB) inhibited nitrogenase activity but not respiratory ATP-formation
in Azotobacter. This hypothesis, though easily reconcilable with modem think-
ing, is difficult to verify experimentally. In particular, reduction of flavodoxin
beyond the semiquinone state by the membrane-bound NADH: ferredoxin oxi-
doreductase (LAANE et al. 1978) has not been demonstrated in vitro. The situa-
tion is even less resolved in Rhizobium. Poly-fi-hydroxybutyrate, the main car-
bon-storage compound of nodules, was thought to be the electron donor for
nitrogenase. However, neither poly-fi-hydroxybutyrate nor the activity of the
NAD +-linked fi-hydroxybutyrate dehydrogenase correlated with nitrogenase ac-
tivity during different experimental conditions (EVANS and PIllLLlPS 1975).
II.1 Physiology, Biochemistry and Genetics of Dinitrogen Fixation 263

In conclusion, many aspects of the electron transfer to nitrogenase have

still to be elucidated. It has become clear, however, that different organisms
utilize different pathways and that one organism does not restrict itself to only
one electron donor.

4 Mechanisms to Protect Nitrogenase Against

Damage by Oxygen

4.1 In Free-Living Organisms

Aerobic nitrogen-fixing bacteria require oxygen to produce ATP and must,

therefore, possess mechanisms to prevent oxygen from inhibiting or damaging
nitrogenase. Even in Azotobacter, the most oxygen-tolerant of nitrogen-fixers,
nitrogenase is inhibited by high concentrations of O 2, This can easily be demon-
strated by measuring nitrogenase activity at increasing oxygen tensions when
a "bell-shaped" curve of activity versus p02 is produced, increasing to a maxi-
mum as the level of available ATP increases and then declining as excess O 2
inhibits nitrogenase activity. The range of oxygen tensions at which the nitrogen-
ase can function increases with the population density and the carbon-substrate
concentration. Nitrogenase activity can also be "switched off" abruptly by
increasing the oxygen solution rate to a level faster than the culture can respire.
The mechanisms which protect nitrogenase from damage by O 2 in aerobes
have been reviewed exhaustively (seeYATES 1977). The simplest concept is that
of respiratory protection: the organism respires rapidly enough to maintain
an anaerobic environment at the nitrogenase site. Azotobacter has the highest
known respiratory quotient amongst bacteria and has a branched electron trans-
port chain which is largely uncoupled to ATP production at high O 2 tensions,
thereby allowing an increased respiration rate.
A second protection mechanism against O 2 in Azotobacter is known as con-
formational protection. This is indicated by:
1. nitrogenase in crude extracts is oxygen-tolerant whereas the purified enzyme
is oxygen-sensitive;
2. nitrogenase activity is inhibited but apparently not damaged by a sudden
increase in the oxygen solution rate in vivo.
This protection is provided by an equimolar complex of the nitrogenase
proteins and a third iron-sulphur protein previously known as the Shethna
[2 Fe-2 S] protein (SHETHNA 1970). Mg2+ ions are necessary for maximum sta-
bility of this complex. It has been suggested that this complex also regulates
nitrogenase activity as follows: the complex forms between the oxidized proteins
and is enzymically inactive; it dissociates and nitrogenase becomes active only
when the [2 Fe-2 S] protein is reduced (SCHERINGS et al. 1977; for a recent
review see ROBSON and POSTGATE 1980). Other organisms which may also possess
some form of conformational protection include X. jlavus (BIGGINS and POST-
GATE 1970) and K. pneumoniae (HILL 1976).
264 H. BOTHE et al. :

Another defence against excess O 2 is the formation of slime, which may

decrease the oxygen solution rate in liquid culture. HILL (1971) suggested that
massive colony formation by Derxia gummosa which involved slime production
under nitrogen-fixing conditions, also protected against excess O 2 during growth
on agar. However, WILCOCKSON (1977) found no correlation between slime
production and O 2 by nitrogen-fixing Klebsiella on solid agar.
The mechanism whereby oxygen causes damage to nitrogenase is not re-
solved. The most favoured hypothesis at present is that the products of O 2
metabolism, hydrogen peroxide, superoxide- and hydroxyl-radicals, are the dam-
aging agents. However, although aerobic organisms possess superoxide dismu-
tase and catalase, enzymes which convert these products to O 2 and H 20, there
is no convincing evidence that they serve to protect nitrogenase.

4.2 The Heterocysts of Blue-Green Algae

In the filaments of blue-green algae, the heterocysts can easily be distinguished

from the bluish vegetative cells by their thick envelope and their more greenish
colour due to the absence (or reduced content) of phycocyanin. There is little
doubt among the different experimenters that heterocysts are the site of nitrogen-
ase under aerobic growth conditions. Heterocysts are often said to provide
an anaerobic compartment for nitrogenase. Clearly, since they are devoid of
the photosynthetic water-splitting system II, nitrogenase inside these cells need
not be protected from oxygen evolved photosynthetically. It has been suggested
that nitrogenase is protected against O 2 from the air by four unique glycolipids,
detected in the laminated layer of the heterocyst envelope, which may bind
oxygen, thus providing a diffusion barrier (LAMBEIN and WOLK 1973). Indeed,
nitrogen fixation is sensitive to air in Anabaena mutants lacking these glycolipids
(HAURY and WOLK 1978). However, heterocysts must have access to O 2, since
they can respire and support N2 fixation in the dark by oxidative phosphoryla-
tion. These observations clearly indicate that oxygen must reach the inside of
heterocysts. The gas does not necessarily penetrate the cell envelope, although
this cannot be ruled out, but can come from vegetative cells to which heterocysts
are connected by plasma bridges called microplasmodesmata (FAY and LANG
1971). Thus, heterocysts, like other aerobes, must protect nitrogenase from
damage by oxygen inside the cell. Little information is available about the
nature of such protective mechanisms; respiratory protection probably plays
an essential role. Site I of the respiratory chain is NADPH-linked in blue-green
algae. This is consistent with the findings that the major route of hexose degrada-
tion in heterocysts is the oxidative pentose phosphate cycle where the activity
levels of the glucose-6-phosphate- and the 6-phosphogluconate dehydrogenases
are particularly high. Thus the oxidation ofNADPH by respiration may remove
oxygen from the nitrogenase site. The Knallgas reaction (see Sect. 6) may also
protect nitrogenase from damage by O 2, Recent investigations (HAURY and
WOLK 1978, GOTTO et al. 1979) with Oz-sensitive mutants from Anabaena indi-
cate that the protective mechanisms for nitrogen fixation in heterocysts involve
different phenomena and that there may be several sites sensitive to oxygen.
II.1 Physiology, Biochemistry and Genetics of Dinitrogen Fixation 265

4.3 The Role of Leghaemoglobin in Legume Nodules

Leghaemoglobin is abundant in legume nodules and, although it was recognized

many years ago to be necessary for effective nitrogen fixation, it is only recently,
through the work of ApPLEBY, BERGERSEN, WITTENBERG and their colleagues,
that its role is understood (ApPLEBY et al. 1976). It has two roles: it facilitates
the transport of oxygen from the bacteroid envelope to the bacteroid surface
and also acts as. an "oxygen buffer" to prevent too great a variation in the
concentration of oxygen at the bacteroid surface. In this second function it
could act to protect nitrogenase from damage by oxygen, but current opinion
is that the oxygen concentration within the nodule is too low to cause damage
to nitrogenase.
One unresolved problem is the location of leghaemoglobin: is it confined
within the bacteroid envelope or also present in the cell cytoplasm? Some investi-
gators have observed leghaemoglobin in nodule electron micrographs to be
present both within the bacteroid envelope and in the cell cytoplasm, but its
presence in the latter may have been due to damaged and "leaky" bacteroid
envelopes.

5 Regulation of Nitrogenase Activity and Biosynthesis

Both biosynthesis and activity of the enzyme are subject to complex regulations
which are not fully understood, and which may differ between organisms.

5.1 Regulation of Nitrogenase Biosynthesis

Nitrogenase, generally, is only formed in the absence of combined nitrogen

(N0 3 , NHt or organic nitrogen) in the culture medium. Earlier physiological
studies of nif regulation in K. pneumoniae which showed that nitrogenase was
repressed in the presence of ammonia have been substantiated by results which
show that nitrogenase polypeptides are not synthesized in cultures grown on
ammonia (EADY et al. 1978). Furthermore, the synthesis of polypeptides recently
assigned to other nif genes is also similarly repressed (ROBERTS et al. 1978).
TUBB and POSTGATE (1973) demonstrated that ammonia represses nitrogenase
synthesis at the transcriptional level using inhibitors of initiation of mRNA
synthesis (rifampicin) and of protein synthesis (chloramphenicol) during dere-
pression of samples taken from fully repressed, sulphate-limited continuous cul-
tures.
The construction of small amplifiable plasmids carrying nifDNA sequences
has made it possible to detect nif-specific mRNA directly using the Southern
hybridisation technique (SOUTHERN 1975). JANSSEN et al. (1980) detected mRNA
which hybridized to DNA restriction fragments representing six of the seven
known nif operons in cultures derepressed for nif. The nif] operon was not
examined since it had not been cloned at the time of the experiment. In contrast
266 H. BOTHE et al. :

nif-specific mRNA was not detected in ammonia-grown cultures which were

repressed for nif. The conclusion from these results is that the nif operons
for which mRNA was detectable under the experimental conditions used are
not transcribed in vivo in ammonia-grown cultures.
N 2 does not induce the synthesis of the enzyme, since this occurs when
the organisms are incubated under argon. Nitrogenase, therefore, is derepressed
in the absence of fixed N. There is evidence that glutamine synthetase (GS)
rather than ammonia regulates nitrogenase biosynthesis. GS plays a key role
in ammonia incorporation by nitrogen-fixing organisms where it catalyzes the
reaction:
glutamate + NHt + ATP --+ glutamine + ADP + Pi (1)
Glutamate is reformed from glutamine by glutamate synthase ( = L-glutamine:
2-oxoglutarate amidotransferase, abbreviated GOGAT) (MEERS et al. 1970):
L-glutamine + a-ketoglutarate ~ 2 L-glutamate (2)
The reductant may either be NADPH, NADH or reduced ferredoxin. Other
ammonia-assimilating pathways (e.g. those catalyzed by glutamate or alanine
dehydrogenases) are of minor physiological significance in N 2-fixing bacteria.
The apparent KM-value of these enzymes for NHt lie between 1 x 10- 4 and
1 x 10 - 3 M, concentrations which are probably higher than those normally pres-
ent in N 2-fixing cells.
That GS regulates nitrogenase biosynthesis is indicated by studies with
mutants (e.g. see STREICHER et al. 1974) and inhibitors (see BISHOP et al. 1975).
Glutamine-requiring mutants from K. pneumoniae that lack active glutamine
synthetase do not express nitrogenase activity. When such mutants reacquire
glutamine synthetase as a result of conjugation with an episome carrying the
genes of this enzyme, they gain the ability to synthesize nitrogenase. In addition
Klebsiella mutants which were unable to regulate GS by adenylylation were
able to synthesize nitrogenase even in the presence of NHt. L-methionine-DL-
sulfoximine and L-methioninesulfone, both analogues of glutamine, specifically
block glutamine synthetase. These compounds also derepress nitrogenase syn-
thesis in Klebsiella, Rhizobium or Anabaena when grown in the presence of
NHt·
The mechanism by which glutamine synthetase affects nitrogenase synthesis
is not clearly understood. Its activity is controlled by its degree of adenylylation;
the deadenylylated enzyme being the active form. At high concentration NHt
ions are assimilated via glutamate dehydrogenase. This results in a high ratio
of glutamine to a-ketoglutarate which, in turn, inhibits the uridylyl transferase
involved in the control of glutamine synthetase activity. This prevents deadenyly-
lation of glutamine synthetase and, consequently, derepression. of nitrogenase.
Apparently, NHt does not repress nitrogenase synthesis via glutamine synthe-
tase in all systems (BISHOP et al. 1976). Carbamylphosphate was also reported
to regulate nitrogenase biosynthesis (see YATES 1977). However, it is not clear
at present whether this compound acts by producing NHt.
Repression of nitrogenase biosynthesis by oxygen has been studied in K.
pneumoniae and A. chroococcum. The method used was to pulse-label the organ-
II.1 Physiology, Biochemistry and Genetics of Dinitrogen Fixation 267

ism with either 14C_ or 35S-labelled amino acids during derepression of nitrogen-
ase. In both instances the repression by O 2 followed different kinetics from
that by NHt. Moreover, mutants of K. pneumoniae exist in which nitrogenase
synthesis is repressed by O 2 but not by NHt (EADY et al. 1978, ROBSON 1979).
These observations indicate clearly that the mechanism of repression by O 2
is different from that by NHt.
There is some evidence to suggest that lack of Mo also represses nitrogenase
synthesis (see YATES and EADY 1980) but other evidence indicates that lack
of Mo merely results in the synthesis of inactive nitrogenase polypeptides. It
is also possible that active MoFe protein of nitrogenase, or a Mo storage protein
act as regulators (see POSTGATE et al. 1981).

5.2 Regulation of Nitrogenase Activity

Ammonium ions do not inhibit nitrogenase activity in vitro but do inhibit
in several organisms in vivo, particularly in the photosynthetic nitrogen-fixing
bacteria Rhodopseudomonas capsulata, R. palustris and Rhodospirillum rubrum.
Complete inhibition of nitrogenase occurs within minutes of adding NHt to
the cells (see MEYER et al. 1978). A possible explanation is that NHt causes
an increase in glutamine synthetase activity thereby depleting the intracellular
ATP pool. The situation in Azotobacter is more complex: addition of NHt
causes a partial inhibition of nitrogenase activity which is reversible in some
instances (see YATES and EADY 1980). LAANE et al. (1979a) showed that NHt
de-energized membranes in A. vinelandii and proposed that NHt inhibited the
energized-membrane-mediated electron transfer to nitrogenase in this organism.
By contrast, NHt did not inhibit membrane energization nor nitrogenase activity
of bacteroids, nor does it inhibit nitrogenase in K. pneumoniae or C. pasteu-
rianum (see YATES and EADY 1980).
Inhibition by oxygen has been considered earlier (Sect. 4), but it is important
to distinguish between damage to the enzyme and fine control of activity. Most
aero bic bacteria can tolerate only low levels of O 2 when fixing N 2; high levels
of O 2 damage nitrogenase within the cell. The exception to this rule is Azoto-
bacter where nitrogenase activity is apparently undamaged at high O2 levels
because of conformational protection. This implies that O 2 , which may inhibit
nitrogenase activity by auto-oxidation of the electron donor flavodoxin hydro-
quinone (yATES and JONES 1974) may also regulate activity through the oxida-
tion states of the proteins comprising the conformationally-protected complex
(Sect. 4.1).
Nitrogenase activity is energy-intensive, hydrolyzing 12-15 molecules of ATP
per molecule ofN 2 reduced. Growth yield measurements indicate that this value
may be even higher in vivo. Because it has a low catalytic turnover rate (",0.5 s)
it is necessary for bacterial cells to synthesize large amounts of the enzyme:
it may comprise as much as 15% of protein in crude cell supernatant (ORME-
JOHNSON et al. 1977). Perhaps the most direct control of nitrogenase activity
is by the ATPjADP ratio: ADP is a potent inhibitor of nitrogenase activity
and its apparent Ki is lower than the KM for ATP. These properties mean
that a small increase in Mg ADP concentration, within the right range, should
268 H. BOTHE et al. :

cause a rapid inhibition of nitrogenase activity (THORNELEY and CORNISH-

BOWDEN 1977).
Depending on the organism, ATP may be provided by fermentation, oxida-
tive phosphorylation or - in the phototrophs - from cyclic photophosphoryla-
tion. In blue-green algae, the reducing equivalents to nitrogenase are not pro-
vided directly through the photolysis of water and via the two photo systems.
Nitrogen-fixation rates in these organisms are dependent on the size of the
reductant pool of carbohydrates (LEX and STEWART 1973). Carbohydrates are
probably the major limiting factor in all nitrogen-fixing organisms outside labo-
ratory-controlled conditions. This is particularly true oflegume-Rhizobium sym-
biosis (see HARDY et a1. 1978). CO 2-enriched atmospheres increased the dry
weight of plants, nodules and the specific nitrogen-fixing activity of legumes.
Thus approaches to improve the availability of photosynthate to the nodule
would enhance nitrogen fixation and nitrogen input to the plant. Such studies,
aiming to construct more efficient plant-Rhizobium symbioses may be more
rewarding than attempts to find mutants with higher nitrogenase activities.

6 The Hydrogenase-Nitrogenase Relationship

Recent compendia in this field are from SCHLEGEL and SCHNEIDER (1978) and
a special issue of Biochimie (Vol. 60: 3, 1978).
The reaction 2 H+ +2e=H 2 is catalyzed by enzymes. Cells may contain
three different proteins that catalyze either uptake, or evolution or both uptake
and evolution of molecular H2 under different physiological conditions. Three
clearly distinguishable proteins have been described to occur in N 2-fixing micro-
organisms:
a) The" classical" hydrogenase from Clostridium, and other anaerobes or
facultative anaerobes, couples to ferredoxin, flavodoxin or a variety of redox
dyes. It has a molecular weight of about 60,000, mediates an exchange reaction
with D 20, is sensitive to inhibition by carbon monoxide and contains three
[4 Fe: 4 S] centres of which only one is believed to undergo oxidation/reduction
during catalysis. Under physiological conditions, the enzyme catalyzes the evolu-
tion of molecular hydrogen to dispose of excess reducing equivalents generated
during fermentation.
b) The unidirectional H 2-evolution catalyzed by nitrogenase can easily be
distinguished from the hydrogenase-dependent formation of the gas by its in-
sensitivity to carbon monoxide and dependence on ATP. The mechanism of
this ATP-dependent reduction of protons is not fully understood at present.
Nitrogenase is believed to have at least two different H 2-evolving sites (Sect. 2).
Clearly, all hydrogen formation catalyzed by nitrogenase is a loss of energy
and reducing power for the organisms. It should be kept in mind that H 2-
evolution always occurs during nitrogenase activity in vitro, though with vari-
able rates, but is often marginal or zero in intact organisms.
c) The third type of enzyme activity is by the unidirectional, H 2-uptake
hydrogenase first observed in Azotobacter by PHELPS and WILSON (1941). This
II.1 Physiology, Biochemistry and Genetics of Dinitrogen Fixation 269

enzyme is present in all aerobic nitrogen-fixing organisms and is membrane-

bound in Azotobacter, blue-green algae and bacteroids. High levels of such
an enzyme have recently been observed in the anaerobic bacterium C. pasteu-
rianum, where it is obviously soluble (CHEN and BLANCHARD 1978). The uptake
hydrogenase activity is not restricted to cells that are actually fixing N 2, but
N 2-fixing conditions enhance specific activities. Enzyme levels also generally
increase after exposure to molecular hydrogen. At present several groups are
attempting to purify such enzymes to homogeneity (see SCHLEGEL and
SCHNEIDER 1978). The enzymes from Rhizobium, Azotobacter and blue-green
algae are stable to oxygen when particulate and do not (or only poorly) couple
with ferredoxin, flavodoxin or other low potential natural electron carriers.
There are four possible roles for the uptake hydrogenase in N 2-fixing aerobes
(mainly suggested by DIXON 1972). Some of these have been verified experimen-
tally:
a) Since H2 is produced by nitrogenase, this gas could be reoxidized by
the uptake hydrogenase. Such a H 2-utilization could be coupled to respiratory
oxygen consumption. It could provide the organisms with ATP and thus mini-
mize the loss of energy caused by the ATP-dependent H 2-formation by nitrogen-
ase. Indeed, A TP formation by the oxyhydrogen reaction was demonstrated
in particles from Azotobacter and heterocysts from Anabaena. Such a function
of hydrogenase, to recycle energy lost by nitrogenase, may be particularly impor-
tant under carbon-limited conditions as shown for Azotobacter, Rhizobium and
blue-green algae (see WALKER and YATES 1978 b, BOTHE et al. 1978).
b) Hydrogen uptake catalyzed by hydrogenase could protect nitrogenase
from damage by oxygen by means of the Knallgas reaction and aid respiratory
protection of nitrogenase at low carbon substrate concentration. Indeed, C2H 2-
reduction was found to be more stable to oxygen when hydrogen was present
in assays with intact Azotobacter, Anabaena variabilis or heterocysts isolated
from the latter organism.
c) Since high concentrations of molecular hydrogen were shown to inhibit
N 2-reduction, hydrogenase could act to remove hydrogen produced by nitrogen-
ase in vivo. Such a function is unlikely, since H2 evolved by nitrogenase probably
disperses too rapidly to the environment and does not reach the concentrations
required to have any effect on nitrogenase in nature (WALKER and YATES 1978b).
d) H2 could be an electron donor to nitrogenase. Hydrogenase, isolated
from Clostridium pasteurianum, was used to provide electrons to cell-free nitro-
genase from Clostridium, Azotobacter or Anabaena. H2 may act as an electron
donor to nitrogenase in Azotobacter chroococcum in vivo, particularly when
grown under low carbon substrate concentrations (WALKER and YATES 1978 a,
b). In photosynthetic bacteria, the electron donation from H2 to nitrogenase
requires light activation: under such conditions, H2 was shown to be the best
electron donor for C 2H 2-reduction in heterocysts isolated from Anabaena
(EISBRENNER and BOTHE 1979). These authors postulated two different pathways
of H 2-utilization in blue-green algae: an oxygen-linked one proceeding in the
dark and a light-dependent one requiring the strict exclusion of oxygen. A
common intermediate of both pathways is presumably plastoquinone, which,
however, is not necessarily the immediate electron acceptor from hydrogenase.
This observation that H2 and hydrogenase feed in electrons close to the plasto-
270 H. BOTHE et al. :

electron donor
(pyruvate,NADH,
NADPH,H2) 2H+

C02 (Calvin cycle)

'----t---e---I-----l e---~s-::;£= NOi (nitrite reductasE

","",,-- ~, NOj (nitrate reductas
(flavodoxin) N2 (nitrogenase)
enzyme ferredoxin nitrogenase hydrogenase ATP":--:::'_ 02

electron donor 2H'

/
02 (respiration)
oxidized

Fig. 2. Scheme: The nitrogenase-hydrogenase relationship

quinone site is consistent with the fact that electron donation from H2 to nitro-
genase is dependent on photosystem I in blue-green algae. It also explains why
hydrogenase cannot catalyze the evolution of H2 in vivo.
In addition to the uptake hydrogenase, a reversible hydrogenase was re-
ported to occur in Anabaena (HALLENBECK and BENEMANN 1978). This blue-
green alga is the only aerobic nitrogen-fixing organism reported to contain
both enzymes. However, BOTHE et al. (1980) were unable to confirm this result
and considered such low activity to be an artefact of an isolated enzyme catalyz-
ing the uptake of molecular hydrogen under physiological conditions.
As already mentioned, uptake hydrogenase is not restricted to nitrogen-
fixing microorganisms. Besides nitrogen fixation, H2 can support several other
reactions (Fig. 2). In green algae, hydrogen acts as an electron donor in photo-
synthetic CO 2-fixation, termed photoreduction (GAFFRON 1940). Such a reaction
is not demonstrable in the blue-green algae commonly used in the laboratory
but was shown to occur in Oscillatoria limnetica (BELKIN and PADAN 1978).
This alga is capable of anoxygenic photosynthesis and can fix CO 2 at the expense
of H 2S or H 2. H2-dependent CO 2-fixation has recently been demonstrated in
Rhizobium japonicum and Derxia gummosa which synthesize ribulose-l,5-bis-
phosphate carboxylase under chemoautotrophic growth conditions (SIMPSON
et al. 1979, PEDROSA et al. 1980). H2 and hydrogenase can also support nitrate
and nitrite reductions in blue-green algae where these reactions are strictly de-
pendent on light activation (BOTHE et al. 1980).
H 2-formation by microorganisms has attracted special attention in the past
years. In particular, BENEMANN and co-workers (see WEISSMAN and BENEMANN
1977) suggested that photoautotrophically growing blue-green algae might
evolve sufficient hydrogen and oxygen to be used as a source for fuel. Among
the different investigators, there is a general agreement that all-hydrogen evolu-
tion is due to nitrogenase in intact cells of aerobic nitrogen-fixing organisms.
However, such yields are variable, and depend on the strain used and the culture
conditions employed. For example, batch-grown cultures of Azotobacter chroo-
coccum did not evolve significant amounts of H 2, whereas O 2-, N 2- as well
as C-limited cultures showed considerable H 2-production (WALKER and YATES
1978b). Nostoc muscorum apparently evolved relatively large quantities of H2
II.1 Physiology, Biochemistry and Genetics of Dinitrogen Fixation 271

at 22° but not at 32°C (ERNST et al. 1979). Clearly, the optimal conditions
for maximal H 2-production have not yet been worked out and the physiology
is not fully understood. A major factor influencing H 2-formations is the uptake
hydrogenase. According to ALBRECHT et al. (1979), Rhizobium strains which
evolve the gas fail to synthesize an active hydrogenase, whereas the enzyme
is effective in those cells which do not produce the gas. The evolution rate
can be increased by adding inhibitors of Hruptake activity. For example, CO
and high concentrations of C 2H 2 unmask the evolution of hydrogen from the
usually faster uptake of the gas by hydrogenase in batch-grown Azotobacter
(SMITH LA et al. 1976) and Anabaena (BoTHE et al. 1977). Similarly, the rate
of Hrformation can be increased in Anabaena when the utilization of the gas
by the Knallgas reaction is prevented by strict exclusion of oxygen from the
assays (ERNST et al. 1979).
It is now clearly established that aerobic nitrogen-fixing microorganisms
have the potential for hydrogen formation, but it should be kept in mind that
the maximal rate is limited by the activity of nitrogenase in the cells. An approxi-
mate value for the maximum possible rate of H2 production can be deduced
from the rate of C 2H 4 production which is an approximate indication of the
maximum electron utilisation by nitrogenase. C 2H 2 reduction rates in blue-green
algae are 1/5 to 1/10 of the CO 2-fixation activity which is in accord with the
requirements of the cell for carbon and nitrogen. Photosynthetic rates are gener-
ally 100-250 Ilmol CO 2 fixed h -1 mg- 1 chlorophyll. Thus H2 productions of
more than 25 Ilmol reported in the literature cannot be sustained by the organ-
isms over a longer period. This low, theoretically achievable maximum value
is the reason for our belief that blue-green algae or other aerobic nitrogen-fixing
organisms are unlikely to produce sufficient H2 to provide power utilizable
for mankind.
In contrast, another application of the studies on the hydrogenase-nitrogen-
ase relationship is conceivable. Rhizobium strains that possess uptake hydrogen-
ase were shown to fix nitrogen more effectively and to grow faster than strains
without an active enzyme (ALBRECHT et al. 1979). Many Rhizobia have poor
uptake hydrogenase activities. Therefore studies to construct new Rhizobia
legume symbioses which have a high efficiency for H 2-utilization could be useful.

7 The Molecular and Genetic Characterization

of Nitrogen Fixation Genes

7.1 Introduction

The genetics of nitrogen-fixing systems developed rapidly during the 1970's

and the majority of genetic studies were carried out with Klebsiella pneumoniae.
The relatively few studies of nitrogen fixation genes (nif) in Azotobacter (SHAH
et al. 1973, GORDON and BRILL 1972, BISHOP and BRILL 1977), Rhizobium (MAIER
and BRILL 1976), Rhodopseudomonas (WALL et al. 1975, MARRS et al. 1977)
272 H. BOTHE et al.:

and cyanobacteria (STEWART and SINGH 1975) have been mainly concerned
with the isolation and preliminary characterization of nif mutants and the devel-
opment of genetic transfer systems. Our review of recent progress in nif genetics
will therefore be confined mainly to K. pneumoniae. Further details and useful
background information on the genetics of nitrogen fixation were published
in HOLLAENDER et al. (1977).
The initial transductional (STREICHER et al. 1971) and transconjugational
(DIXON and POSTGATE 1972) transfer of nif showed that the genes for histidine
biosynthesis (his) and nif were linked in K. pneumoniae. Large numbers of K.
pneumoniae nif mutants were subsequently isolated and characterized and the
mutations in all of these have been mapped in one region close to the operator
end of his. Furthermore, the isolation of n'-.r hybrids (DIXON and POSTGATE
1972) by the intergeneric transfer of his and nif from K. pneumoniae to a non-
nitrogen-fixing Escherichia coli indicated that all the genes specific to nitrogenase
synthesis were located near his.

7.2 The nif Genes

The close linkage of his and nif facilitated the construction of self-transmissible
plasmids which carry the his-nif region of the K. pneumoniae genome (CANNON
et al. 1976 and DIXON et al. 1976). Plasmid pRD1 carries the transfer and antibi-
otic resistance genes of the parent plasmid, RP4 and also the chromosomal
genes gnd rjb his nif shiA. This plasmid and several derivatives have been used
for complementation analyses and fine structure mapping of nif mutations.
pRD1 has also been used to study the expression of nif genes in other bacteria
(POSTGATE and KRISHNAPILLAI 1977), and was the source of DNA for cloning
the nif genes.
Fifteen nif genes (n ifQ , nifB, nifA, nifL, nifF, nifM, nijV, nifS, nijU, nifN,
nifE, nifK, nifD, nifH and nifJ) have now been identified primarily by comple-
mentation tests between plasmid and chromosomally located nif mutations
(DIXON et al. 1977, ELMERICH et al. 1978, MACNEIL et al. 1978, MERRICK et al.
1978, 1980). These mutations included nifpoint mutations on the K. pneumoniae
chromosome, on pRD1 and its derivatives and insertion mutations obtained
by bacteriophage Mu- and transposon-(Tn4, Tn7 and Tn10) directed mutagene-
sis. Nif deletions, derived from insertion mutants, were used for fine structure
mapping of nif point mutations. The gene order obtained from these studies
was, his ... nifQBALFMVSUNEKDHJ (Fig. 3). Complementation analysis of
insertion mutations which exert transcriptional polarity indicated that the 15
nif genes were arranged in 7 operons (Fig. 3).
Genetic studies of polar mutations in the multicistronic nif operons have
shown that these operons are transcribed in the same direction as the his operon
of K. pneumoniae. RNA-DNA hydridization studies using separated strands
of DNA which represented the his and six of the seven nif operons showed
that his-specific and nif-specific mRNA synthesized in vivo hybridized exclusive-
ly to one strand. The conclusions from these results is that those nif operons
11.1 Physiology, Biochemistry and Genetics of Dinitrogen Fixation 273

his
DG
-- QBALFMV5UNE KDH
---------- I
n i f operons.
approximate
I I I I II I I II I II gene locations
o ;. 8 12 16 20 -1.4
I I i
32 kilobasepairs40
I
I
I
I
I
II I i
I I I
I I I
I
I
I
!
I I I
Q
I
I
I
I
A L
III!!!
M
I IIUI
5' N
I lin! III!!! II
K 'H
I 1/11
I insertions
I I
I III III I iii 1111111 I II! 1111 1111111
BiI F V :U E D J :I locati ons II
I
I
I
I
I
I
I
I
I I !
I
I
I I
I I :
R R R B H I RB B !BBR BR HBI R I restriction B I H
I I I II I
I I I I I III II I II I I II I
I I I I t i j I I I I I I I I I I
5 55 5 5Bg X5X 5 5 5 5 Bg5 Bg X map
R R pCRA37
I I
H pCMl
I ~
R R pSA30
I I
R R pGR 112
I I
R H pSA32
L--1
B X
I I
pMCl
H X
I I pMC2
B B pMC3
I I

Fig. 3. Maps of; (1) his and nif genes, (2) nifinsertion mutations (±250 basepairs), and
(3) restriction sites in the his-nif region of the K. pneumoniae chromosome. Arrows indicate
nif operons and their direction of transcription - nifF and nifJ are monocistronic operons.
Symbols for restriction endonucleases are: R EcoR1; H HindIII; B BamH1; S SaIl;
X Xhol; Bg BglIl. pCRA37, pCMl, pSA30, pGRl12, pSA32, pMCl, pMC2 and pMC3
are small amplifiable plasmids carrying the nifDNA restriction fragments indicated

for which detectable levels of mRNA were present under the conditions used
are transcribed from the same DNA strand as the his operon (JANSSEN et al.
1980).

7.3 Nif Gene Products

K. pneumoniae nitrogenase is composed of two redox proteins (see Sect. 2).

The two identical sub-units of Kp 2 are specified by nifH. The two non-identical
subunits of Kp 1 which have molecular weights of 56,000 and 60,000 are the
products of nifD and nifK respectively. The low molecular weight co-factor
(FeMoco) which can be extracted from Kp 1, partly restores activity to nitrogen-
ase in cell extracts of nifB, nifE and nifN mutants (ROBERTS et al. 1978). Polypep-
tides with molecular weights of 46,000 and 50,000 have been assigned to nifE
and nifN (ROBERTS et al. 1978). Nitrogenase activity, which is impaired in nifM
mutants, is partially restored in cell extracts of these mutants by the addition
of purified Kp 2.
The products of nifJ (120,000 m.w.) and nifF (17,000 m.w.) are thought to
be involved in the electron transport system for nitrogenase. The low levels
274 H. BOTHE et al.:

Table 6. Nif gene product

Gene Polypeptide Function

Molecular weight

nifJ 120,000 Electron input to nitrogenase

nifH 35,000 Kp2 structural gene
nifD 56,000 Kpl (X subunit
nifK 60,000 Kp1 p subunit
nifE 46,000 Synthesis or processing of FeMoCo
nifN 50,000 Synthesis or processing of FeMoCo
nifU . Unknown Unknown
nifS Unknown Unknown
nifV Unknown Unknown
nifM Unknown Kp2 processing
nifF 17,000 Electron transfer
nifL 45,000 Regulation
nifA 57,000 Activation of niftranscription
nifE Unknown Synthesis or processing of FeMoCo
nifQ Unknown Unknown

of nitrogenase activity in vivo of nifJ and nifF mutants are significantly enhanced
in cell extracts which are assayed in vitro using sodium dithionite as an electron
donor for nitrogenase. Pyruvate can also act as an electron donor to nitrogenase
in extracts of wild-type but not of nifF and nifJ mutants. Pyruvate-supported
activity can be restored in extracts of nifF but not nifJ mutants by the addition
of A. chroococcum flavodoxin which suggests that the nifFproduct is an electron-
transport protein (HILL and KAVANAGH 1980).
The biochemical phenotype of nifA and nifL mutants suggests that the
products of these genes have regulatory functions. The nifA product is required
for the derepression of the nif genes since all nif-specific proteins so far identified
are absent in nifA mutants. Complementation studies suggest that the nifA
product acts as a positive activator of nif transcription. A specific regulatory
function has not yet been assigned to the nifL product mainly because the
phenotype of the nifL mutations characterized so far is complicated by their
polarity on nifA. Polypeptides of 57,000 and 45,000 molecular weight have
been assigned to nifA and nifL respectively (F.C. CANNON unpublished).
The products of the remaining genes, nifQ, nijU, nifS and nifV have not
yet been identified and their functions are unknown. Genetic and biochemical
characterization of nifQ, nijU and nifV mutants are complicated by their high
levels of nitrogenase activity compared to other mutants. The molecular weights
and probable functions of most of the nif gene products are listed in Table 6.

7.4 Cloning of K. pneumoniae nif Genes

The source of DNA for the primary cloning of nif genes was pRD1, a plasmid
with a molecular weight of 10 8 . This phase of cloning involved the following
steps (CANNON et al. 1977, 1979); (1) Digestion of pRD1 and cloning vector
ILl Physiology, Biochemistry and Genetics of Dinitrogen Fixation 275

DNAs with restriction endonucleases. (2) Ligation of the resulting fragments

with T4 polynucleotide ligase. (3) Transformation of His- Nlf derivatives of
K. pneumoniae with the resulting mixture of recombinant molecules. (4) Selection
of transformants carrying genetic determinants on the cloning vectors followed
by screening for a Ntr+ phenotype. In some cases, transformants carrying a
genetic determinant (his+) on the "passenger" DNA were selected and subse-
quently screened for Ntr+. This procedure was used to construct a group of
plasmids (pCRA37, pCM1, pSA30 and pMC1) which carry overlapping frag-
ments of nifDNA and which collectively carry the complete cluster of K. pneu-
moniae nif genes (Fig. 3). A small amplifiable plasmid (M.W. 18.5 x 10 6 ) carry-
ing the complete cluster of nif genes was constructed by combining the nif
DNA components of pCMl and pMC2 on the cloning vector, pACYCl77
(Fig. 3 and unpublished results of M.C. CANNON and F.C. CANNON). The entire
nif region has also been cloned by inserting two HindlII restriction fragments
(Fig. 3) from pRDl into a cloning vector (PUHLER et al. 1979).

7.5 A Physical Map of nif Genes

The sites at which several restriction endonucleases cleave the nif region of
the K. pneumoniae chromosome were mapped using plasmids carrying cloned
DNA fragments (Fig. 3). It was then possible to determine the distribution
of nif genes on DNA restriction fragments using genetic and molecular tech-
niques (CANNON et al. 1979). The precise locations (± 250 basepairs) of nifinser-
tion mutations were subsequently obtained by determining the molecular dis-
tances between the points of insertion and restriction endonuclease cleavage
sites which had previously been mapped on the DNA of the his-nif region
(RIEDEL et al. 1979).
The locations of 43 n!l: : Tn5, 9 n!f: : Tnl 0 and 17 nif:: M u mutations are
shown in Fig. 3. The map of nif genes derived from these results and the order
of the mutations examined are the same as those obtained from genetic fine
structure mapping. The minimum size of the nif gene cluster derived from these
results is 23 kilobases which is in good agreement with that obtained from
Pl transduction studies (20 kilobases) by KENNEDY (1977). The maximum sepa-
ration between insertions in two adjacent genes (nifK-E) is 1.6 kilo bases - this
and smaller gaps between other genes could accommodate small nif genes which
have not yet been identified. The boundaries between nifL and F, nifN and
E, and nifD and H were accurately determined, because in these cases, insertions
in the two adjacent genes mapped within 250 base pairs of each other. From
13 insertions mapped in nifJ, an expected size of 124,000 (m.w.) was determined
for the gene product which is in good agreement with the size of the nifJ
product (120,000 m.w.) obtained from SDS polyacrylamide gel electrophoresis
(MERRICK et al. 1978, ELMERICH et al. 1978, ROBERTS et al. 1978). However,
useful estimates of the expected size of gene products could not be determined
for other nif genes because the number of insertions mapped were insufficient.
KLIPP and PUHLER (personal communication) have recently developed a proce-
dure by which the size of nif polypeptides and their DNA coding regions can
be accurately determined.
276 H. BOTHE et al.:

It has been possible from our knowledge of the physical locations of nif
genes to clone small fragments of nifDNA coding for nifpromoters, or individ-
ual or groups of genes such as the seven nif operons (unpublished results of
M.C. CANNON and F.C. CANNON). These plasmids are suitable for, (1) transcrip-
tion studies in vivo and in vitro with purified nif promoters, (2) sequencing
of nifDNA, and (3) identification of nifpolypeptides and mapping their DNA
coding regions.

7.6 Interspecies Homology of Nitrogenase Genes

Biochemical studies of nitrogenase from different prokaryotes has suggested

a high probable degree of sequence homology between their corresponding
DNA's (EADY 1977). If the DNA sequences which encode nitrogenase in these
organisms were conserved they should therefore hybridize.
NUTI et al. (1979) reported hybridization between restriction fragments of
plasmid DNA from Rhizobium leguminosarum and DNA encoding part of the
structural genes for K. pneumoniae nitrogenase. These results provide evidence
that at least some of the Rhizobium nif genes are carried on plasmids indigenous
to the strains examined. MAZUR et al. (1980) used cloned fragments of K. pneu-
moniae nif genes in similar hybridization studies to identify and clone Anabaena
7120 nif genes. Their results suggest that the order of nif genes is different
in these two organisms. RUVKUN and AUSUBEL (1980) found that part of the
structural genes for K. pneumoniae nitrogenase hybridized to total DNA from
19 out of 21 divergent species of nitrogen-fixing bacteria but not to DNA from
12 different non-nitrogen-fixing species. These results suggest that some nitro-
genase genes have either been highly conserved or exchanged between different
nitrogen-fixing species in their recent evolutionary history.
These hybridization results demonstrate that a combined molecular and gen-
etical analysis of nif genes in Ntr bacteria on which little or no nif genetic
studies have been done is now possible. Probes of Klebsiella nifDNA-encoding
homologous parts of the structural genes for nitrogenase could be used to clone
nif genes from other microbial genera, and all the nif mutants of K. pneumoniae
which have been characterized in the past decade are potential hosts for examin-
ing such cloned material.
Acknowledgements. The authors are grateful for support from NATO research grant 856
and from the Deutsche Forschungsgemeinschaft (to H.B.).

References
Akkermans ADL, Abdulkadir S, Trinick MJ (1978) N 2 -fixing root nodules in Ulmaceae:
Parasponia or (and) Trema? Plant Soil 49:711-715
Albrecht SL, Maier RJ, Hanus FJ, Russell SA, Emerich DW, Evans HJ (1979) Hydrogen-
ase in Rhizobium japonicum increases nitrogen fixation by nodulated soybeans. Science
203: 1255-1257
Appleby CA, Bergersen FJ, Macnicol PK, Turner GL, Wittenberg BA, Wittenberg JB
(1976) Role of leghaemoglobin in symbiotic 'N 2 fixation. In: Newton WE, Nyman
II.1 Physiology, Biochemistry and Genetics of Dinitrogen Fixation 277

CJ (eds) Proc 1st Int Symp Nitrogen Fixation, Vol I. Washington State Univ Press,
Pullman, pp. 274-292
Ayanaba A, Dart PJ (eds) (1977) Biological nitrogen fixation in farming systems of
the tropics. Wiley and Sons, New York, pp. 1-377
Belkin S, Padan E (1978) Hydrogen metabolism in the facultative anoxygenic cyanobac-
teria (blue-green algae) Oscillatoria limnetica and Aphanothece halophytica. FEBS
Lett 94:291-294
Benemann JR, Yoch DC, Valentine RC, Arnon DI (1971) The electron transport in
nitrogen fixation by Azotobacter III. Requirements for NADPH-supported nitrogen-
ase activity. Biochim Biophys Acta 226:205-212
Bergersen FJ, Turner GL (1973) Kinetic studies of nitrogenase from soyabean root nod-
ules. Biochem J 131: 61-75
Berndt H, Lowe DJ, Yates MG (1978) The nitrogen-fixing system of Corynebacterium
autotrophicum. Eur J Biochem 86: 133-142
Biggins DR, Kelly M (1973) Interaction of nitrogenase from Klebsiella pneumoniae with
ATP or cyanide. Biochim Biophys Acta 205: 288-299
Biggins DR, Postgate JR (1970) Nitrogen fixation by extracts of Mycobacteriumjlavum
301. Use of natural electron donors and oxygen sensitivity of cell-free preparations.
Eur J Biochem 19:408-415
Bishop PE, Brill WJ (1977) Genetic analysis of Azotobacter vinelandii mutant, strains
unable to fix nitrogen. J Bacteriol130:954-956
Bishop PE, McParland RH, Evans HJ (1975) Inhibition of the adenylylation of glutamine
synthetase by methionine sulfone during nitrogenase repression. Biochem Biophys
Res Commun 67:774-781
Bishop PE, Guevara JG, Engelke JA, Evans HJ (1976) Relation between glutamine
synthetase and nitrogenase activities in the symbiotic associations between Rhizobium
japonicum and Glycine max. Plant Physiol 57: 542-546
Bishop EO, Lambert MD, Orchard D, Smith BE (1977) Nitrogenase of Klebsiella pneu-
moniae: water proton NMR relaxation studies on the binding of divalent metal ions
and nucleotides to the iron protein. Biochim Biophys Acta 482: 286-300
Bothe H (1970) Photosynthetische Stickstoffixierung mit einem zellfreien Extrakt aus
der Blaualge Anabaena cylindrica. Ber Dtsch Bot Ges 83:421-432
Bothe H, Floener L (1978) Physiological characterization of Cyanophoraparadoxa, a
flagellate containing cyanelles in endosymbiosis. Z Naturforsch 33e:981~987
Bothe H, Yates MG (1976) The electron transport to nitrogenase in Mycobacterium
jlavum. Arch MikrobioI107:25-31
Bothe H, Falkenberg B, Nolteernsting U (1974) Properties and function of the pyruvate:
ferredoxin oxidoreductase from the blue-green alga Anabaena cylindrica. Arch Micro-
bioi 96:291-304
Bothe H, Tennigkeit J, Eisbrenner G, Yates MG (1977) The hydrogenase-nitrogenase
relationship in the blue-green alga Anabaena cylindrica. Planta 133: 237-242
Bothe H, Distler E, Eisbrenner G (1978) Hydrogen metabolism in blue-green algae.
Biochimie 60: 277-289
Bothe H, Neuer G, Kalbe I, Eisbrenner G (1980) Electron donors and hydrogenase
in nitrogen-fixing microorganisms. In: Stewart WDP, Gallon JR (eds) Nitrogen fixa-
tion. Academic Press, London New York, pp. 83-112
Burns RC, Hardy RWF (1975) Nitrogen fixation in bacteria and higher plants. Springer.
Berlin Heidelberg New York, pp. 1-189
Burris RH (1971) Fixation by free-living micro-organisms: enzymology. In: Postgate
JR (ed) The chemistry and biochemistry of nitrogen fixation. Plenum, London New
York, pp. 106-160
Burris RH (1972) Nitrogen fixation, assay methods and techniques. In: Colowick SP,
Kaplan NO (eds) Methods in enzymology, Vo124B. Academic Press, London New
York, pp. 415-431
Burris RH (1975) The acetylene reduction technique. In: Stewart WDP (ed) Nitrogen
fixation by free-living organisms. Int Bioi Programme, Vol VI. Cambridge U niv Press,
London,pp.249-257
278 H. BOTHE et al.:

Cannon FC, Dixon RA, Postgate JR (1976) Derivation and properties of F-prime factors
in Escherichia coli carrying nitrogen fixation genes from Klebsiella pneumoniae. J
Gen Microbiol 93: 111-125
Cannon FC, Riedel GE, Ausube1 FM (1977) Recombinant plasmid that carries part
of the nitrogen fixation (nif) gene cluster of Klebsiella pneumoniae. Proc Natl Acad
Sci USA 74:2963-2967
Cannon FC, Riedel GE, Ausubel FM (1979) Overlapping sequences of Klebsiella pneu-
moniae nifDNA cloned and characterised. Mol Gen Genet 174:59-66
Carnahan JE, Mortenson LE, Mower HF, Castle JE (1960) Nitrogen fixation in cell-free
extracts of Clostridium pasteurianum. Biochim Biophys Acta 44: 520-535
Carter KR, Rawlings J, Orme-Johnson WH, Becker RR, Evans HJ (1980) Purification
and characterization of a ferredoxin from Rhizobium japonicum bacteroids. J BioI
Chern 255:4213-4223
Chatt J (1980) Chemistry relevant to the biological fixation of nitrogen. In: Stewart
WDP, Gallon JR (eds) Nitrogen Fixation. Academic Press, London New York, pp.
1-18
Chatt J, Dilworth JR, Richards RL (1978) Recent advances in the chemistry of nitrogen
fixation. Chern Rev 78: 589-625
Chen JS, Blanchard DK (1978) Isolation and properties of an undirectional H2 oxidizing
hydrogenase from the strictly anaerobic N 2-fIxing bacterium Clostridium pasteurianum
W 5. Biochem Biophys Res Commun 84:1144-1150
Christou G, Garner CD, Mabbs FE (1978) Crystal structure oftris(tetra-n-butylammonium)
tri-~-benzenethiolato-bis-{ tri-~-sulphido-[~3 -sulphido-tris(benzenethiolatoiron)]-
molybdenum}[Bu~Nh[{(PhSFC3)MoS4h(SPhh]; an Fe3MoS4 cubic cluster dimer.
J Chern Soc Chern Commun 17:740-741
Cramer SP, Gillum WO, Hodgson KO, Mortenson LE, Steifel EI, Chisnell JR, Brill
WJ, Shah VK (1978) The molybdenum site of nitrogenase 2. A comparative study
of MoFe proteins and the iron-molybdenum cofactor by X-ray absorption spectros-
copy. J Am Chern Soc 100:3814-3819
Dalton H, Mortenson LE (1972) Dinitrogen (N2) fixation (with a biochemical emphasis).
Bacteriol Rev 36:231-260
Davis LC, Henzl MT, Burris RH, Orme-Johnson WH (1979) Iron-sulphur clusters in
the molybdenum-iron protein component of nitrogenase: Electron paramagnetic reso-
nance of the carbon monoxide-inhibited state. Biochemistry 18: 4860-4869
DeBont JAM, Mulder EG (1974) Nitrogen fixation and co-oxidation of ethylene by
a methane-utilizing bacterium. J Gen Microbiol 83: 113-121
Dilworth MJ (1966) Acetylene reduction by nitrogen-fixing preparations from Clostridium
pasteurianum. Biochim Biophys Acta 127:285-294
Dixon R, Kennedy C, Kondorosi A, Krishnapillai V, Merrick M (1977) Complementation
analysis of Klebsiella pneumoniae mutants defective in nitrogen fixation. Mol Gen
Genet 157: 189-198
Dixon RA, Postgate JR (1972) Genetic transfer of nitrogen fixation from Klebsiella
pneumoniae to Escherichia coli. Nature 237: 102-103
Dixon RA, Cannon FC, Kondorosi A (1976) Construction of a P plasmid carrying
nitrogen fixation genes from Klebsiella pneumoniae. Nature (London) 260:268-271
Dixon ROD (1972) Hydrogenase in legume root nodule bacteroids, occurrence and prop-
erties. Arch Mikrobiol 85: 193-201
Eady RR (1973) Nitrogenase of Klebsiella pneumoniae: Interaction of the component
proteins studied by ultracentrifugation. Biochem J 135: 531-535
Eady RR (1977) Aspects of the evolution of nitrogenase. In: Leigh GJ (ed) The evolution
of metalloenzymes, metalloproteins and related materials. Symposium Press, London,
pp 67-84
Eady RR (1980) Methods for studying nitrogenase. In: Bergersen FJ (ed) Methods for
evaluation of biological nitrogen fixation. Wiley and Sons, New York, pp. 213-264
Eady RR, Smith BE (1979) Physico-chemical properties of nitrogenase and its compo-
nents. In: Hardy RWF, Bottemeley F, Burns RC (eds) A treatise on dinitrogen fIxa-
11.1 Physiology, Biochemistry and Genetics of Dinitrogen Fixation 279

tion, Sect I, II. Inorganic and physical chemistry and biochemistry. Wiley and Sons,
New York, pp. 399-490
Eady RR, Smith BE, Cook KA, Postgate JR (1972) Nitrogenase of Klebsiella pneumoniae:
Purification and properties of the component proteins. Biochem J 128: 655--675
Eady RR, Thorneley RNF, Lowe DJ (1979) Nitrogenase of Klebsiella pneumoniae. A
pre-steady-state burst of A TP hydrolysis coupled to electron transfer between the
component proteins. FEBS Lett 95: 211-213
Eady RR, Issack R, Kennedy C, Postgate JR, Ratcliffe H (1978) Nitrogenase synthesis
in Klebsiella pneumoniae. Comparison of ammonium and oxygen regulation. J Gen
MicrobioI104:277-285
Eisbrenner G, Bothe H (1979) Modes of electron transfer from molecular hydrogen
in Anabaena cylindrica. Arch Microbiol123: 37-45
Elmerich C, Houmard J, Sibold L, Manheimer I, Charpin N (1978) Genetic and biochemi-
cal analysis of mutants induced by bacteriophage Mu DNA integration into Klebsiella
pneumoniae nitrogen fixation genes. Mol Gen Genet 165: 181-189
Ernst A, Kerfin W, Spiller H, Boger P (1979) External factors influencing light-induced
hydrogen evolution by the blue-green alga, Nostoc muscorum. Z Naturforsch 34c: 820-
825
Evans HJ, Phillips DA (1975) Reductants for nitrogenase and relationship to cellular
electron transport. In: Stewart WDP (ed) Nitrogen fixation by free-living microor-
ganisms. Int BioI Programme, vol VI. Cambridge Univ Press, Cambridge, pp 389-420
Fay P, Lang NJ (1971) The heterocysts of blue-green algae. I. Ultrastructural integrity
after isolation. Proc R Soc London Ser B 178: 185--192
Gaffron H (1940) Carbon dioxide reduction with molecular hydrogen in green algae.
Am J Bot 27:273-283
Gallon JR (1980) Nitrogen fixation by photoautotrophs. In: Stewart WDP, Gallon JR
(eds) Nitrogen fixation. Academic Press, London New York, pp. 197-238
Gordon JK, Brill WJ (1972) Mutants that produce nitrogenase in the presence of ammo-
nia. Proc Nat! Acad Sci USA 69:3501-3503
Gordon JK, Brill WJ (1974) Derepression of nitrogenase synthesis in the presence of
excess NH: . Biochem Biophys Res Commun 59: 967-971
Gotto JW, Tabita FR, van Baalen Ch (1979) Mutants of Anabaena strain CA altered
in their ability to grow under nitrogen-fixing conditions. J Bacteriol 140: 327-332
Green M, Wilson PW (1953) Hydrogenase and nitrogenase in Azotobacter. J Bacteriol
65:511-517
Haaker H, de Kok A, Veeger C (1974) Regulation of dinitrogen fixation in intact Azoto-
bacter vinelandii. Biochim Biophys Acta 357: 344-357
Hageman RV, Burris RH (1978 a) Nitrogenase and nitrogenase reductase associate and
dissociate with each catalytic cycle. Proc Nat! Acad Sci USA 75:2699-2702
Hageman RV, Burris RH (1978b) Kinetic studies on electron transfer and interaction
between nitrogenase components from Azotobacter vinelandii. Biochemistry
17 :4117-4124
Hallenbeck PC, Benemann JR (1978) Characterization and partial purification of the
reversible hydrogenase of Anabaena cylindrica. FEBS Lett 94:261-264
Hallenbeck PC, Kostel PJ, Benemann JR (1979) Purification and properties of nitrogenase
from the cyanobacterium Anabaena cylindrica. Eur J Biochem 98:275--284
Hardy RWF (1977-1979) A treatise on nitrogen fixation Vol I-IV. Academic Press,
London New York
Hardy RWF, Burns RC (1968) Nitrogen fixation. Annu Rev Biochem 37:331-358
Hardy RWF, Holsten RD (1977) Methods for measurement of dinitrogen fixation. In:
Hardy RWF, Gibson AH (eds) A treatise on dinitrogen fixation, vol IV. Wiley and
Sons, New York, pp. 451-486
Hardy RWF, Holsten RD, Jackson EK, Burns RC (1968) The acetylene-ethylene assay
for N 2-fixation: laboratory and field evaluation. Plant Physiol 43: 1185--1207
Hardy RWF, Burns RC, Holsten RD (1973) Applications of the acetylene-ethylene assay
for measurements of nitrogen fixation. Soil BioI Biochem 5:47-81
280 H. BOTHE et al.:

Hardy RWF, Havelka UD, Quebedeaux B (1978) Increasing crop productivity: the prob-
lem strategies, approach and selected rate-limitations related to photosynthesis. In:
Hall DO, Coombs J, Goodwin TW (eds) Proc 4th Int Congr Photosynthesis. Biochem
Soc, London, pp. 695-720
Haury JF, Wolk CP (1978) Classes of Anabaena variabilis mutants with oxygen-sensitive
nitrogenase activity. J Bacteriol136: 688-692
Hill S (1971) Influence of oxygen concentration on the colony type of Derxia gummosa
grown in nitrogen-free media. J Gen Microbiol 67: 77-83
Hill S (1976) Influence of atmospheric oxygen concentration on acetylene reduction and
efficiency in intact Klebsiella pneumoniae. J Gen Microbiol 93: 335-345
Hill S, Kavanagh E (1980) Some biochemical properties of nijF and nifJmutants relevant
to electron transport to nitrogenase in Klebsiella pneumoniae. J Bacteriol141 : 470-475
Hollaender A, Burris RH, Day PR, Hardy RWF, Helinski DR, Lamborg MR, Owens
L, Valentine RC (eds) (1977) Genetic engineering for nitrogen fixation. Plenum, New
York London
Huyhn BH, Munck E, Orme-Johnson WH (1979) Nitrogenase XI: Mossbauer studies
on the cofactor centers of the MoFe protein from Azotobacter vinelandii OP. Biochim
Biophys Acta 527: 192-203
Janssen KA, Riedel GE, Ausubel FM, Cannon FC (1980) Transcriptional studies with
cloned nitrogen-fixation genes. In: Orme-Johnson WH, Newton WE (eds) Nitrogen
fixation, vol I. Univ Park Press, Baltimore, pp. 85-93 .
Jeng D, Morris JA, Mortenson LE (1970) The effect of reductant in inorganic phosphate
release from adenosine-5-triphosphate from purified nitrogenase of Clostridium pas-
teurianum. J BioI Chern 245:2809-2813
Kennedy C (1977) Linkage map of nitrogen fixation (nif) genes in Klebsiella pneumoniae.
Mol Gen Genet 157: 199-204
Knight E Jr, D'Eustachio AJ, Hardy RWF (1966) A flavoprotein with ferredoxin activity
from Clostridium pasteurianum. Biochim Biophys Acta 113: 626-628
Kurtz DM, McMillan RS, Burgess BK, Mortenson LE, Holm R (1979) Identification
of iron-sulfur centers in the iron-molybdenum proteins of nitrogenase. Proc Natl
Acad Sci USA 76:4986-4989
Laane C, Haaker H, Veeger C (1978) Involvement of the cytoplasmatic membrane in
nitrogen fixation by Rhizobium leguminosarum bacteroids. Eur J Biochem 87: 147-153
Laane C, Haaker H, Veeger C (1979a) On the efficiency of oxidative phosphorylation
in membrane vesicles of Azotobacter vinelandii and of Rhizobium leguminosarum bac-
teroids. Eur J Biochem 97: 369-377
Laane C, Krone W, Konings WN, Haaker H, Veeger C (1979b) The involvement of
membrane potential in nitrogen fixation by bacteroids of Rhizobium leguminosarum.
FEBS Lett 103:328-332
Lambein F, Wolk CP (1973) Structural studies on the glycolipids from the envelope
of the heterocysts from Anabaena cylindrica. Biochemistry 12: 791-798
Leach CK, Carr NG (1971) Pyruvate: ferredoxin oxidoreductase and its activation by
ADP in the blue-green alga Anabaena variabilis. Biochim Biophys Acta 245: 165-
174
Lex M, Stewart WDP (1973) Algal nitrogenase, reductant pools and photosystem I activi-
ty. Biochim Biophys Acta 292:436-443
Lowe DJ (1979) Simulation of the electron-paramagnetic resonance spectrum of the
iron-protein of nitrogenase: A prediction of the existence of a second paramagnetic
centre. Biochem J 175:955-957
Lowe DJ, Eady RR, Thorneley RNF (1978) Electron paramagnetic resonance on nitro-
genase of Klebsiella pneumoniae. Evidence for acetylene and ethylene-nitrogenase tran-
sient complexes. Biochem J 173: 277-290
Ludden PW, Burris RH (1976) Activating factor for the iron protein of nitrogenase
from Rhodospirillum rubrum. Science 194:424-426
Ludden PW, Burris RH (1978) Purification and properties of nitrogenase from Rhodospir-
ilium rubrum and evidence for phosphate, ribose and an adenine-like unit covalently
bound to the iron protein. Biochem J 175:251-259
II.1 Physiology, Biochemistry and Genetics of Dinitrogen Fixation 281

Lundell DJ, Howard JB (1978) Isolation and partial characterisation of two different
subunits from the molybdenum iron protein of Azotobacter vinelandii nitrogenase.
J BioI Chern 253: 3422-3426
MacNeil T, MacNeil D, Roberts GP, Supiano MA, Brill WJ (1978) Fine-structure
mapping and complementation analysis of nif (nitrogen ftxation) genes in Klebsiella
pneumoniae. J. Bacteriol136:253--266
Maier RJ, Brill WJ (1976) Ineffective and non-nodulating mutant strains of Rhizobium
japonicum. J. Bacteriol127: 863--869
Marrs B, Wall JD, Gest H (1977) Emergence of the biochemical genetics and molecular
biology of photosynthetic bacteria. Trends Biochem Sci 2:105-108
Mayhew SG, Ludwig ML (1975) Flavodoxins. In: Boyer PO (ed) The Enzymes. Academic
Press, London New York, pp. 57-118
Mazur BI, Rice D, Haselkorn R (1980) Identiftcation of blue-green algal ftxation genes
by using heterologous DNA hybridisation probes. Proc Natl Acad Sci USA
77:186-190
McKenna CE, Huang CW (1979) In vivo reduction of cyclopropene by Azotobacter
vinelandii nitrogenase. Nature 280: 609-610
Meers JL, Tempest DW, Brown CM (1970) "Glutamine (amide) 2-oxyglutarate amino
transferase oxidoreductase (NADP)", an enzyme involved in the synthesis of gluta-
mate by some bacteria. J Gen Micro bioI 64: 187-194
Merrick M, Filser M, Kennedy C, Dixon R (1978) Polarity of mutations induced by
insertion of transposons Tn5, Tn7 and Tn10 into the nif gene cluster of Klebsiella
pneumoniae. Mol Gen Genet 165:103--111
Merrick M, Pilser M, Dixon R, Elmerich C, Sibold L, Houmard J (1980) The use
of translocatable genetic elements to construct a ftne-structure map of the Klebsiella
pneumoniae nitrogen ftxation (nij) gene cluster. J Gen Microbiol117:509-520
Meyer J, Kelley BC, Vignais PM (1978) Nitrogen ftxation and hydrogen metabolism
in photosynthetic bacteria. Biochimie 60: 245-260
Millbank JW (1969) Nitrogen ftxation in moulds and yeasts - a reappraisal. Arch Mikro-
bioI 68: 32-39
Mishustin EN, Shil'nikova VK (1971) Biological ftxation of atmospheric nitrogen. Mac-
Millan, London, pp. 1-420
Mortenson LE, Thorneley RNF (1979) Structure and function of nitrogenase. Annu
Rev Biochem 48:387-418
Mortenson LE, Valentine RC, Carnahan JE (1962) An electron transport factor from
Clostridium pasteurianum. Biochem Biophys Res Commun 7: 448-452
Munck E, Rhodes H, Orme-Johnson WH, Davies LC, Brill WJ, Shah VK (1975) Nitro-
genase VIII. Mossbauer and e.p.r. spectroscopy. The MoFe protein component from
Azotobacter vinelandii OP. Biochim Biophys Acta 400: 32-53
Newton WE, Bulen WA, Hadfteld KL, Stiefel EI, Watt GD (1977) HD formation as
a probe for intermediate in N2 reduction. In: Newton WE, Postgate JR, Rodriguez-
Barrueco C (eds) Recent advances in nitrogen ftxation. Academic Press, London
New York, pp. 119-130
Nordlund S, Eriksson U, Baltscheffsky H (1977) Necessity of a membrane component
for nitrogenase activity in Rhodospirillum rubrum. Biochim Biophys Acta 462: 187-195
Nuti MP, Lepidi AA, Prakash RK, Schilperoort RA, Cannon FC (1979) Evidence for
nitrogen ftxation (nif) genes on indigenous Rhizobium plasmids. Nature 282:533--535
O'Donnell M, Smith BE (1978) Electron-paramagnetic-resonance studies on the redox
properties of the molybdenum-iron protein of nitrogenase between + 50 and
-450mV. BiochemJ 173:831-839
Okon S, Houchins JP, Albrecht SL, Burris RH (1977) Growth of Spirillum /ipoferum
at constant partial pressures of oxygen, and the properties of its nitrogenase in cell-free
extracts. J Gen Microbiol 98: 87-93
Orme-Johnson WH, Davis LC, Henzl MT, Averill BA, Orme-Johnson NR, Munck E,
Zimmerman R (1977) Components and pathways in biological nitrogen ftxation. In:
Newton WE, Postgate JR, Rodriguez-Barrueco C (eds) Recent advances in nitrogen
ftxation. Academic Press, London New York, pp. 131.:..178
282 H. BOTHE et al. :

Pedrosa FO, Dobereiner J, Yates MG (1980) Hz-dependent growth and autotrophic

COz-fixation by Derxia. J Gen Microbiol119:547-551
Peterson RB, Burris RH (1976) Conversion of acetylene reduction rates to nitrogen
fixation rates in natural populations of blue-green algae. Anal Biochem 73 :404-410
Phelps AS, Wilson PW (1941) Occurrence of hydrogenase in nitrogen-fixing organisms.
Proc Soc Exp Bioi 47:473-476
Postgate JR (ed) (1971) The chemistry and biochemistry of nitrogen fixation. Plenum
Press, New York London, p. 1-326
Postgate JR (1974) Evolution within nitrogen-fixing systems. In: Carlile MJ, Skehel JJ
(eds) Evolution within the microbial world. Cambridge Univ Press, Cambridge, pp.
263-292
Postgate JR (1978) Nitrogen fixation. Stud BioI No 92. Arnold, London, pp 1-67
Postgate JR (1982) Biological nitrogen fixation: fundamentals. Phil Trans R Soc Lon
B 296:375-385
Postgate JR, Krishnapillai V (1977) Expression of Klebsiella niland his genes in Salmonel-
la typhimurium. J Gen Microbiol 98: 379-385
Postgate JR, Eady RR, Dixon RA, Hill S, Kahn D, Kennedy C, Partridge P, Robson
R, Yates MG (1981) Some aspects of the physiology of dinitrogen fixation. In: Bothe
H, Trebst A (eds) Biology of inorganic nitrogen and sulfur. Springer, Berlin Heidelberg
New York, p. 103-115
Piih1er A, Burkardt HJ, Klipp W (1979) Cloning in Escherichia coli the genomic region
of Klebsiella pneumoniae which encodes genes responsible for nitrogen fixation. In:
Timmis KN, Piihler A (eds) Plasmids of medical, environmental and commercial
importance. Developments in genetics, vol I. Elsevier/North-Holland, Amsterdam,
pp.435-447
Quispel A (1974) The biology of nitrogen fixation. North-Holland Res Monogr Frontiers
BioI, Amsterdam Oxford, pp. 1-769
Rawlings J, Shah VK, Chisnell JR, Brill WJ, Zimmerman R, Munck E, Orme-Johnson
WH (1978) Novel metal clusters in the iron-molybdenum cofactor of nitrogenase.
J BioI Chern 253:1001-1004
Rennie RJ, Funnell A, Smith BE (1978) Immunochemistry of nitrogenase as a probe
for the enzyme mechanism: Evidence for multiple enzyme forms and an MgATP
binding site on the MoFe protein. FEBS Lett 91: 158-161
Riedel GE, Ausubel FM, Cannon FC (1979) Physical map of chromosomal nitrogen
fixation (nif) genes of Klebsiella pneumoniae. Proc Nat! Acad Sci USA 76:2866-
2879
Rippka R, Waterbury JB (1977) The synthesis of nitrogenase by non-heterocystous cyano-
bacteria. FEMS Lett 2: 83-86
Rippka R, Deruelles J, Waterbury JB, Herdman M, Stanier RY (1979) Generic assign-
ments, strain history and properties of pure cultures of cyanobacteria. J Gen Microbiol
111: 1-61
Rivera-Ortiz JM, Burris RH (1975) Interactions among substrates and inhibitors of nitro-
genase. J Bacteriol 123: 537- 545
Roberts GP, MacNeil T, MacNeil D, Brill WJ (1978) Regulation and characterisation
of protein products coded by the nif(nitrogen fixation) genes of Klebsiella pneumoniae.
J Bacteriol 136: 267-279
Robson RL (1979) Oz-repression of nitrogenase synthesis in Azotobacter chroococcum.
FEMS Microbiol Lett 5: 259-262
Robson RL, Postgate JR (1980) Oxygen and hydrogen in biological nitrogen fixation.
Annu Rev Microbiol 34: 183-207
Ruvkun GB, Ausubel FM (1980) Interspecies homology of nitrogenase genes. Proc Nat!
Acad Sci USA 77:191-195
Scherings G, Haaker H, Veeger C (1977) Regulation of nitrogen fixation by Fe-S protein
II in A. vinelandii. Eur J Biochem 77: 621-630
Schlegel HG, Schneider K (1978) Hydrogenases: Their catalytic activity, structure and
function. Goltze, Gottingen, pp. 1-453
11.1 Physiology, Biochemistry and Genetics of Dinitrogen Fixation 283

Schollhorn R, Burris RH (1967) Acetylene as a competitive inhibitor of N 2 -fixation.

Proc Nat! Acad Sci USA 58:213-216
Shah VK, Brill WH (1977) Isolation of an iron-molybdenum cofactor from mtrogenase.
Proc Nat! Acad Sci USA 74: 3249-3253
Shah VK, Davis LC, Gordon JK, Orme-Johnson WH, Brill WJ (1973) Nitrogenaseless
mutants of Azotobacter vinelandii: Activities, cross-reactions and EPR spectra.
Biochim Biophys Acta 292: 246-255
Shanmugan KT, Buchanan BB, Arnon DI (1972) Ferredoxins in light- and dark-grown
photosynthetic cells with special reference to Rhodospirillum rubrum. Biochim Biophys
Acta 256:477-486
Shethna YI (1970) Non-heme (iron-sulfur) proteins of Azotobacter vinelandii. Biochim
Biophys Acta 205: 58-62
Siefert E (1976) Die Fixierung von molekularem Stickstoff bei phototrophen Bakterien
am Beispiel von Rhodopseudomonas acidophila. Thesis, Univ Gottingen, pp. 1-149
Silver WS, Postgate JR (1973) Evolution of asymbiotic nitrogen fixation. J Theor BioI
40:1-10
Simpson FB, Maier RJ, Evans HJ (1979) Hydrogen-stimulated CO 2 -fixation and coordi-
nate induction of hydrogenase and ribulosebisphosphate carboxylase in a Hruptake
positive strain of Rhizobiumjaponicum. Arch MicrobioI123:1-8
Singh PK (1973) Nitrogen fixation by the cellular blue-green alga Aphanothece. Arch
Microbiol 92: 59-62
Smith BE (1980) Studies on the iron molybdenum cofactor from the nitrogenase MoFe
protein of Klebsiella pneumoniae. In: Newton WE, Otsuka S (eds) Molybdenum
chemistry of biological significance. Plenum, New York London, pp. 179-190
Smith BE, Lang G (1974) Mossbauer spectroscopy of nitrogenase components from
Klebsiella pneumoniae. Structural assignments and mechanistic conclusions. Biochem
J 137:169-180
Smith BE, Eady RR, Thorneley RNF, Mortenson LE (1976) Nitrogenases from Klebsiella
pneumoniae and Clostridium pasteurianum. Kinetic investigations of cross reactions
as a probe of the enzyme mechanism. Biochem J 157:439-447
Smith LA, Hill S, Yates MG (1976) Inhibition by acetylene of conventional hydrogenase
in nitrogen-fixing bacteria. Nature (London) 262:209-210
SOderlund R, Svensson BH (1976) The global nitrogen cycle. In: Svensson BH, Soderlund
R (eds) Nitrogen, phosphorus and sulfur-Global cycles. Scope Rep 7. Ecol Bull (Stock-
holm) 22:23-74
Southern EM (1975) Detection of specific sequences among DNA fragments separated
by gel electrophoresis. J Mol BioI 98: 503-517
Sprent JI (1979) The biology of nitrogen-fixing organisms. McGraw-Hill, Maidenhead,
pp.1-196
Stanier RY, Cohen-Bazire G (1977) Phototrophic procaryotes: the cyanobacteria. Annu
Rev Microbiol31 :225-274
Stephens PJ, McKenna CE, Smith BE, Nguyen HT, McKenna MC, Thomson AJ, Devlin
F, Jones JB (1979) Circular dichroism and magnetic circular dichroism of nitrogenase
proteins. Proc Nat! Acad Sci USA 76:2585-2589
Stewart WDP, Lex M (1970) Nitrogenase activity in the blue-green alga Plectonema
boryanum strain 594. Arch Microbiol 73: 250-260
Stewart WDP, Singh HN (1975) Transfer of nitrogen-fixing nif genes in the blue-green
alga Nostoc muscorum. Biochem Biophys Res Commun 62: 62-69
Streicher SL, Gurney EG, Valentine RC (1971) Transduction of nitrogen fixation genes
in Klebsiellapneumoniae. Proc Nat! Acad Sci USA 50:1174-1177
Streicher SL, Shanmugam KT, Ausubel FM, Morandi C, Goldberg RB (1974) Regulation
of nitrogen fixation in Klebsiella pneumoniae: evidence for a role of glutamine synthe-
tase as a regulator of nitrogenase synthesis. J Bacteriol120: 815-821
Sweeney WV, Rabinowitz JC, Joch DC (1975) High and low reduction potential4Fe-
4S* clusters in Azotobacter vinelandii. J BioI Chern 250:7842-7847
Tanaka M, Mitsura H, Yasunobu KT, Mortenson LE (1978) The amino acid sequence
284 H. BOTHE et al. :

of Clostridium pasteurianum iron protein, a component of nitrogenase. J BioI Chem

252:7081-7100
Tel-Or E, Luijk LW, Packer L (1977) An inducible hydrogenase in cyanobacteria en-
hances N 2 -fIxation. FEBS Lett 78:49--51
Thauer RK, Rupprecht E, Jungermann K, Decker K (1969) Hydrogen formation from
NADH in cell-free extracts of Clostridium kluyveri. FEBS Lett 4: 108-112
Thomeley RNF (1975) Nitrogenase of Klebsiella pneumoniae. A stopped-flow study of
magnesium-adenosine triphosphate-induced electron transfer between the component
proteins. Biochem J 145: 391-396
Thomeley RNF, Cornish-Bowden A (1977) Kinetics of nitrogenase from Klebsiella pneu-
moniae. Heterotrophic interactions between magnesium adenosine-5-diphosphate and
magnesium-adenosine-5' -triphosphate. Biochem J 165: 255-262
Thomeley RNF, Eady RR, Lowe DJ (1978) Biological nitrogen fIxation by way of
an enzyme-bound dinitrogen-hydride intermediate. Nature 272: 557-558
Trinick MJ (1973) Symbiosis between Rhizobium and the non-legume, Trema aspera.
Nature 244:459--460
Tso M-Y, Burris RH (1973) The binding of ATP and ADP by nitrogenase components
from Clostridium pasteurianum. Biochim Biophys Acta 309:263-270
Tubb RS (1974) Glutamine synthetase and ammonia regulation of nitrogenase synthesis
in Klebsiella. Nature 251 :481-485
Tubb RS, Postgate JR (1973) Control of nitrogenase synthesis in Klebiella pneumoniae.
J Gen Micro bioI 79: 103-117
Tyler B (1978) Regulation of the assimilation of nitrogen compounds. Annu Rev Biochem
47:1127-1162
Uyeda K, Rabinowitz JC (1971) Pyruvate-ferredoxin oxidoreductase: studies on the reac-
tion mechanism. J BioI Chem 246:3120-3125
Vincent JM (1974) Root-nodule symbiosis with Rhizobium. In: Quispel A (ed) The biology
of nitrogen fIxation. North-Holland Res Monogr Frontiers BioI Vol 33. Amsterdam
Oxford
Walker CC, Yates MG (1978a) H2 evolution and acetylene effects in N 2-fIxing Azoto-
bacter chroococcum under different nutrient limitations. Proc Steenbock-Kettering
Int Symp Nitrogen Fixation, Madison, Wisconsin, Abstract A-2
Walker CC, Yates MG (1978b) The hydrogen cycle in nitrogen-fIxing Azotobacter chroo-
coccum. Biochimie 60:225-231
Wall JD, Weaver PF, Gest H (1975) Genetic transfer of nitrogenase-hydrogenase activity
in Rhodopseudomonas capsulata. Nature 258: 630-631
Weissman JC, Benemann JR (1977) Hydrogen production by nitrogen-starved cultures
of Anabaena cylindrica. Appl Environ Microbiol 33: 123-131
Wilcockson J (1977) The apparent influence of atmospheric oxygen concentration on
nitrogenase activity and slime production in Klebsiella pneumoniae grown in solid
medium. J Gen MicrobioI101:311-317
Wilson PW (1958) Asymbiotic nitrogen fIxation. In: Ruhland W (ed) Encyclopedia of
plant physiology, Vol VIII. Springer, Berlin Gottingen Heidelberg, pp. 9--47
Winter RH, Burris HC (1976) Nitrogenase. Annu Rev Biochem 45:409--426
Wyatt JT, Silvey JKG (1969) Nitrogen fIxation by Gloeocapsa. Science 165:908-909
Yates MG (1971) Electron transport to nitrogenase in Azotobacter chroococcum. Eur
J Biochem 24:347-357
Yates MG (1977) Physiological aspects of nitrogen fIxation. In: Newton WE, Postgate
JR, Rodriguez-Barrueco C (eds) Recent advances in nitrogen fIxation. Academic
Press, London New York, pp. 219--270
Yates MG (1980) Biochemistry of nitrogen fIxation. In: Miflin BJ (eds) The biochemistry
of plants, Vol V. Academic Press, London New York, pp. 1-64
Yates MG, Eady RR (1980) The physiology and regulation of nitrogen fIxation. In:
Subba-Rao NS (eds) Recent advances in nitrogen fIxation. Oxford and IBH Publ,
New Delhi, pp. 88-120
Yates MG, Jones C (1974) Respiration and nitrogen fIxation in Azotobacter. Adv Micro-
bioI Physiol11 :97-136
II.1 Physiology, Biochemistry and Genetics of Dinitrogen Fixation 285

Yates MG, Lowe DJ, Thorne1ey RNF (1975) Nitrogenase of Azotobacter chroococcum:
Inhibition by ADP of the reduction of oxidised Fe protein by sodium dithionite.
FEBS Lett 60: 89-93
Yoch DC, Carithers RT (1979) Bacterial iron-sulfur proteins. Microbiol Rev 43: 384-421
Yoch DC, Valentine RC (1972) Ferredoxin and flavodoxins of bacteria. Annu Rev Micro-
bioI 26: 139-162
Yoch DC, Carithers RP, Arnon DI (1977) Isolation and characterization of bound iron-
sulfur proteins from bacterial photosynthetic membranes. J BioI Chern 252: 7453--7460
Zumft WG (1976) The molecular basis of biological dinitrogen ftxation. Struct Bonding
29:1-66
Zumft WG, Mortenson LE (1975) The nitrogen-ftxing complex of bacteria. Biochim
Biophys Acta 416: 1-52
11.2 Dinitrogen-Fixing Symbioses with Legumes,
Non-Legume Angiosperms and Associative Symbioses
A. QUISPEL

1 Introduction

As fixation of dinitrogen, N 2, is limited to certain prokaryotic organisms, other

organisms like higher plants can only profit from N 2-fixation if they live in
close proximity to such prokaryotes. On the other hand, N 2-fixing organisms
with a heterotrophic type of metabolism will need a good provision of organic
substrates in order to obtain the energy for the highly energy-demanding process
ofN 2-fixation. Both types of organisms can therefore benefit from mutual asso-
ciations and the more intimate these associations are, the better results of their
interactions may be expected. The examples of symbiotic associations which
will be discussed in this chapter range from rather loose associations through
systems with more intimate cellular and intercellular contacts to intracellular
endosymbioses in which, in the highly developed types of root nodule symbioses,
the prokaryotic microsymbionts function more or less as N 2-fixing organelles
in a well-adapted functional organ as part of the whole plant.
In many natural ecosystems symbiotic associations with N 2-fixing organisms
play an important role in the N economy. In agriculture the leguminous nodu-
lated plants were used as food producers and green manurial systems already
in antiquity. In our century the demands for N are partially covered by fertilizers
produced by chemical-technological processes of N 2 fixation. However, the in-
creasing demands for N in food production and the increased costs of fossil
energy in the production of fertilizers recently have stimulated research on bio-
logical, and especially, on symbiotic N2 fixation. This is reflected in many recent
symposia (NUTMAN 1976, NEWTON and NYMAN 1976, NEWTON et al. 1977, Do-
BEREINER et al. 1977, NEWTON and ORME-JOHNSON 1980). The most recent sym-
posium (GIBSON and NEWTON 1981) could only partially be included in this
review. As the former edition of this encyclopedia contained an excellent and
very comprehensive survey of N 2-fixing symbioses (ALLEN and ALLEN 1958)
and recent extensive summaries are available (QUISPEL 1974, HARDY and SILVER
1977) we will restrict ourselves especially to those results of recent physiological
research which are representative for present interest and progress in this field.
The topics to be discussed will be chosen with regard to their, relevance to

Abbreviations. ALA b-amino levulinic acid; ALAD b-amino levulinic acid dehydrogenase;
ALAS b-amino levulinic acid synthetase; ATP adenosine triphosphate; DNA deoxyribo-
nucleic acid; EPS exopolysaccharide; FITC fluoresceine isothiocyanate; GDH glutamate
dehydrogenase; GOGAT glutamate synthase; GS glutamine synthetase; lAA indole
acetic acid; ICA indole carbonic acid; LPS lipopolysaccharide; 2-i-P isopentenyl adenine;
PEP phosphoenolpyruvic acid; PHBA polyhydroxybutyric acid; RNA ribonucleic acid.
11.2 Dinitrogen-Fixing Symbioses 287

our understanding of the symbiotic interactions which lead to an efficient N 2-

fixing system. The influence of environmental factors will only be discussed
in relation to these interactions, as the ecological aspects as such will be the
main topic of another chapter in this encyclopedia (GmsoN and JORDAN Chap. 8,
Vo1. 12C). For fundamental aspects of the biochemistry of the N 2-fix'ing process
we refer to BOTHE et a1. (Chap II.1, this Vo1.).

2 Description of the Main Symbiotic

Dinitrogen-Fixing Systems

2.1 Associative Symbioses

There is no need to discuss the rather loose symbiotic associations in the rhizo-
sphere and the phyllosphere of higher plants as the next chapter of this volume
by DOBERElNER (Chap. II.3) will be devoted to this subject. There are some
reasons to mention them in this chapter. The first step in the establishment
of more intimate symbioses such as root nodules of leguminous plants consists
of an accumulation of the bacteria in the rhizosphere. It is tempting to speculate
that these more complicated symbioses developed during evolution from more
associative types of symbioses. This may be indicated by the fact that even
in rhizosphere associations bacteria tend to enter the outer root cell layers
and occasionally even inner cells of the cortex and the endodermis. Some data
on N2 fixation values were already given in Table 2 in BOTHE et a1. (Chap. 11.1,
this Vo1.).

2.2 Symbioses with Cyanobacteria

2.2.1 Distribution
Symbioses with N 2-fixing Cyanobacteria (blue-green algae) are found with quite
different types of organisms: with fungi in different lichens, in liverworts like
Anthoceros, Blasia and Cavicu/aria, in Sphagnum, in the water fern Azolla and
in the Cycadeae and as only representative of the Angiospermae, Gunnera (Ha-
loragaceae). All symbiotic Cyanobacteria belong to the genera Nostoc or Ana-
baena. For a more extensive discussion we refer to MILLBANK (in QUISPEL 1974)
and reviews by STEWART in NEWTON et a1. (1977) and by PETERS et al. in NEWTON
and ORME-JOHNSON (1980).

2.2.2 Description and Development

In lichens the Cyanobacteria are either the only "phycobiont", or they are
localized in special organs, called cephalodia, in those lichens where the main
phycobiont is a green alga such as Trebouxia. Though there is a close juxtaposi-
tion of the fungal and cyanobacterial cells, no haustoria are formed. Some
288 A. QmSPEL:

authors emphasize the thick extracellular sheath which surrounds the Cyanobac-
teria. In most symbioses with green lower or higher plants the Cyanobacteria
occupy cavities which already were formed before invasion with the algae. In
many cases growth of these cavities is stimulated by the presence of the cyano-
bacteria. In Azolla the correlated growth of the cavities and the Anabaena cells
indicates the regulatory influence of some common factor. Contact between
the Cyanobacteria and the host cells may be stimulated by the induced develop-
ment of hairs (Azolla) or papillae (liverworts). Such structures act in the transfer
of substances between the symbiotic partners. Though in most cases the Cyano-
bacteria remain intercellular, in Sphagnum and especially in Gunnera they live
intracellularly after dissolving host cell walls.
In Cyanobacteria heterocysts are regarded as special sites for N 2 fixation.
We refer to the chapter by STEWART in the second photosynthesis volume
(Vol. 6) of this Encyclopedia and reviews by FOGG et al. (1973) and STEWART
(in QUISPEL 1974).
In most symbiotic Cyanobacteria the percentage of heterocysts is unusually
high. In lichens a high percentage of heterocysts is only observed in cephalodia,
which is explained by the good C-nutrition and consequently high C/N ratio
due to the photosynthesis of the green algae.

2.2.3 N 2 Fixation (C 2H 2 Reduction)

In all cases N2 fixation has been demonstrated by the 15N2 and/or the C 2H 2
reduction method.

2.3 Root Nodules with Actinomycetes: Actinorhizas

2.3.1 Distribution
Root nodules with actinomycete-like organisms as endophytes, nowadays
termed actinorhizas, were first described for Alnus glutinosa by WORONIN in
1866 and HILTNER in 1896. The importance of the nodules as sites ofN 2 fixation
was established by KREBBERS in 1932 and ROBERG in 1934. The search after
plants with actinorhizas was highly stimulated by BOND (see e.g. in QUISPEL
1974, NUTMAN 1976, NEWTON et al. 1977). The lists of plants in Table 1 will
certainly prove to be incomplete as regularly new additions are discovered.
It is evident from this table that actinorhizas are found among plants belonging
to taxonomically unrelated families. Even within one genus, species with and
without actinorhizas are known, sometimes depending on the geographical situa-
tion (e.g. the European Dryas octopetala which is the only Dryas species without
root nodules). With the only exception of Datisca all actinorhizas are found
on woody plants. Their ecological importance is considerable as many of the
plants are pioneers in natural ecosystems. Their practical importance for forestry
was emphasized by GORDON et al. (1979).
IL2 Dinitrogen-Fixing Symbioses 289

Table 1. Incidence of Nrfixing root nodules in Angiospermae. (Data based on ALLEN

and ALLEN 1958 and AKKERMANS in GORDON et al. 1979)
Micro- (Sub )family Approximate Reported number Nodulated
symbiont number of of nodulated genera
genera species genera species

Frankia Casuarinaceae Casuarina

Coriariaceae Coriaria
Betulaceae 2 Alnus
Datiscaceae 2 1 2 Datisca
Myricaceae 2 2 Myrica
Rosaceae 124 4 Dryas
Purshia
Cercocarpus
Rubus
Elaeagnaceae 3 3 Elaeagnus
Hippophae
Shepherdia
Rhamnaceae 58 3 Ceanothus
Discaria
Colletia
Rhizobium Ulmaceae 16 230 1 2 Parasponia
Leguminosae:
Mimosoidae 1,500 127
Caesalpinoidae 1,300 27
Papilionatae 10,000 952

2.3.2 Description
Actinorhizas are typical metamorphosized roots with an apical meristem and
a centre vascular bundle, surrounded by an endodermis. The enlarged cortex
contains the cells with the endophyte. Immediately behind the meristem the
newly formed cortex cells are penetrated by hyphae growing from the already
infected cells. More to the base senescence and disintegration occurs. The mer-
istems branch in a dichotomous way, giving rise to a system of lobes with
a clustered or coralloid form called rhizothamnia. In some plants like Myrica,
Comptonia and Casuarina the apical meristem may give rise to normal, though
negatively geotropic, roots. In the latter plant the repeated initiation of lateral
roots leads to the formation of a complex truncate system. We refer to SCHAEDE
(1962) for a survey of the older literature on nodule cytology. SCHAEDE'S light
microscopical studies of the nodules of Alnus, Hippophae and Myrica from
1933 to 1938 still form the base for all further work on actinorhizae. Studies
on ultrastructural aspects are available for Alnus glutinosa, A. crispa, Myrica
cerifera, M. pennsylvanica, Hippophae rhamnoides, Datisca cannabina, Purshia
tridentata, Comptonia peregrina, Ceanothus integerrimus (see e.g. BECKING et al.
290 A. QUISPEL:

Fig. 1. Cross-section through cells of a spore (+) rhizothamnion of Alnus glutinosa

showing cells with hyphae and vesicles and the typical sporangia formation. (photo
C. vAN DUK) Bar = 10 11m

1964, VAN DIJK and MERKUS 1976, STRAND and LAETSCH 1977, and articles
in TORREY and TJEPKEMA 1979, and GORDON et al. 1979). Notwithstanding the
taxonomic diversity of the plants studied the pictures of the endophyte-contain-
ing cells show a surprising uniformity (Fig. 1). The cells are filled with dense
clusters of hyphae which at the cell surface swell into vesicular (most species)
or club-shaped (Myricaceae and Coriaria) structures. Only in Casuarina are
such swellings absent or inconspicuous, while in Datisca they are situated in
the central part of the cells. The vesicles are characterized by incomplete septae
and inclusions with an unknown character called striated bodies. All hyphae
and vesicles are surrounded by a capsule which is continuous with the host
cell walls and can be dissolved by pectic enzymes. Most probably these capsules
are of host cell origin, as is the plasmalemma, which is present at the side
of the host-cell cytoplasm (LALONDE and DEVOE 1976).
In some cases a third type of cell is observed, originally called bacteroid,
bacteria-like cells or granulae. They are typically formed in thickenings of
hyphae which are characterized by transversal cell divisions (VAN DIJK and
MERKUS 1976). The detailed descriptions leave no doubt that these swellings
are sporangia and the small cells formed are the spores of the endophytes.
This spore formation inside the nodules appears to be more the exception than
the rule. It has been observed up till now in only part of the population of
Alnus glutinosa in the Netherlands, in Scottish Myrica gale, but not in Myrica
gale in the Netherlands, in Dryas drummondii from Alaska and American
Purshia tridentata. Most nodules are practically devoid of spores, though they
may be never entirely absent (VAN DIJK in GORDON et al. 1979).
11.2 Dinitrogen-Fixing Symbioses 291

2.3.3 Infection and Development

Initiation and development of actinorhizal root nodules was first studied in
Alnus glutinosa by POMMER and TAUBERT in 1959 (QUISPEL 1974) and more
recently by ANGULO-CARMONA (1974). There are no observations ona develop-
ment of the Actinomycetes in the rhizosphere. Pictures of polar adhesion and
encapsulation on the root hairs (LALONDE in NEWTON et al. 1977) appear to
be not specific and are not correlated with subsequent infections. Infection
is presumed to take place in deformed root hairs. Nodule formation is a two-step
process. After penetration, cell divisions are initiated in the cortex after which
the newly formed cells are occupied by the endophytes. These swellings we
call the primary nodules. As they are already Nz-fixing small nodules the term
prenodules is less preferable. Under the influence of these primary nodules
the nearby formation of lateral roots is initiated. These root initials are infected
from the primary nodules and are transformed into the definite nodule system,
the rhizothamnia. Studies with Casuarina, Ceanothus, Comptonia and Myrica
gale confirm this general pattern of development (see TORREY and TJEPKEMA
1979).

2.3.4 N z Fixation (CzH z Reduction)

With only a few exceptions, N z fixation has been confirmed in the plants men-
tioned in Table 1 by 15N z uptake or CzH z reduction. Typical examples of
CzH z reduction range from 9-92.0 J..lm h -1 g-l fresh weight for seedlings of
Alnus glutinosa.
No differences were observed between CzH z reduction of nodules with or
without spores. This indicates that spores are not N z-fixing cells (AKKERMANS
and VAN DIJK in NUTMAN 1976).

2.4 Leguminous Root Nodules with Rhizobium

2.4.1 Distribution
Root nodules produced by an infection with bacteria belonging to the genus
Rhizobium are widespread among plants belonging to the families of the Legu-
minosae (Table 1). Their agronomical and ecological importance is considerable
(HARDY and GIBSON 1977).
The bacterial origin of the root nodules and their role in the fixation of
atmospheric dinitrogen was established by the classical work of HELLRIEGEL
and WILLFAHRT in 1888. In the same year Beyerinck isolated the responsible
bacteria. The older literature was extensively summarized in the excellent books
of FRED et al. (1932) and WILSON (1940) and the review of ALLEN and ALLEN
(1958).

2.4.2 Description
Leguminous root nodules are more or less spherical to cylindrical structures,
in the latter case sometimes branched into finger-like or coralloid forms. In
292 A. QUISPEL:

most cases they are exogenous to the root pericycle, either one-sided laterally
adhered to the roots or encircling them. Exceptional nodule types are formed,
e.g. in Arachis hypogaea, (CHANDLER 1978). All nodules are formed by some
meristematic activity. In many plants such as, e.g. peas, the meristematic activity
continues some time at the apex of the nodule. In such nodules the youngest,
recently infected cells are found immediately behind the meristem, while old,
more or less senescent parts are present at the base. In other plants such as,
e.g. soybeans, the meristematic activity is not persistent after the nodules are
formed. The development of the nodules up till the senescent state then takes
place more or less synchronously in all cells of the nodule. All nodules contain
some small vascular bundles which form the connection with the vascular system
of the root.
The main part of the healthy, actively N 2-fixing, nodules is occupied by
the bacteroid-containing cells. These are large, parenchymatous cells which are
densely packed with the actively N 2-fixing cells of Rhizobium called bacteroids.
Compared to normal bacteria, as they are observed in Rhizobium cultures, these
bacteroids are swollen, and in most cases more or less pleomorphic. They will
be discussed more extensively in Section 5.2 (Fig. 3). In all cases the bacteroids
are surrounded by an envelope membrane or peribacteroid membrane, which
is derived from the host cell plasmalemma. Another characteristic of these cells
is the presence of the red pigment leghemoglobin (see Sect. 5.5). The characteris-
tics of the bacteroid-containing cells can be best understood by studying their
development.
Microscopical and ultrastructural descriptions of leguminous root nodules
and their development are available for species of Pisum, Trifolium, Medicago,
Glycine, Arachis, Lupinus, Phaseolus, and members of the cowpea group. We
refer for recent discussions and detailed references to KUNE (1975a, b), KIJNE
and PLANQuE (1979), NEWCOMB (1976), ROBERTSON et al. (1978a, b), Tv (1979)
and WERNER and MORSCHEL (1978).

2.4.3 Infection and Nodule Development

If rhizobia from an infective strain are inoculated on the roots of a susceptible
host plant they start developing in the rhizosphere and get adsorbed to the
root hairs. Then the normally straight-growing root hairs are deformed, during
which a phase called "marked curling" is especially important (YAO and
VINCENT 1976). The metabolized cell wall, which is correlated with this deforma-
tion, is the place where the bacteria enter the root hair by way of an invagination
of the cell wall producing an infection thread. Such threads contain a mucilagin-
ous mass in which the bacteria multiply and which is surrounded by a cell
wall deposited by the host cell (Fig. 2). In some plants, like lupin or Arachis,
infection threads could not be observed. Inward growth of the infection thread
is preceded by the host cell (root-hair) nucleus. For a more extensive description
of this infection process we refer to DART in QmsPEL (1974) and HARDY and
SILVER (1977). If the infection thread reaches the opposite cell walls the bacteria
multiply in the intercellular space from which new infection threads are formed
in the next cell. In this way the infection threads grow towards the inner part
II.2 Dinitrogen-Fixing Symbioses 293

nucleus

curled
root hair

Rhlzobial infection of
leguminous root hair o ()
slime
Fig. 2. Rhizobial infection of leguminous root hair: schematic representation of the bind-
ing of the bacteria in the slime at the root hair tip and the invagination of the infection
thread. (Courtesy of J.W. KIJNE)

of the cortex. In most cases the infection threads grow towards the nuclei of
the host cells which are swollen, synthesize DNA but do not yet divide (LIB-
BENGA and BOGERS, in QmsPEL 1974). In peas, cell divisions are only initiated
in the inner cortex of the infected roots (LIBBENGA and BOGERS, in QmsPEL
1974); in other plants the central or outer cells of the cortex are involved.
These cell divisions are the beginning of the nodule meristem.
Up till now the bacteria are surrounded by a cell-wall deposit and thus
are still intercellular. The change to the intracellular situation is limited to the
cells which are formed by the meristems, either the meristematic centres formed
in the cortex, or, during further growth of the nodules, the apical meristems.
In these cells the cell wall deposit around the infection thread becomes very
thin and finally is absent. Now the bacteria are taken up by the host-cell cyto-
plasm in a process of endocytosis of the plasmalemma. The endocytotic mem-
brane encloses the bacteria and grows with the multiplying bacteria till they
are transformed into the bacteroids. Bacteroids are still surrounded by the mem-
brane, now called peribacteroid membrane. Either every bacteroid is surrounded
by its own peribacteroid membrane or, as in soybeans, the peribacteroid mem-
brane encloses a small group of bacteroids. This phase of development more
or less coincides with the synthesis of leghaemoglobin (Sect. 5.5).
When the bacteroids leave the infection threads by endocytosis, more or
less dramatic events take place in the host cell cytoplasm. The vacuoles may
become filled with granulous material resembling cytoplasm with ribosomes,
which in peas coincides with a disrupted vacuole membrane (KIJNE ·1975a).
At a later phase the cell recovers and new vacuoles are formed. There is a
marked enlargement of the cells with an increase of ribosomes, endoplasmatic
reticulum, mitochondria and plasmids. In the fully developed bacteroid-contain-
ing cell (Fig. 3) the mitochondria occupy the outward surface of the cytoplasmic
cell contents. The nuclei swell, while the nuclear membranes are unaltered and
are surrounded by the expanding endoplasmatic reticulum. The role and func-
294 A. QUISPEL:

Fig. 3. Electron microscopic photograph of a root nodule cell from a nodule of Pisum
sativum filled with bacteroids. The adjoining cell remained uninfected. (KUNE 1975a)
Bar=5l1m

tion of small vesicles is uncertain. They may be formed from the endoplasmatic
reticulum and the dictyosomes, and might play a role in the synthesis and
the removal of the infection thread cell wall material. This latter activity might
be indicated by the presence of vesicles with fibrillous material. Another function
might be the transport to or from the peri bacteroid membranes (ROBERTSON
etal.1978a).
In older nodules or at the base of nodules with an apical meristem senescence
takes place. Depending on the host-bacterial strain combination studied, degen-
eration starts with the bacteroid, with the host cell protoplasm or there is a
simultaneous breakdown. Senescent nodules are characterized by the green bili-
verdin-like substances derived from the breakdown of leghaemoglobin.
The normal sequence of events of infection and nodulation only occurs
in compatible combination of host plants in Rhizobium strains. In incompatible
or less compatible combinations either no infection is possible (non-infective
combinations) or the infection and nodulation process is somewhere interrupted
or develops in an abnormal way. Growth of infection threads may be inter-
rupted, the bacteria may remain in the infection threads or, if endocytosis is
11.2 Dinitrogen-Fixing Symbioses 295

possible, they do not change into typical bacteroids, no synthesis of leghaemo-

globin may be possible or there are disfunctions in the N 2-fixing or assimilatory
system. Very often a too early disintegration and degeneration of bacteroids
takes place. Typical examples of ultrastructural events leading to ,abnormal
nodulation were described by NEWCOMB et al. (1977). In all cases where no
or insufficiently N 2-fixing bacteroids are formed the nodules are ineffective.
Effectivity may be further influenced by environmental factors. Especially
the presence of bound N, e.g. in the form of higher concentrations of NH:
or NO; inhibits infection and nodulation as well as the N2 fixing activity
of the nodules formed. Some possible explanations for this inhibiting effect
will be discussed later (e.g. in Sect. 6).

2.4.4 N2 Fixation (C 2H 2 Reduction)

The presence of N 2 fixation, already extensively proven by the practice of agri-
cultural N yields and older experiments with growth on N-poor soils or nutrient
solutions, has been extensively confirmed by numerous 15N2 experiments and
determinations of C 2H 2 reduction under laboratory and field conditions. N 2-
fixing activity of course depends on the plant - Rhizobium combination studied
and the environmental conditions. Some extrapolations to annual N increases
under field conditions are based on the N 2 (C 2H 2) reduction as measured under
field conditions, the environmental situation and the life span of the active
root nodules (see Table 2, Chap. 11.1, this Vol.).

2.5 Non-Leguminous Root Nodules with Rhizobium

The only non-leguminous plants in which the presence of effective root nodules
with Rhizobium is well-documented belong to the genus Parasponia of the family
Ulmaceae (TRINICK 1979, TRINICK and GALBRAITH 1976). They were initially
identified as Trema species, which however do not possess root nodules. The
root nodules look more like those of the non-legumes, as they have a central
vascular bundle. The bacteria remain in the infection threads though their cell
wall deposits may become very thin and finally even disappear. Yet no real
endocytosis was observed nor was formation of bacteroids obvious. The leaf
nodules which are found in the leaves of many non-leguminous tropical plants
will not be discussed in this chapter, as previous claims of N2 fixation could
not be confirmed by modern studies (BECKING in QmsPEL 1974).

3 The Dinitrogen-Fixing Micro-Symbionts:

Isolates and Cultures
3.1 Introduction

Studies of symbiotic associations must be based on experiments with the sym-

biotic system and on experiments with the isolated symbiotic partners. Usually
there are few problems in obtaining the plants without their. micro-organisms.
296 A. QUISPEL:

It may be far more difficult to obtain cultures of the micro-symbionts. Moreover,

it soon appears that isolated micro-symbionts may differ in many aspects from
these organisms in the symbiotic situation. It is necessary to make a clear distinc-
tion between isolations of the micro-symbionts out of the symbiotic situation
and further cultivation of these isolates. Isolates may be obtained after micro-
dissection or crude homogenization after which the cells of the micro-symbiont
can be freed from debris of the partner by the usual methods for the preparation
of cell organelles like filtration, centrifugation, buoyant density centrifugation
or chromatography on suitable Sephadex columns. After isolation we can try
to make cultures in suitable nutrient media. This is not yet possible for all
micro-symbionts. When cultures are obtained it is always necessary to prove
the identity of the cultivated organisms with the micro-symbionts by application
of Koch's criteria. The cultivated organisms must be able to infect axenic plants
belonging to the specific symbiotic partner after which they can again be isolated
from this reconstructed symbiosis.

3.2 Cyanobacteria

Though the Cyanobacteria from most symbiotic systems described above

(Sect. 2.1) can be readily isolated, cultivation and reinfection was seldom suc-
cessful. Only from some liverworts were Cyanobacteria isolated, cultivated and
used for successful re-inoculations. Reinfections of Azolla plants with "Ana-
baena azollae" have never been successful up till now, so that their identity
with the endophytic Anabaena from Azolla must be doubted (for references
see PETERS et al. in NEWTON and ORME-JOHNSON 1980).
Freshly isolated Nostoc from the liverworts Anthoceros punctatus and Blasia
pusilla did not show any photosynthetic activity. This suggests that in the sym-
biosis they obtain organic substrates from their partners, the photosynthetically
active liverworts. They were able to reduce N 2 with an active nitrogenase, but
the assimilatory enzymes glutamine synthetase (GS) and glutamate dehydroge-
nase (GDH) were hardly or not active. All NH 3, which was formed by N2
fixation, was excreted into the medium (STEWART and RODGERS 1977). Similar
observations were made with isolated Cyanobacteria from lichen-cephalodia
and Azolla (PETERS et al. l.c.). The high activity of nitrogenase coincided with
the high percentage of heterocysts.
After cultivation, as it was possible with the liverworts', Nostoc strains, photo-
synthetic activity recovers, the percentage ofheterocysts and nitrogenase activity
decreases, while the activity of as and the use of the NH3 for protein synthesis
recovers.
There are only few data about the specificity of the Cyanobacteria. The
liverworts studied by STEWART and RODGERS could be infected by an isolate
from Gunnera, while infections with isolates from Cycadeae were unsuccessful.
The cultivated cyanobacterial microsymbionts are in no way different from
other heterocyst-forming cyanobacteria. We refer to Chapter 6, Volume 6, this
Series.
11.2 Dinitrogen-Fixing Symbioses 297

3.3 Frankia, the Endophyte from the Actinorhizas

3.3.1 Isolation and Cultivation

In most cases crushed nodule suspensions can be used for infection and nodula-
tion experiments. Such suspensions thus contain the endophytes in a vital, infec-
tive form. Clusters of hyphae with vesicles can be isolated by filtration through
Nylon filters, by Sephadex G50 chromatography or sucrose density gradient
centrifugation (for references AKKERMANS et aI. in NEWTON et aI. 1977, AKKER-
MANS et aI., BAKER et aI. in GORDON et aI. 1979). Though many disputable claims
have been made about successful, but mostly non-infective cultures of the endo-
phytes, it is only since 1978 that infective cultures forming effective root nodules
were obtained by QrnsPEL and T AK for Alnus glutinosa, by TORREY and co-
workers for A. rubra, A. viridis and Comptonia peregrina and by LECHEVALIER
for Myrica pennsylvanica (further references in TORREY and TJEPKEMA 1979,
GORDON et aI. 1979) while an infective, but not effective, strain was isolated
from Elaeagnus umbellata, BAKER et aI. 1979). All cultivated strains grow as
thin hyphae with a diameter less than 111m, consisting of typical prokaryotic
cells, sometimes forming small vesicle-like cells but most characterized by spo-
rangia of the same type as described inside root nodules by VAN DIJK and
MERKUS (1976) (Fig. 4). A quite similar organism was already described by
POMMER in 1959 as an infective culture from A. glutinosa nodules, which alas
could not be maintained. We may honour POMMER as haven given the first

Fig. 4. Cultivilted form of a Frankia, isolated from spore ( - ) nodules of Alnus glutinosa.
Note the formation of a sporangium by actinomycete-like I:).yphae. (Photo J. VALSTAR)
BAR=10 11m
298 A. QUiSPEL:

description of the real endophytes of Actinorhizas. The generic name Frankia

is now generally accepted (BECKING in BERGEY 1974). Its taxonomic position
will be near the Dermatophyllaceae (VAN DUK and MERKUS 1976, LECHEVALIER
and LECHEVALIER in GORDON et al. 1979).

3.3.2 Specificity
Most studies on specificity of the endophytes were done before cultivated
Frankia strains were available, and were based on cross-inoculation experiments
with crushed nodule suspension. Infection and nodulation appeared to be possi-
ble if plants were inoculated with crushed nodules from the same or related
species. Positive results were obtained in cross-inoculations between the Elaeag-
naceae: Elaeagnus, Hippophae and Shepherdia, but no results were obtained
between Alnus, Myrica, Casuarina, Ceanothus, Coriaria and other unrelated
species (BOND in QUISPEL 1974). Positive results were claimed for cross-inocula-
tion between Alnus glutinosa with Myrica gale (RODRIGUEZ-BARRUECO and BOND
in NUTMAN 1976) and Elaeagnus umbellifera (MIGUEL et al. in DOBEREINER et al.
1977). Not all related species yield good results. Sometimes cross-inoculation
between different species of one genus like Alnus or Myrica or between geo-
graphically distinct plants yields non-effective nodules (BOND). VAN DUK (1978)
showed that the presence or absence of spore formation in the nodules [Spore
( +) or Spore (-) type] is a characteristic of the endophyte strains used for
inoculation. This is the more remarkable since all Frankia strains which are
isolated from spore (-) types of nodules abundantly produce sporangia and
spores in cultures.
Cross-inoculation with the isolated and cultivated Frankia strains has shown
that cross-inoculations have positive results when the strain from Comptonia
peregrina is inoculated on roots of different Alnus species (LALONDE in TORREY
and TJEPKEMA 1979, LALONDE and CALVERT in GORDON et al. 1979). Frankia
cultures from A. rubra infect Myrica spp. but no Ceanothus plants (BERRY
and TORREY in GORDON et al. 1979).
The Frankia cultures from A. glutinosa infect Myrica gale but are not infec-
tive to Hippophae rhamnoides. The structures, which are formed inside the nod-
ules: vesicles or club-shaped swellings, are specific for the host plant and are
not determined by the Frankia strains (LALONDE and CALVERT, BERRY and
TORREY in GORDON et al. 1979). They thus cannot be used for taxonomic pur-
poses as was done by BECKING (in BERGEY 1974).

3.3.3 Nutrient Requirements

The strains which were isolated up till now can be divided into three classes:
1. The strains isolated by TORREY and collaborators from, e.g. Comptonia and
American Alnus species grow well, though slowly, on simple nutrient solu-
tions with salts, yeast extract and casamino acids. According to LALONDE
and CALVERT (in GORDON et al. 1979), they show a much better growth
after addition of commercial lecithin to the nutrient solutions.
11.2 Dinitrogen-Fixing Symbioses 299

2. The strains isolated from Alnus glutinosa spore ( - ) type are absolutely depen-
dent on the presence of root lipids.
3. Some other strains, like an isolate from the spore ( + ) type of Alnus glutinosa,
needed these root lipids only during isolation (QUISPEL and BURGGRAAF in
GIBSON and NEWTON 1981).
All strains grow well on Tween 80 or fatty acids as C and energy source
(BLOM 1980).

3.3.4 Metabolic Activities

Cultivated Frankia strains reduce CzH z and fix 15N z so that they can form
an active nitrogenase. This formation of nitrogenase is repressed by ammonium,
but its activity is remarkably insensitive to oxygen. The correlation between
nitrogenase induction and the formation of vesicles in these cultures is a strong
indication that these vesicles are the sites of nitrogenase activity and might
to some extent be comparable to the heterocysts of Cyanobacteria (TJEPKEMA
et al. 1980). Isolated clusters of hyphae with vesicles from root nodules of A.
glutinosa contain nitrogenase, an active hydrogenase, glutamate dehydrogenase
and glutamate synthase but no glutamine synthetase, which is present in the
plant cell filtrate together with enzymes of citrulline biosynthesis (AKKERMANS
et al. in NEWTON et al. 1977, GORDON et al. 1979).

3.4 Rhizobium

3.4.1 Isolation and Description

After the first isolation of "Bacterium radicicola "by Beyerinck, strains of Rhizo-
bium, the generic name given by FRANK in 1890, were isolated from root nodules
of many leguminous plants. No special difficulties were encountered in isolation,
cultivation and reinfection experiments. In most cases rather simple nutrient
solutions may be used, such as yeast-extract/mannitol. Completely defined
media must contain a suitable sugar, the usual salts and some vitamins like
biotin or thiamin. For some strains, and especially for single cell cultures, more
complicated media are recommended. There are no special characteristics to
identify Rhizobium strains besides their possibility to infect roots of susceptible
plants. Non-infective strains therefore can only be identified as rhizobia if it
is absolutely certain that they originated, e.g. by mutation, from infective strains.
For a more extensive discussion of all aspects of Rhizobium we refer to VINCENT
(1970, in QUISPEL 1974, in HARDY and SILVER 1977).
Rhizobium species are aerobic non-spore-forming bacteria, consisting of
Gram-negative short rods with one or more flagellae. On agar media they form
white or occasionally pink colonies with a more or less slimy surface, especially
on carbohydrate-containing media. The cells contain a nucleoid and many ribo-
somes, spherical electron-dense bodies which most probably consist of glycogen,
and older cells contain poly hydroxybutyric acid (PH B) as a reserve material.
The cell envelope has the composition and structure which is typical for
Gram-negative bacteria. At the outside of the cytoplasmic membrane a small
300 A. QmSPEL:

layer of mucopeptide is covered by an outer membrane with protein, lipids,

polysaccharides and lipopolysaccharide (LPS). Several studies were made of
the LPS to study the possible importance for taxonomic classifications, to
explain the morphological changes during bacteroid formation (see Sect. 5.2)
and more recently to find antigenic binding sites for lectins (see Sect. 4.1). For
further references see PLANQuE et al. 1979, CARLSON et al. 1978). The sugar
analysis indicates a composition which is not essentially different from the best-
studied LPS of coliform bacteria. Differences between strains may be consider-
able, between strains isolated from one cross-inoculation group no less than
between strains from different groups. The outer membrane may be covered
by a polysaccharide capsule and extra cellular slime (EPS). In liquid nutrient
solutions this EPS can be precipitated from the medium. Both PHB and EPS
may function as reserve substances.
Periplasmic enzymes, such as alkaline phosphatase, have been found (VAN
BRUSSEL 1978, GLENN and DILWORTH 1979) in R. leguminosarum. They can
be liberated from the cells by osmotic shock methods. Binding activities, suggest-
ing the presence of binding proteins, have been found for glucose, glutamine
and other substrates (VAN BRUSSEL in preparation).

3.4.2 Taxonomy
Rhizobium belongs to the family Rhizobiaceae (BECKING in BERGEY 1974). The
only other genus of this family is Agrobacterium of which A. tumefaciens, the
responsible agent of crown galls, is the best-known representative. Both bacteria
are so much related that some bacteriologists prefer to classify all of them
in the genus Rhizobium.
There is a marked difference in growth velocity between different Rhizobium
species on the usual nutrient solutions. This leads to a sharp distinction between
fast-growing and slow-growing species. This difference is reflected in other char-
acteristics, like the insertion of flagellae, acid-production, sugar-substrate speci-
ficity, metabolic pathways, DNA-composition, serological differences and phage
specificity. All further distinctions between Rhizobium "species" are based on
host specificity. It is usual to classify the leguminous plant genera as well as
their specific bacteria in cross-inoculation groups (Table 2). Rhizobium strains
isolated from plants of a certain cross-inoculation group are infective for other
plants. However, there are many exceptions. Nevertheless, these cross-inocula-
tion groups and the Rhizobium "species" based on these groups have proven
their practical usefulness.

3.4.3 Metabolism
Rhizobium strains are all aerobic bacteria with a respiratory type of metabolism.
They cannot live anaerobically by fermentation, but some strains can use N0 3
for anaerobic dissimilatory N0 3 reduction. For most strains the presence of
the Embden-Meyerhof pathway is doubtful. Certainly many strains can use
the Pentose Phosphate pathway and the Entner-Doudoroff pathway, the latter
being the predominant one for most slow growers (for further references see
11.2 Dinitrogen-Fixing Symbioses 301

Table 2. Cross-inoculation groups in the Rhizobium symbioses, in which mutual infections

by the corresponding Rhizobium are possible

Group Rhizobium Plants which are Occasional

regularly infected infections

Fast growing
Pea group R.leguminosarum Pisum, Vicia, Lens Trifolium
Lathyrus, Cicer
Clover group R. trifolii Trifolium
Bean group R. phaseoli Phaseolus
Medicago group R. meliloti Medicago, Melilotus,
Trigonella

Slow growing
Soybean group R. japonicum Glycine
Lupinus-Lotus R.lupini Lupinus, Ornithopus,
group Lotus, Anthyllis,
Astragalus, Caragana,
Ononis, Dorycnium
Cow-pea R. strains Stylosanthes, Centrosema,
miscellany Desmodium, Lotononis,
specialized Leucaena
promiscuous R. strains Vigna sinensis, Other Phaseolus
Phaseolus lathyroides species Medicago,
Ph. atropurpurea, Glycine
Arachis hypogaea
Parasponia

VINCENT and BERGERSEN in QUISPEL 1974, RONSON and PRIMROSE 1979). In

R. japonicum a direct oxidation of gluconate has been postulated while the
presence of the tricarbonic acid cycle has been demonstrated.
At normal Oz concentrations the respiratory pathway consists of cytochro-
mes b, c and a-a3. Under reduced Oz concentration cytochrome a-a3 is no
longer formed, while there is a production of protoporphyrin IX, the haem
of leghaemoglobin, due to an increase of the activities of the enzymes 6-amino-
levulinic synthetase (ALAS) and -dehydrogenase (ALAD), (AVISSAR and
NADLER 1978).
Different substrates can be used as respiratory substrates, hexoses, pentoses
and organic acids from the tricarbonic acid cycle (SKOTNICKI and ROLFE 1978).
Many catabolic enzymes, as well as some transport systems, are induced by
their specific substrates (DE VRIES 1980). Because a typical catabolite repression
by glucose is absent, simultaneous use of different substrates is possible (DE
HOLLANDER and STOUTHAMER 1979), while growth, and as we shall see in
Section 3.4.4, N z fixation, may be stimulated by the combination of different
substrates like pentoses, glycerol and succinic acid (WILCOCKSON and WERNER
1979). A quite unexpected substrate for some Rhizobium strains was recently
discovered by HANUS et al. (1979). In these strains Hz can induce hydrogenase
302 A. QmSPEL:

and ribulose bisphosphate carboxylase. They can use the energy obtained by
the oxidation of H2 for CO 2 assimilation in the Calvin pathway like typical
hydrogen bacteria.

3.4.4 N 2 Fixation (C 2H 2 Reduction)

Up till 1975 nobody succeeded in demonstrating N2 fixation (C 2H 2 reduction)
in cultivated rhizobia. After successful efforts to inoculate rhizobia into plant
callus cultures, it was observed that mere contact with these plant cells was
sufficient to induce nitrogenase activities in some Rhizobium strains (see CHILD
and KURZ 1978). This led in 1975 to the independent discovery in five laborato-
ries that these strains reduced C 2H 2 in certain nutrient solutions either on agar
media (see GIBSON et al. 1976) or in stationary liquid of chemostat cultures
(EVANS and KEISTER 1976). Up till now C 2H 2 reduction has been demonstrated
in a number of strains of slow-growing rhizobia, mainly belonging to the cowpea
group and R. japonicum. Requirements are a nutrient solution with a suitable
sugar, e.g. arabinose instead of glucose, the simultaneous presence of succinic
acid or a related organic acid and the presence of a nitrogenous substance
such as glutamate, glutamine or NH; salts in concentrations below 0.1 mM.
The most important prerequisite for nitrogenase activity is the establishment
of a very low O 2 concentration. In agar and stationary liquid cultures this
low O 2 concentration is obtained by the respiration of the cells.
The fact that C 2H 2 reduction is first detected at the end of the exponential
growth period might be explained by the decrease of O 2 in dense bacterial
populations. It has been suggested that the combination of sugars and organic
acids enables the optimal respiratory activity to remove most of the O 2. Another
explanation is that, due to certain regulatory mechanisms in this situation, the
energy of respiration is chanelled to N 2 fixation instead of the biosynthesis
of cell material (WILCOCKSON and WERNER 1979). Indeed N2 fixation is only
detected in phases of the growth curve where growth is severely impeded, so
that most of the NH3 formed is excreted into the medium (O'GARA and SHANMU-
GAM 1976). Several authors observed that in N 2-fixing cultures the bacterial
cells may be swollen and pleomorphic, more or less resembling the bacteroids
in the nodules (PANKHURST and CRAIG 1978, VAN BRUSSEL et al. 1979, WILCOCK-
SON and WERNER 1979). This again is an indication of the absence of normal
cell growth. The nitrogenase activity in cultures of Rhizobium appears to be
rather insensitive to the presence of NH; in the medium as compared to other
N 2-fixing bacteria. This is related to the low O 2 concentrations; at higher O 2
content nitrogenase activity is far more sensitive to NH; (BERGERSEN and
TURNER 1978).
The original observations on the induction of nitrogenase activity in Rhizo-
bium by plant callus cultures is not only explained by the provision of the
right substrates or the establishment of the optimal low O 2 concentration, as
suggested by CHILD and KURZ (1978). REpORTER developed a system in which
Rhizobium cultures were separated from callus cell suspensions. In this system
Rhizobium strains were activated to reduce C 2H 2, which otherwise did not show
any nitrogenase activity. Two active principles were produced by these callus
11.2 Dinitrogen-Fixing Symbioses 303

cultures which stimulated nitrogenase activity in the bacteria. It was claimed

that the production of such active factors was induced by the Rhizobium cells
or nutrient solutions in which Rhizobium had been cultivated (BEDNARSKI and
REpORTER 1978).
The development of nitrogenase activity in Rhizobium cultures is a real de
novo synthesis. This is shown by inhibition with chloramphenicol or rifampicin
(WERNER 1978) and incorporation of L- 3s S-methionine in nitrogenase compo-
nent 1 (LIM et al. 1979). The latter experiments showed that nitrogenase synthesis
is repressed by high O 2 concentration and derepressed under microaerophilic
conditions. Derepression is moreOVer inhibited by N0 3, most probably through
N0 2 , and by cyclic GMP.
The NH3 formed is further assimilated by the enzymes glutamine synthetase
(GS). Rhizobium and the closely related Agrobacterium species are unique in
possessing two quite distinct forms of GS: GS I which is inactivated by adenyly-
lation in the presence of non-limiting concentrations of NH 3, and GS II, which
is not adenylylated but is very sensitive to higher oxygen concentrations, higher
temperatures and certain chemical agents. There are no indications about a
regulatory role of the GS operon for the induction of nitrogenase synthesis
as has been postulated for Klebsiella (references LUDWIG 1978, RANGA RAo
et al. 1978).
N 2 fixation (C 2 H 2 reduction) and N assimilation in fresh isolates from root
nodule bacteroids will be extensively discussed in Sections 5.4 and 5.5

3.4.5 Genetics
The use of mutants which are defective in infectivity or effectivity opens up
very promising possibilities for analysis. Especially when non-infective mutants
are obtained, it is essential to start with strains characterized by genetic markers
to be sure that the mutants which are obtained really are rhizobia and not
contaminants.
Mutants are usually obtained with nitrosoguanidine as mutagenic agent.
Most isolated mutants are auxotrophs without further defects in their symbiotic
behaviour (JOHNSTON and BERINGER in HOLLJENDER 1977). More interesting
are mutants which are symbiotically defective (e.g. DENARm et al. in NUTMAN
1976), or even show better symbiotic performance (MAIER and BRILL 1978).
A very promising method to obtain mutants using transposon mutagenesis was
described by JOHNSTON etal. (1978). Upon transfer of a RP4 plasmid, containing
phage Mu with an inserted Tn5 transposon, selection is possible for Kanamycin
resistance coded by Tn5. As a result of the suicide character of this plasmid
upon transfer to hosts without Mu repressor, the transposon Tn5 is found
inserted at other sites (e.g. in the chromosome). A high frequency of auxotrophs
was reported. As these mutations can be mapped by mapping the position
of the drug-resistant marker, they must be useful for isolation and mapping
of symbiotically defective mutants, with a reduction in time-consuming plant
tests. Genetic exchange among mutants only recently led to convincing results.
Successful transformation depends on a most probably periplasmatic compe-
tence factor (DANDEKAR et al. 1978). Only a few successful transduction experi-
304 A. QUISPEL:

ments were published (e.g. by BUCHANAN-WOLLASTON 1979). Most experiments

on genetic recombination were performed by conjugation with plasmids e.g.
the P-plasmid R 68.45 (JOHNSTON et al. 1978, BERINGER et al. 1978). Many chro-
mosomal genes were transferred, but a gene for host specificity co-transferred
with the chromosomal alleles was never tested. High frequencies of transfer
of "nodulation genes" higher than those of the transfer of chromosomal
markers are strong evidence that such genes are present on a plasmid. Plasmids
are indeed found in all Rhizobium species tested up till now. Many strains
even contain more than one big plasmid. Many, if not all, genes which code
for factors essential for infection and subsequent nodulation appear to be located
on one of such plasmids in R. leguminosarum and other strains. Such plasmids
are therefore called sym-plasmids (see articles by BERINGER and PRAKASH et al.
in GIBSON and NEWTON 1981).
The presence of nif genes on plasmids was made probable by the demonstra-
tion of homologous sequences between plasmid DNA from R. leguminosarum
and Klebsiella nif gene DNA (Nun et al. 1979). For further discussion of nif
genes we refer to Chapter II.l, this Volume).

4 Symbiotic Relations

Most research on the physiological and biochemical aspects of the symbiotic

relations, which lead to the establishment of a N 2 -fixing symbiotic system, was
and is based on the root nodule symbiosis of legumes. Therefore this chapter
will be mainly based on research with the legume-Rhizobium symbiosis with,
where possible, incidental comparisons with other symbioses.

4.1 Chemotaxis and Rhizosphere Accumulation

There are some indications that Rhizobium is chemotactically attracted by roots

and root extracts. As non-mobile mutants of Rhizobium strains are usually
as infective as mobile strains, we must conclude that chemotaxis will play only
a minor role (NAPOLI et al. in NEWTON and ORME-JOHNSON 1980). After contact
of Rhizobium cells with the root surface, they start to multiply in the rhizosphere.
Some analyses have been made of the substances which are exudated by the
roots and may stimulate bacterial growth (e.g. EGERAAT 1972). All analysis
indicated the presence of many amino acids with some smaller p,eptides. Homo-
serine may playa dominating role. Though rhizosphere accumulation certainly
is an important phase for the infection process, there are no indications that
specific differences in root exudate composition and rhizobial growth stimula-
tion play a role in the specific recognition process between leguminous roots
and Rhizobium strains. Nor is there any evidence that the rhizobia in the rhizo-
sphere already develop nitrogenase activity.
II.2 Dinitrogen-Fixing Symbioses 305

4.2 Binding of Rhizobium to Root Hairs

LI and HUBBELL in 1969 suggested that the specific recognition between plant
roots and bacteria, which determines whether infection will follow, already takes
place at a very early phase. It has been repeatedly observed that Rhizobium
cells may adhere to root hairs or plant callus cells. This adhesion, mostly as
a polar binding of one of the tips of the bacterial cell, is not a specific process
but may take place at different types of surfaces. On the other hand DAZZO
et al. 1976) observed a four to five times as abundant binding to clover root
hairs by infective strains of R. trifolii than by non-infective strains.
In peas CHEN and PHILLIPS (1976) could not find any difference in binding
of infective and non-infective Rhizobium strains. In non-leguminous plants the
remarkable polar exo-encapsulation threads, described by LALONDE et al. in
NEWTON et al. (1977) turned out to be non-specific and no indication for a
successful infection (QmsPEL in GORDON et al. 1979). Yet it is very well possible
that a microscopically visible adhesion of Rhizobium cells to root hairs, though
generally non-specific, may include specific recognition aspects which are deter-
minative for the initiation of the infection.
HAMBLIN and KENT (1973) were the first to consider the possibility that
phyto-haemagglutinins, presently called lectins, playa role in a specific recogni-
tion process between root hairs and Rhizobium cells. Lectins are sugar-binding
proteins, which are usually extracted from seeds, and which are important tools
in medical-biological research as agglutinants for e.g. red blood cells and lym-
phocytes (see also Chap. 5 in Vol. 13 B, Plant Carbohydrates II, of this Encyclo-
pedia). They bind to the surface of such cells and may initiate cell divisions
in lymphocytes. Little is known about their possible functions in the plant.
BOWLES and KAUSS (1976) suggested that they are involved in cell-wall synthesis
as sugar-transporting agents. Indeed their most characteristic aspect is the bind-
ing to specific sugar configurations. Binding to polysaccharides, e.g. cell wall
constituents, can be prevented or removed by sugars which as haptens compete
with the specific groups of the polysaccharides. A very attractive theory about
the possible role of lectins in the very early phases of infection was developed
by DAZZO and HUBBELL (1975). By serological methods they demonstrated
the presence of common cross-reactive antigens on roots of white clover and
the cell surface of R. trifolii. These antigenic sites were found both on infective
and non-infective strains of R. trifolii, but infective strains displayed a distincti-
vely greater degree of antigenic reactivity. On the other hand, FITC-labeled
antisera, prepared against infective R. trifolii strains, were bound to the tips
of growing root hairs. These tips were the sites where during infection the
most abundant binding of bacteria was observed. A soluble, non-dialyzable
clover lectin was capable of binding to the antigens of both root hairs and
bacteria. Lectins are typically bivalent, which explains their agglutinating activi-
ty. On these data DAZZO and HUBBELL proposed the hypothesis that lectins
connect the common antigenic sites and in this way bind the bacteria to the
root hair tips (Fig. 5). The lectin was isolated from clover seeds and purified
as a protein called trifolin. A similar protein was eluted from clover roots
with solutions of the hapten for clover lectin, 2-deoxY-D-glucose. This same
306 A. QUISPEL:

Fig. 5. Schematic representation of the

lectin-binding theory. (Courtesy of
J.W. KUNE)

rhizobium

!IE! slime r-< lectin o lectin receptor

hapten prevents binding of clover lectin to R. trifolii, as well as the binding

of infective bacteria to the root hairs (DAZZO et al. 1976). Antiserum to trifolin
binds at the root hair region of the roots of 24-h-old clover seedlings, but
does not bind to root hairs of Medicago sativa and roots of Aeschynomene
americana and Lotus corniculatus (DAZZO in NEWTON and ORME JOHNSON 1980).
The bacterial antigenic determinants are found at the surface in a morphological
distinct capsule. This capsular material could be extracted by 1 % phenol, which
certainly will dissolve part of the LPS as well. The extracted material could
be separated into three fractions. Two of these fractions precipitated with anti-
clover root serum (DAZZO and BRILL 1979). WOLPERT and ALBERSHEIM observed
that LPS from different Rhizobium species specifically was bound to lectin from
plants belonging to the same cross-inoculation group. No binding occurred
between these leetins and the extra-cellular polysaccharides (BPS). More recent
work from the same laboratory demonstrated, however, that most mutants
which lacked BPS were non-infective but regained infectivity after genetic rever-
sion (NAPOLI et al. in NEWTON and ORME-JOHNSON 1980). Mutants without
a capsule had lost their infectivity (BAUER in NEWTON and ORME-JOHNSON 1980).
This confusing situation will only be elucidated after a still better insight into
the chemistry of the polysaccharides of the outer membrane, capsule and extra-
cellular slime, their situations in the structure of the cell wall and their ontogenet-
ical relations.
Most work on lectins was done with lectins which had been isolated from
seeds. It must be emphasized that we are only interested in lectins which are
eventually present at the root hair surface. In eluates of pea roots carbohydrate
containing lectin was found (KIJNE et al. 1980). The carbohydrate composition
resembled in some essential aspects the carbohydrates in a hot water extract
of young pea root cell walls. The hypothesis was put forward that lectins,
which normally bind sugars for cell wall synthesis, recognize bacterial polysac-
charides with the same configuration as part of their" self".
It must be admitted that the theory of DAZZO and HUBBELL is still disputed,
as different authors failed to find a correlation between the binding of lectins
and specificity of infection. We may again refer to the work of CHEN and
PmLLIPs (1976) who observed that less than 10% of infective strains of R.
leguminosarum bound pea lectin. Though FITC conjugates of antiserum against
R. leguminosarum indeed adhered to pea root hairs, these conjugates were bound
11.2 Dinitrogen-Fixing Symbioses 307

in the same way to roots of many other plants. There are some explanations
for such conflicting results. The antigenic sites at the bacterial surfaces may
be covered by non-specific slime. BAUER and coworkers (BAUER in HOLLAENDER
1977, in NEWTON and ORME-JOHNSON 1980) observed that correlations between
cross-inoculation specificity and lectin binding increased if better-purified lectin
preparations were used. The presence of lectin-binding sites on the bacterial
surface strongly depends on the phase of development. Best lectin-binding was
mostly observed during mid-log phase. In some cases binding of lectin depends
on the presence of an additional factor from the plant root culture solution.
It must be stressed that correlations between lectin binding and infection speci-
ficity must be based on direct observations of the root hair infection. A correla-
tion with nodule formation is not necessary, as many other factors may interfere
in later phases of the nodulation process.

4.3 Root Hair Deformation and Infection-Thread Formation

Root hair deformation is an essential prerequisite for infection, both in legumin-

ous (see DART in QUISPEL 1974, in HARDY and SILVER 1977) as in non-legumin-
ous infections, (CALLAHAN et al. in TORREY and TJEPKEMA 1979). Infection
threads were correctly interpreted by NUTMAN already in 1956, as an invagina-
tion process, though we must realize that this invagination can only proceed
by means of a continuous deposit of cell-wall substances around the slimy
mass of bacteria. Root hair deformation and infection thread formation must
be regarded as two aspects of a process of disregulation in the normal process
of root hair growth.
Much confusion has arisen around the question of what types of substances
are responsible for root hair deformation. Part of this confusion was caused
by the fact that no distinction was made between different types of deformations
such as curled, helical, branched etc. According to YAO and VINCENT (1976)
the type of deformation described as "marked curling" is the only one which
is directly related with the subsequent infection thread formation. Soluble factors
never produce real "marked curling", which is only obtained after direct cell-to-
cell contact with the infective bacteria. It is tempting to relate this effect to
the specific binding of Rhizobium to root hairs according to DAZZO and HUB-
BELL'S lectin bridge theory. The importance of this binding is not the binding
as such but the reactions it evokes in the cells (BAUER in HOLLAENDER 1977).
If KIJNE et al. (1980) are right in assuming that plant root lectins erroneously
recognize bacterial surface sites as carbohydrates for cell wall synthesis, we
may understand why normal cell wall growth is disrupted and channelled into
infection thread formation.

4.4 Cell Wall Degrading Enzymes

For most phytopathogenic organisms penetration of, for example, the leaf sur-
face demands the activity of cell wall degrading enzymes. Formation and activity
308 A. QUISPEL:

of such enzymes is a complex process in which cell-wall substances from the

host induce the synthesis of the enzymes in the pathogens, often in a sequential
way. On the other hand, the activity of the enzymes is susceptible to inhibition
by substances produced in the host cells (ALBERSHEIM and ANDERSON-PROUTY
1975, COOPER in SOLHEIM and RAA 1977).
It is not immediately obvious to expect similar processes in Rhizobium infec-
tion, as the infection proceeds not by penetration but by invagination. Yet
we must realize that growth of the infection thread must be accompanied by
cell wall metabolism, probably more of a synthetic than a degradative nature.
In the deformation of root hairs, degradative aspects may be dominating.
In much-discussed experiments F AHRAEUS and LJUNGGREN observed an in-
crease of "polygalacturonase" activity in the medium or in superficial phos-
phate-buffer washings of roots which were infected with Rhizobium. This activity
was only observed after addition of infective Rhizobium strains. The method
which they used does not distinguish between enzymes produced by the bacteria
or enzymes which are exuded by the roots. As nobody had ever succeeded
in demonstrating pectolytic enzymes in Rhizobium cultures, they preferred the
plant root origin of the enzymes. The specific effect of infective Rhizobium
strains was explained by the assumption that polysaccharides from the bacteria
induced "polygalacturonase" formation at the root surface (LJUNGGREN 1969).
Most other authors did not succeed in reproducing these results. The influence
of plant growth conditions, as used by the different authors, may explain some
of the discrepancies. The original supposition, that Rhizobium does not produce
pectinolytic enzymes may be erroneous. HUBBELL et al. (1978), when applying
a very sensitive agar plate assay, succeeded in demonstrating low levels of pectin-
olytic enzymes in Rhizobium strains. This was only demonstrated in Rhizobium
strains from temperate regions. In the Rhizobium strains which produced pectin-
olytic enzymes, no correlation with their infectivity was observed. In any case
these results open up new possibilities for research and a re-interpretation of
older experiments. The results of LJUNGGREN and F AHREAUS may be equally
well explained by the usual induction by plant substances of microbial enzyme
production. Nothing is known about cell wall metabolism during growth of
the infection thread nor about the reason that cell wall deposits stop as soon
as the infection reaches the cells which just were formed by the meristem. The
role of plant factors is indicated by genetic mutations in the plant genome
which lead to continued growth of the infection thread. In other cases growth
of the infection threads stops already in the root hair. Nothing is known about
the immediate expression of the genes involved (see Sect. 4.6). For a survey
of all factors controlling the nodule formation we refer to VINCENT (in NEWTON
and ORME-JOHNSON 1980).

4.5 The Role of Plant Hormones in Nodule Formation

The role of plant hormones in the initial phases of infection is doubtful. Though
indole acetic acid may be formed in the rhizosphere its influence may be at
best of a secondary nature. Cytokinins may also be produced in the rhizosphere
11.2 Dinitrogen-Fixing Symbioses 309

and may lead to root thickenings and local malformations (PuPpo and RIGAUD
1978). There are no indications that such swellings on roots of leguminous
and non-leguminous plants (BERMUDEZ DE CASTRO in NEWTON et al. 1977) bear
any resemblance to real root nodules. Certainly plant hormones are of funda-
mental importance in later phases of the nodulation process (LIBBENGA and
BOGERS in QmsPEL 1974). While in peas the cortex cells which are passed by
the infection thread react by a swelling of the nuclei due to DNA synthesis,
they do not yet divide. The first pattern of cell division is observed ahead
of the penetrating infection thread in the inner cortex cells opposite the xylem
strands. This pattern could be reproduced in excised root cortex cylinders by
application of plant hormones. Addition of auxins only stimulated divisions
in the pericycle, combination of auxins with kinetin produced the same pattern
of cell divisions in the inner cortex as is observed after Rhizobium infection.
The remarkable localization of cell divisions to the region opposite the xylem
strands indicated that factors from the vascular bundles are limiting the de-
differentiation process. Indeed it appeared that a mixture of auxins, kinetin
and an extract from the stele initiated cell divisions all over the cortex. The
initiation of cell divisions which lead to the nodule meristem therefore could
be explained by a gradient hypothesis. According to this hypothesis de-differenti-
ation of cortex cells depends on the optimal interaction of a gradient of auxins
and cytokinins from the infected cells with a gradient of as yet unidentified
factors from the xylem. The much-discussed polyploid character of most root
nodule cells could be explained by the reaction of tetraploid cells, which are
already most abundant in the outer cortex, on combinations of auxins and
cytokinins with or without root stele extracts (BoGERS et al. 1976). In non-
leguminous plants like Alnus, the initial cell divisions are observed in the outer
and central cortex cells, while in the neighbouring pericycle lateral roots are
initiated. ANGULO-CARMONA (1974) concluded that they are formed in excess
of normal lateral roots and at sites where normally no roots should have been
initiated. As the formation of root initials is stimulated by auxins, it is tempting
to suggest that the produc~ion of auxins in the infected cells is responsible
for the initiation of these lateral roots, which after infection from the primary
nodule develop into the rhizothamnia.
It is difficult to demonstrate whether the cells with an infection thread really
are centres of auxin and cytokinin production. However, it is a well-known
fact that young root nodules are hyperauxinic. This high auxin content may
be the result of synthesis by the bacteria or by an increased synthesis of the
invaded root cells. Cultivated rhizobia synthesize indole acetic acid (IAA), but
only if a precursor such as tryptophan is present in the culture solution. The
biochemical pathways for lAA synthesis are basically similar in plants and
bacteria, but there is a difference in the production of the side-product. indole
carbonic acid (ICA). A comparison of the lAA/ICA ratio led to the conclusion
that lAA synthesis in root nodules is mainly a plant-cell process (DULLAART
1970). This is in accordance with observations on phytopathogenic organisms
which induce a higher IAA synthesis in the host cells. Even if bacteria synthesize
part of the lAA, the production of the precursor tryptophan will be the main
limiting factor.
310 A. QUISPEL:

A higher production of IAA may be caused by a higher synthesis rate or

by a decreased rate of destruction. MENNES (1973) indeed observed that in
peas the specific activity of IAA-oxidase is lower in the nodule than in the
roots. Differences in IAA-oxidase activity can be explained by differences in
phenolic contents. Cytochemical studies indicated that peroxidase activity
(which can be used as an indication of IAA-oxidase) is absent in the cytoplasm
of the bacteroid-containing cells (OOSTROM et al. 1975). More difficult to under-
stand was the high peroxidase activity in the cells of the infected root cortex
area. This poses the problem whether indeed enough IAA may reach the target
cells in the inner cortex as is expected on the base of the gradient hypothesis.
A comparison with other symbiotic systems shows that hyperauxiny is found
in Alnus nodules (DULLAART 1970) but not in nodules of Cycadeae. This may
reflect the difference between nodules which are formed by the infection and
preformed nodules which are invaded by the microsymbiont. Several authors
observed an increased content of cytokinins in root nodules as compared with
uninfected root tissue. The cytokinins were identified as zeatin and its riboside,
which are normal products of root metabolism, and iso-pentenyladenine (2iP)
which is produced in Rhizobium cultures. Bacterial cytokinin production may
be stimulated by roots. In pea nodules a positive correlation was observed
between cytokinin activity and the mitotic index in the nodule meristems. It
was suggested that the bacteria in the infection thread form 2iP and the mer-
istems are responsible for the production of zeatin and its derivative. As both
substances are closely related it might be possible that the meristems convert
the 2iP from the bacteria into zeatin. For detailed references see e.g. SYONO
et al. (1976), PuPPo and RIGAUD (1978), NEWCOMB in NEWTON and ORME-
JOHNSON (1980) and HENSON and WHEELER (1977b) for non-legumes. In animal
cells mitotic activity can be stimulated by LPS from bacterial cell walls. KIJNE
et al. in SOLHEIM and RAA (1977) reported that Rhizobium- and E. coli LPS,
if added together with auxins, can stimulate cell divisions in root cortex cells
and in this respect can replace cytokinins. The active part of LPS is the lipid
A moiety. It is not yet known whether LPS stimulates cytokinin production
or directly stimulates cell divisions. High contents of gibberellins in root nodules
were found in Lupinus luteus (DULLAART and DUBA 1970) and Alnus glutinosa
(HENSON and WHEELER 1977a). DULLAART and DUBA (1970) observed a stimula-
tion by gibberellic acid of the synthesis of IAA from tryptophan by cell-free
nodule extracts. This and the well-known auxin-sparing effects of gibberellins
makes it worthwhile to pay more attention to gibberellins in the nodulation
process.

4.6 Miscellaneous Problems

The physiological studies described in the preceding paragraphs were only a
small part of the wealth of problems which arise from the description of the
infection and nodulation process (see Sect. 2.4.3). The majority of these prob-
lems can only be stated with few or no possibilities for experimental approach.
Much inspiration might be obtained from comparisons with related phenomena
in phytopathology. Certain ultrastructural events bear some resemblance with
events during, for example, hypersensitive reactions. The fundamental difference
IL2 Dinitrogen-Fixing Symbioses 311

is that hypersensitive reactions lead to disruption of cell structures, while in

nodule formation the reactions are regulated and channelled into the formation
of new functional structures. Phytopathogens may be inhibited by the produc-
tion of phyto-alexins which are formed by the host and elicited by glucan-like
substances from the cell walls of the infecting micro-organisms (ALBERSHEIM
and ANDERSON-PROUTY 1975). Nothing is known of such events during Rhizo-
bium infections, but the attention paid to cell-surface binding might lead to
some insight in these aspects as well. Here we must not limit ourselves to the
events at the root hair surface. A highly crucial moment certainly is the endocy-
tosis of bacteria, as at that moment the cell surface of the bacteria comes
in direct contact with the plasmalemma of the host cells. The discovery of
a mitogenic activity of rhizobia1 LPS (KUNE et al. in SOLHEIM and RAA 1977)
may be a first indication of the interactions involved. In the formation of crown
gall by the related Agrobacterium tumefaciens it is now well established that
plasmid DNA of the bacteria is taken up and brought to expression by the
plant cells. The first indication that Rhizobium-DNA may be taken up by legume
roots was obtained by LEPIDI et al. (1976). Our insight into complicated biologi-
cal systems is always highly increased by comparisons with systems where devia-
tions in the normal process take place. In root nodule symbioses such compari-
sons are possible with plant-bacteria combinations where the normal infection
process is somewhere disturbed. Such deviations may be based on mutations
of the genome of either the plants, or the bacteria or to unfavourable environ-
mental conditions. VINCENT (in NEWTON and ORME-JOHNSON 1980) distinguished
the following genetically determined moments in infection and nodulation:
pre-infection (root surface colonisation, root adhesion, root hair branching and
marked curling), infection (formation of the infection thread, development of
meristematic activity, release of bacteria, bacteroid development), nodule func-
tion (nitrogen fixation, complementary biological and biochemical functions
and nodule persistence). Certainly these moments can be further divided into
different gene functions. Many of such deviations were studied with Rhizobium
strains or plant species whose genetical relations were unclear. The methods
developed for isolation and genetical analysis of Rhizobium mutants (see
Sect. 3.4.5) open up possibilities for better genetic-physiological comparisons.
The same may be expected from plant molecular-genetic studies. A better study
of the genes involved and their biochemical expression will certainly be one
of the most important tools for a better understanding of the physiological
backgrounds of root nodule formation (VERMA 1980). Such studies already led
to the discovery of the de novo synthesis of specific nodule proteins.

5 The N2-Fixing System

5.1 Introduction

After our discussion of the symbiotic interactions which lead to the development
of the active N 2-fixing root nodule we will now consider the biochemical and
physiological aspects of the structures which are functional for the N 2-fixing
312 A. QUISPEL:

process in leguminous root nodules. These are the bacteroids, surrounded by

the peribacteroid membranes in the bacteroid-containing host cells. In this sys-
tem the exchange of substances between host and bacteria, which is essential
for N 2-fixing activity, must take place through a complicated structure of mem-
branes (Fig. 6). Our aim must be to understand the biochemical and transport
processes which take place in the bacterial and host cytoplasm and the mem-
brane systems between.

5.2 Bacteroids

For practical reasons some authors prefer to call all bacteria inside the root
nodules bacteroids. This is an unfortunate definition, as it neglects the great
difference between the still-multiplying bacteria and the final forms which are
the actively Nz-fixing sites in the nodules. The gradual change from bacteria
to bacteroids in the transition process may be emphasized by distinguishing
bacteria, transforming bacteria and bacteroids, the more so as these phases
can be approximately separated by sucrose gradient centrifugation (TE MAY
CHING et al. 1977). We must admit that in some plants the morphological differ-
ences between bacteria and bacteroids are far more pronounced than in others.
The old dogma that the bacteroids are no longer viable in the sense that
they can no longer develop in nutrient solutions has been questioned by GRES-
HOFF et al. (1978). If the bacteroids of clover and soybean nodules are very
carefully prepared from bacteroid-containing protoplasts, isolated by macera-
tion, they are all able to multiply on suitable nutrient media. Bacteroids are
only very susceptible, especially to osmotic damage. This coincides with the
observation of LAANE et al. (1978) that during preparation of bacteroids, mem-
branes may be damaged by the effect of fatty acids and phospholipase from
the plant cells. This can be prevented with bovine serum albumin.
The DNA content of bacteroid cells was determined in bacteroids of many
species and compared with the DNA content of bacteria in mid log and early
stationary phase. Determinations were made in isolated bacteroids or by micro-
fluorometry in situ (for detailed references see PAAU et al. 1979, SUTTON et al.
1978). The DNA content in bacteroids either equals the content of bacteria
or exceeds it by factors varying from 1.5~3 or occasionally even 6 times, depend-
ing on the leguminous association studied. It is generally observed that the
RNA content, especially the ribosomal RNA in bacteroids, is low (BERGERSEN,
in QUISPEL 1974, BISSELING et al. 1979). Protein synthesis has been demonstrated
by the incorporation oflabeled amino acids in proteins in situ and after isolation.
SHAW and SUTTON (1979) observed a change in the labeling pattern of different
polypeptides during bacteroid development. After 12 days, two, polypeptides
were especially labeled which have the molecular weight of the nitrogenase
components.
Bacteroids have a normal respiratory activity with a similar glycolytic system
as in the cultivated bacteria (e.g. KURZ and LA RUE 1977), a TCA cycle with
a partial glyoxylate cycle (STOVALL and COLE 1978) and a terminal electron
flow to 02 or NO.3. There are some changes in the cytochrome content, which,
11.2 Dinitrogen-Fixing Symbioses 313

at least partially, coincide with changes in cultivated bacteria due to anaerobic

conditions (see Sect. 3.4.3). The pigments P-420 and P-450 are supposed to
be coupled to the nitrogenase system (BERGERSEN in QUISPEL 1974). Flavins
may be present in high concentration, while a non-haem protein has been found
(PmLLIPs et al. 1973). Especially important is the presence of an intact, inte-
grated cytoplasmic membrane, which can be energized. This membrane energiza-
tion is essential to deliver the energy to the cytoplasmic nitrogenase (LAANE
et al. 1979). The mediation between the energized membrane and the nitrogenase
may be provided by the flavodoxin which is the most powerful reductant for
nitrogenase or by the non-heme iron protein. Though reserve substances like
PHBA are broken down during active respiration and N 2-fixation, the normal
substrate for respiration must be provided by the carbohydrates (sucrose) of
the phloem sap. However, all studies on respiratory substrates for bacteroids
agree that sugars are far less suitable substrates than TCA-cycle acids such
as succinate. As the catabolic enzymes for glucose degradation are present
(KURZ and LA RUE 1977) a plausible explanation may be that glucose transport
is inhibited. Indeed DE VRIES (1980) observed that bacteroids of R. legumino-
sarum no longer were able to transport the glucose analogue 2-deoxY-D-glucose,
which is well transported by bacteria. On the other hand these bacteroids pos-
sessed a highly active transport system for succinate~ Either the glucose transport
system is repressed in bacteroids by the presence of organic acids like malic
acid in the plant cells, or this transport is impeded by the absence of binding
proteins.
That carbohydrates are not essential for bacteroids is, moreover, indicated
by the fact that mutants which are defective in sugar metabolism still form
effective root nodules (RONSON and PRIMROSE 1979). If organic acids are the
favourite substrates for bacteroids, the surrounding root cell has to metabolize
the phloem sucrose to these acids (see Sect. 5.3).
Other transport systems must be present for the transport of amino acids.
In R. leguminosarum bacteroids, glutamate and aspartate are rapidly used for
CO 2 production and thus may be considered as respiratory substances. Other
amino acids like leucine are, after transport, used for protein synthesis (STAAL
in preparation). Since it is doubtful whether the NH3 formed from N 2-reduction
is used in bacteroids, the question arises whether bacteroids depend on plant
amino acids for their protein synthesis. Bacteroids possess a weakened, more
permeable, outer membrane. Direct isolation of LPS with the aqueous phenol
method showed a far smaller amount of LPS than was observed in bacteria
cultivated on different nutrient solutions (VAN BRUSSEL et al. 1977). Further
studies by PLANQuE et al. (1979) did not indicate essential differences in the
sugar and fatty acid composition of LPS of bacteria and bacteroids. There
were differences in physical characteristics, for example the behaviour <,luring
ultracentrifugation and the solubility. Part of the LPS is already lost during
isolation of bacteroids, which certainly does not happen after similar treatments
of bacteria. This altered, more soluble LPS and the more permeable structure
of the bacteroid outer membrane may have important consequences. It may
explain the swollen pleomorphic forms. The solubilized part of the LPS may
have effects on the host cells (KIJNE et al. in SOLHEIM and RAA 1977). The
314 A. QUISPEL:

increased permeability to macromolecules must have consequences for the local-

ization of periplasmatic enzymes and binding proteins which now may be dis-
solved in the whole space between bacteroid cytoplasmic membrane and host
cell peribacteroid membrane. For the time being we can only speculate about
the consequences for the symbiotic relations.

5.3 The Bacteroid-Containing Plant Cells

The cytoplasm of the bacteroid-containing plant cells has two important func-
tions: it must provide the bacteroids with the respiratory substrates and most
probably the amino acids for protein (nitrogenase) synthesis. On the other hand
they must absorb the products of N 2fixation, most probably NH 3, and assimi-
late this NH3 into amino acids. Both transport processes must take place in
the peribacteroid membrane. Cytochemical evidence, e.g. the presence of adenyl-
cyclase and staining with phosphotungstic acid (Tu 1979, ROBERTSON et al.
1978a) is in accordance with the plasmalemma origin. This, of course, does
not mean that the peribacteroid membrane is identical with the plasmalemma.
ROBERTSON et al. (1978b) prepared peribacteroid membrane fractions after
osmotic shock of membrane-surrounded bacteroid preparations.
Their identity was indicated by staining with phosphotungstic acid and silver
salts. They consisted of lipoids and protein in a 6.1: 1 ratio. As succinic acid
and related acids are the obvious substrates for bacteroids, the host cells must
produce these acids from the phloem sucrose. Moreover organic acids are neces-
sary to provide the C-skeletons for amino acid synthesis. Already in 1960
MULDER and VAN VEEN observed an increase of N2 fixation in root nodules
of peas by increased concentrations of CO 2, which promotes organic acid syn-
thesis. More recently a direct correlation was observed between C 2H 2 reduction
and CO 2 binding. Both CO 2 binding in vivo and PEP-carboxylase activity in
vitro were far higher in nodules than in roots. Studies with differently 14C
labeled glucose confirmed the role of glycolysis with PEP-carboxylation, with
only a minor role of the pentose phosphate pathway (LAING et al. 1979). The
low oxygen concentration due to the high respiration of the bacteroids may
lead to a shift in enzyme activities towards more fermentative organic acid-
producing processes. The organic acids produced, like malic acid, might have
inducing or depressing effects on metabolic and transport processes in the bac-
teroids (DE VRIES 1980).
The study of the plant cell part in the metabolic activities of bacteroid-
containing cells is extremely difficult because root nodules consist of so many
different cell types. Better progress may be expected from studies with isolated
bacteroid-containing protoplasts. Contrary to isolated bacteroid~ C 2H 2 reduc-
tion by these protoplast suspensions was markedly stimulated by sugars. More-
over C 2H 2 reduction was stimulated by nucleotide triphosphates and different
amino acids, but strongly inhibited by asparagine and glutamine (WOOl and
BROUGHTON 1979).
The function of nodule-specific plant proteins, called nodulines, is not yet
understood (LEGOCKl and VERMA 1979).
II.2 Dinitrogen-Fixing Symbioses 315

5.4 Nitrogenase

The first definite proof that bacteroids are the sites of N 2 fixation was given
by the work of KENNEDY et al. in 1966 and BERGERSEN and TURNER in 1967.
The first cell-free system of bacteroid nitrogenase was obtained by KOCH et al.
in 1967. Highly purified preparations can be obtained by purification with
DEAE-cellulose and Sephadex G 150 (WHITING and DILWORTH 1974). The ni-
trogenase resembles the nitrogenase from other N 2-fixing organisms in all as-
pects, so that we can refer to the discussion in Chapter II.l, this Volume).
The older discussions about the question whether bacteroid formation
is essential for N 2 fixation by Rhizobium (BERGERSEN, QUISPEL, in QUISPEL,
1974) have lost part of their meaning since some strains can fix N2 in cultures.
Moreover, in Parasponia N 2 fixation takes place without a conspicuous bacter-
oid formation. On the other hand in legumes the formation of nitrogenase
is only derepressed at a certain phase in the development which more or less
coincides with the development of bacteroids. This derepression was studied
by many authors by determining nitrogenase activity in bacteria and bacteroids
in different phases of development (e.g. BISSELING et al. 1979, PAAU et al. 1979,
WERNER and STRIPF 1978). In all cases a clear picture resulted of the development
of nitrogenase during bacteroid development. In peas it appeared that the two
protein components of nitrogenase were not coordinatedly synthesized, young
bacteroids containing relatively more protein II, older bacteroids relatively more
protein I (BISSELING et al. 1979).
No direct experimental evidence is available about the factors which lead
to derepression of nitrogenase but much may be learned from the regulatory
factors which influence (de)-repression in cultivated strains. Microaerophilic
conditions may be among the most important factors to derepress nitrogenase
(see Sect. 5.6). The role of GS in transcription is uncertain (see Sect. 5.5). Very
suggestive are the observations on plant factors which might derepress nitrogen-
ase (BEDNARSKI and REpORTER 1978). Though NH3 is a repressor in most N 2-
fixing organisms, either directly or by promoting GS-adenylylation, Rhizobium
appears to be rather insensitive. Though BISHOP et al. (1976) demonstrated a
65% reduction of nitrogenase activity in soybean nodules after a 4-day period
in 5 mM (NH4hS04, this reduction was more probably caused by an effect
on nodule formation than by a direct effect on nitrogenase. Other inhibiting
effects ofNHt are explained by a deviation of ATP supply (HOUWAARD 1978).
This aspect will be further discussed in Section 6. For detailed references we
refer to DILWORTH (in NEWTON and ORME-JOHNSON 1980).

5.5 NH3 Assimilation

In bacteria NH3 is assimilated by the GS-GOGAT pathway with GDH only

playing a role at high NH3 concentration. The role of GS in cultivated Rhizo-
bium bacteria was already discussed in Section 3.4.4. Especially important as-
pects are the inactivation of GS by adenylylation which is stimulated at higher
NH3 concentrations, and the suggested role of deadenylylated GS in transcrip-
tion of the nif genes.
316 A. QUISPEL:

For plants as well it is now generally accepted that GDH mainly may have
an anabolic function, while GS-GOGAT is the most active NH3 assimilatory
system (MIFLIN and LEA 1977). Affinities for NH3 are far greater for GS than
for GDH. Km values ranging from 0.02-0.5 mM and from 0.03-0.1 have been
reported. There is no evidence in plants for NH3 repression, nor for an adenyl-
ylation cascade system of GS, as in bacteria.
In bacteroids GDH is present, but usually in too low concentrations to
playa role in NH3 assimilation. Data on GS were rather conflicting (BROWN
and DILWORTH 1975), which may result from the use of different assay methods.
The often used transferase reaction is not fully specific; it measures activities
both of the adenylylated and the deadenylylated form, depending on the addi-
tion of Mg2+ or Mn 2+ and has a higher turnover rate than the biosynthetic
assay. From the data of BROWN and DILWORTH (1975), we may conclude that
the activity of the biosynthetically active de-adenylylation form is small or
negligible. When bacteroid samples of different phases of development were
compared, it appeared that activity of GS, even when measured with the trans-
ferase assay, was negatively correlated with the activity of nitrogenase (PLANQuE
et al. 1978).
Microaerophilic conditions, as they are supposed to be present in the nodules
and essential for nitrogenase derepression, decrease GS activity. As RANGA
RAo et al. (1978) observed a complete loss of activity of GS II at p02 beneath
0.4% it is not probable that this type of GS can play a role in bacteroids.
However, further studies on these two forms during bacteroid development
are required. Studies on the effect of excess NH3 or N depletion in vivo with
intact nodules, did not show any effect on the level of bacteroid GS nor on
its level of adenylylation (BISHOP et al. 1976). Glutamate synthase (GOGAT)
has been observed in bacteroids of lupin, soybeans and peas, but in most cases
the activities were very low. The general conclusion must be that the activities
of NH 3-assimilating enzymes in bacteroids are too low to match those of the
N 2-fixing rates and even get lost during nitrogenase derepression. This explains
the observation that isolated bacteroids excrete the fixed N as NH3 (for further
references see BOLAND et al. in NEWTON and ORME-JOHNSON 1980). All investiga-
tions agree that there is a high activity of GS in the supernatant which is
left after sedimentation of bacteroids from nodule homogenates and which is
considered to represent the plant cell cytosol (e.g. BROWN and DILWORTH 1975,
ROBERTSON et al. 1975, SCOTT et al. 1976). Though of plant origin, the activity
of GS depends on the development of effective bacteroid tissue. Increase of
plant GS correlates with the increase of bacteroid nitrogenase (ROBERTSON et al.
1975, STRIPF and WERNER 1978). No increase or even decrease ofGS is observed
in ineffective root nodules (WERNER et al. 1979).
Initially controversial data on plant GOGA T may be explained by methodo-
logical difficulties. The role of plant GDH is generally considered to be of
minor importance. Sometimes high activities were obtained which were even
correlated with bacteroid nitrogenase activity (STRIPF and WERNER 1978,
KENNEDY 1979). The latter author suggests that the occurrence of both comple-
mentary pathways of NH3 assimilation may be important as a regulated re-
sponse to gradients of NH3 concentrations around bacteroids in nodule tissues.
11.2 Dinitrogen-Fixing Symbioses 317

5.6 Oxygen Regulation and Legbaemoglobin

One of the most central problems for all aerobic N 2-fixing organisms is the
regulation of O 2. This O 2 is necessary for respiration which produces the energy
for N 2 fixation though the nitrogenase must be protected against inactivation
by O 2. In the root nodules we have seen that there is strong evidence that
micro-aerophilic conditions are among the most important factors to derepress
nitrogenase formation.
TJEPKEMA and YOCUM (1974) concluded from a deflection in the oxygen
concentration curve of root nodule respiration that there must be a diffusion
barrier against O 2 somewhere in the cortex. This was directly confirmed by
measurements of the O 2 concentration by micro-electrodes in the cortex. BER-
GERSEN and TURNER (1975) concluded from the O 2 concentration effect on respi-
ration of bacteroid suspensions that besides a high-affinity system which is
linked with N2 fixation through P-450, and consequently inhibited by N-phenyl
imidazole, there is a low affinity respiratory system which might act as a respira-
tory protector of nitrogenase. There is no evidence for a "switch on-switch
off" type of regulation (CRISWELL et al. 1976) (see Chap. II.1, this Vol.).
There is now general consensus that besides these types of O 2 protection
the most important regulation of O 2 provision and O 2 protection is provided
by leghaemoglobin. Since its discovery by KUBo in 1939 and identification by
KElLIN and WANG in 1945 and VIRTANEN in 1947, it was soon recognized that
leghaemoglobin content was directly correlated with effectivity. The correlation
between development of nitrogenase activity and leghaemoglobin synthesis dur-
ing nodule development is not as absolute (e.g. NASH and SCHULMAN 1976),
but we must be careful, as our methods for measuring the first nitrogenase
activity are far more sensitive than the methods for leghaemoglobin determina-
tion.
As several extensive reviews summarized our knowledge about the chemistry
and physico-chemistry of leghaemoglobin (e.g. ApPLEBY in QUISPEL 1974,
HARDY and SILVER 1977) we can restrict ourselves here to the remark that
leghaemoglobin consists of an apoprotein which has a striking similarity with
animal haemoglobins and especially myoglobins, while the haem group is a
proto haem IX.
All evidence shows that the apoprotein is coded by plant DNA and synthe-
sized on plant ribosomes. We refer to the cross-inoculation experiments of
CUTTING and SCHULMAN, which showed that leghemoglobin is determined by
the host plant, and the inhibition of 14C incorporation in leghemoglobin by
cycloheximide, but not by chloramphenicol (references: DILWORTH in NEWTON
and ORME-JOHNSON 1980). On the other hand NADLER and AVISSAR (1977) ob-
tained strong evidence that the bacteroids are the sites of haem synthesis. The
enzyme o-aminolevulinic synthetase (ALAS) could only be demonstrated in the
bacteroids while the activity of this enzyme, which plays a role in haem synthesis,
increased during leghaemoglobin synthesis. Another enzyme of the haem-synthe-
sis pathway, o-aminolevulinic dehydrogenase (ALAD), increased in the bacter-
oids during effective nodule development but decreased in the plant fraction.
Plant extracts do stimulate ALA incorporation in haem in bacteroids, which
318 A. QUISPEL:

indicates a cooperative effort of plant and bacteroids in haem synthesis. We

already saw in Section 3.4.3 that cultivated bacteria are stimulated to synthesize
protoporphyrin IX under micro-aerophilic conditions (AVISSAR and NADLER
1978). This confirms an earlier suggestion by NASH and SCHULMAN (1976) that
the synthesis of leghaemoglobin might be a reaction to a low oxygen concentra-
tion. Under such conditions both ALAS and ALAD activities increase.
The localization of leghaemoglobin is still controversial. Micro-autoradio-
grams after labelling with 59Fe showed the main activity between the bacteroids
and the peri-bacteroid membrane. Peroxidase activity, one aspect of leghaemo-
globin, was demonstrated at the same site. X-Ray probe analysis, however,
indicated the presence of leghaemoglobin in the cytoplasm of the bacteroid-
containing cells (references in ApPLEBY in QUISPEL 1974). Localization with
ferritin-labelled antiserum showed activity in the host cell cytoplasm and at
the outside of the peribacteroid membrane (VERMA and BAL 1976). The exact
localization must be important for our interpretations of leghaemoglobin func-
tion. These interpretations are based on the work of BERGERSEN et al. (1973),
WITTENBERG, APPLEBY and co-workers (for recent discussions and references
see DILWORTH and WITTENBERG in NEWTON and ORME-JOHNSON 1980). Oxygen-
ated leghaemoglobin stimulates respiration of succinate and nitrogenase activity
of bacteroids of R. leguminosarum at low O 2 concentration in such a way that
respiration far better supports N 2 fixation with leghaemoglobin than without.
Oxygenated leghaemoglobin increases ATP content and the ATP/ADP ratio,
which is counteracted by the P-450 inhibitor N-phenylimidazole. Ferric leghae-
moglobin and other oxidized ferric haemoproteins are without any effect. Sever-
al explanations have been put forward. Leghaemoglobin might act as an oxygen
carrier enabling a facilitated diffusion of oxygen to the bacteroids. This function
might be especially important as the diffusion barrier in the cortex and respira-
tion in the cortex already decreased the oxygen concentration around the bacter-
oids. In this way it is difficult to understand why ferrioxy-Ieghaemoglobin should
be inactive. Another explanation claims that leghaemoglobin acts as an O 2
buffer and establishes the optimal O 2 concentration for nitrogenase derepression
and activity while still sustaining enough respiratory activity. It cannot yet be
excluded that leghaemoglobin still has a more specific function in direct relations
with certain components of the high affinity respiratory pathway which leads
to energization of the bacteroid membrane. Finally it is interesting to make
a comparison with the non-leguminous Frankia nodules. With the exception
of one positive claim for Casuarina by DAVENPORT in 1960, no haemoglobin
has ever been demonstrated, though the haem content may be elevated (refer-
ences in QUISPEL 1974). Moreover no diffusion barrier to oxygen is present
in the cortex (TmPKEMA in GORDON et al. 1979). Regulation of O 2 therefore
is far less developed, though the activity of nitrogenase shows that some un-
known protective mechanism must be active.

5.7 Hydrogen Production and Hydrogen Uptake

Production of hydrogen (H2) is a natural consequence of nitrogenase activity.

This ATP-dependent H2 production is a waste of energy which is used for
11.2 Dinitrogen-Fixing Symbioses 319

reduction of H+ ions, while it could have been used for N2 reduction. DIXON
in 1972 was the first who drew the attention to the possible role of a conventional
hydrogenase in re-utilizing this H2 in an oxidative process which again yields
ATP and thus makes the nitrogenase reaction more efficient. SCHUBERT and
EVANS (1976) therefore introduced the concept of relative efficiency as the factor:

1- H2 evolved 1 rate of H2 evolution in air

or - - - - - - - " - - - - - - - - -
C 2H 2 reduced rate ofH 2 evolution in Ar-COz-0 2

It appeared that in different leguminous symbiotic combinations this factor

varied from 0.20 (in· Trifolium subterraneum) to 0.99 (in cowpea). Very high
values were obtained for most actinorhizas, with the exception of Ceanothus.
These differences were caused by the bacterial partner and were correlated with
similar results in the cultivated N 2-fixing strains of R. japonicum. Oxidation
of H2 by hydrogenase could indeed sustain nitrogenase activity and thus par-
tially could regain the lost energy. Hydrogenase is an inducible enzyme which
thus can be induced by the H2 produced by nitrogenase. This induction is
repressed by succinate, and this repression is in its tum counteracted by cAMP
(LIM and SHANMUGAM 1979).
Up till now it remains to be decided whether hydrogenase plays a really
important role in determining the efficiency of N 2-fixation in the field as com-
pared with other factors. For more detailed discussions and references we refer
to EVANS et al. (in NEWTON and ORME-JOHNSON 1980).

6 Root Nodules as Part of the Whole Plant

In all symbiotic N 2 fixing plant systems the micro-symbiont fixes N 2 under

circumstances where the NH3 formed is hardly or not at all used for the micro-
symbiont itself, but is for the most part transported to the host. On the other
hand, the micro-symbionts depend on the host for organic substrates for produc-
tion of energy and reducing power. These organic substrates must, moreover,
produce the C skeletons for amino acid formation. During nodule formation
organic substrates are essential for growth of the nodules and the development
of the Nz-fixing systems. These organic substrates provide energy, C and N,
and it is well possible that even in the fully developed bacteroids organic N
substances from the host are still necessary for nitrogenase synthesis if the
bacteroids are unable to assimilate the NH3 formed. Therefore root nodules
must be considered as integrated parts of the whole plant in which' growth
and N 2 'fixation are regulated and coordinated by the plant (Fig. 6).
This integration is achieved by the presence of nodular vascular bundles.
Markedly developed transport cells around these vascular bundles indicate the
important role of transport processes either towards or from the nodules system.
The functioning of this transport system has been extensively studied by
PATE and co-workers, who calculated flow-sheets for C and N during different
phases of nodule development in peas, cowpea and lupin (e.g. PATE et al. 1979)
320 A. QUISPEL:

transport transport
from to
phloem xylem

plant cell
plasmalemma

plant cell
cytoplasm

peri bacteroid
membrane
leghemo-
intersymbiontic space globin
bacteroid
outer membrane

bacteroid membrane ==I==9F~~~?=~:;:=I=+~F==:F==F

cytoplasm

bacteroid cytoplasm

Fig. 6. Highly simplified scheme of the main interactions between root nodule cells and
bacteroids at the interface between bacteroid and plant cell cytoplasm in its relation
to the long-distance transport. Disputed aspects like the use of sugar by bacteroids,
the formation of amino acids inside the bacteroids and the localization of leghemoglobin
are indicated by dotted line

(Fig. 7). We refer to the contribution by PATE in Volume 1, this Series (1975).
Dependence on the flow of photosynthetic carbohydrates through the phloem
may be direct or indirect after a previous formation of root reserve carbohy-
drates. There are several indications that this flow of phloem carbohydrates
forms the main limiting factor for nodule growth and N 2 fixation. Studies
on nodulation and N 2 fixation during growth and development of leguminous
and non-leguminous plants have been made by many investigators (e.g. PATE
and co-workers, as above, LAWRIE and WHEELER 1975, BETHLENFALVAY and
PmLLIPS 1978, RYLE et al. 1979). .
II.2 Dinitrogen-Fixing Symbioses 321

SHOOT GROWTH
® ..........

SUGARS:
COa
DOWNWARD- --
ATMOSPHERE
TRANSPORT AMINO ACIDS:
@ ---UPWARD c

i
TRANSPORT ~

Fig. 7. The nodule as part of the whole plant transport system with data on the C-flow
in percentages of the daily C increase by photosynthesis. (Data from MINCHIN and PATE
1973)

The general pattern emerges that N 2 fixation per plant as well as per nodule
increases during growth till the flowering period, after which a decrease may
be observed during fruit and seed formation. Removal of flower buds may
prevent this decrease. Though there are marked differences between different
species and environmental circumstances, this general pattern is easily under-
stood by considering root nodules as sinks in the phloem transport, system
which competes with other sinks such as growing fruits. Interruption of the
phloem transport by excision of nodulated roots or stem-ringing leads to a
rapid decrease and even a complete annihilation of N2 fixation (see QUISPEL
in QUISPEL 1974).
Rapid reductions in N 2 fixation are obtained by darkness, shading or remov-
al of leaves.
322 A. QUISPEL:

Increase of photosynthesis has been obtained by increasing the CO 2 supply

to the leaves or by decreasing the O 2 concentration to reduce photorespiration
(HARDY and HAVELKA in NEWTON and NYMAN 1976). Both methods lead
to increased N 2 fixation which indicates that under normal conditions N 2
fixation is limited by photosynthesis.
The conclusion that N 2 fixation in root nodules is primarily limited by
the direct flow of assimilates appears to be in contradiction to the conclusion
based on C-flow sheets that in lupin root, respiration relied heavily on earlier
received carbon in the reserve carbohydrates. This might be explained by assum-
ing that the bacteroid tissue is more directly linked to the flow of assimilates
in the phloem· than the remaining parts of the root or nodule tissue which
is responsible for most of the total nodule and root respiration. As the latter
respiration is not affected in the same degree by darkening, excision or stem-
ringing as N2 fixation or N assimilation, it has been suggested that either a
special part of respiration is linked to N 2 fixation or N assimilation, or that
these processes have a lower affinity for A TP and thus are more susceptible
to reduction in ATP production.
The well-know inhibiting effect of NHt has been explained as well by a
competition for ATP, in a diversion of energy from the Nrfixing system to
the NHt assimilating reactions (HoUWAARD 1978). This is in accordance with
the general observation that the inhibiting effect of NHt may be counteracted
by increased photosynthesis (WILSON 1940).
All considerations of the relation between root and nodule respiration with
nodulation and N 2 fixation must take due account of the existence of different
pathways of respiration in plants. LAMBERS recently observed that in many
different plant tissues a CN-insensitive respiration is found which is not con-
nected with energy production. He considers this part of respiration as an over-
flow system. If more energy is needed, respiration shifts from this overflow
to the normal, energy-producing system. Such shifts occur during development
and when N 2 is the main N source (DE VISSER and LAMBERS 1979).
Such considerations may be helpful in elucidating the controversies about
the differences in energy consumption between N 2 and NO; -utilizing organisms.
While some investigators did not observe a difference in energy demands on
theoretical grounds or experimental observations, other authors arrived at the
conclusion that N 2 fixation is a less efficient, more energy-demanding process
(SILSBURY 1977). RYLE et al. (1979), observed a higher respiration of nodulated
plants of about 10%-15%, as compared to equivalent non-nodulated plants
on NO;. As this difference is independent of efficiency of N 2 fixation, it must
be more related to the respiration of root nodules. A better insight will be
obtained if such observations are analyzed with regard to eventual non-energy-
producing overflow types of respiration. There are no indicatiol).s that the re-
moval of NH3 from the N 2 fixing nodule parts plays an important role in
the regulation of N 2 fixation.
The general picture emerges of a plant system in which the flow of assimilates
to the root nodules is the main regulating system. The efficient use of these
assimilates depends on the efficiency of the respiratory system and on the relative
efficiency of the bacteroids which may be related to the H 2-recycling possibilities.
11.2 Dinitrogen-Fixing Symbioses 323

7 Concluding Remarks

The different symbiotic systems described in this chapter offer a range of evolu-
tionary adaptations. Starting with loose associations in which N 2-fIxing bacteria
may enter cortex cells, the symbioses with Cyanobacteria give us examples of
inter- and intracellular organs which for the main part were formed before
the invasion with the N 2-fIxing bacteria. In the real root nodule symbioses
the specialized organs are formed under the influence of the infecting microsym-
biont. There are several reasons like the less-developed O 2 protection system
to consider the actinorhizas as a less-adapted system than the legume-Rhizobium
symbioses. The complicated interactions in these symbioses are schematically
summarized in Fig. 6. In all symbiotic systems N 2 fIxation is directed to the
benefIt of the host plant. In Cyanobacteria this direction is based on phenotypic
reactions by the host plant, as the bacteria themselves are able to use N2 under
free-living conditions. In Rhizobium the adaptation must moreover have a genet-
ic base as the free-living bacteria profIt hardly or not at all from their N 2-fIxing
capacities.
The highly complicated interactions involved in nodule formation and sym-
biotic N 2 fIxation must be still far better resolved to understand why they are
limited to specialized bacteria and specialized plants and before we can dream
of possibilities to extend symbiotic N 2 fIxation to other plants (HOLLAENDER
1977). Without any doubt the further studies on symbiotic interactions leading
to nodule formation and N 2 fIxation, together with genetic research and selec-
tion of plant and bacterial germ plasm will lead to the most efficient systems
under different environmental situations and therewith a better use of these
extremely important systems for N enrichment in the biosphere.

References

Albersheim P, Anderson-Prouty A (1975) Carbohydrates, proteins, cell surfaces and the

biochemistry of pathogenesis. Annu Rev Plant Physiol 26: 31-52
Allen EK, Allen ON (1958) Biological aspects of symbiotic nitrogen fixation. In: Ruhland
W (ed) Encyclopedia of plant physiology, vol X. Springer, Berlin G6ttingen Heidel-
berg
Angulo-Carmona A (1974) La formation des nodules fixateurs d'azote chez Alnus glutin-
osa (L.). Vill. Acta Bot Neerl23:257-303
Avissar YT, Nadler KD (1978) Stimulation of tetrapyrrole formation in Rhizobiumjaponi-
cum by restricted aeration. J Bacteriol 135: 782-789
Baker D, Torrey JG, Kidd GH (1979) Isolation by sucrose-density fractionation and
cultivation in vitro of Actinomycetes from nitrogen-fixing root nodules. Nature
(London) 281: 76-78 '
Becking JH, Boer WE de, Houwink AL (1964) Electron microscopy of the endophyte
of Alnus glutinosa. Antonie van Leeuwenhoek J Microbiol Serol 30: 343-376
Bednarski MA, Reporter M (1978) Expression of rhizobial nitrogenase: influence of
plant cell-conditioned medium. Appl Environ Microbiol 36: 115-120
Bergersen FJ, Turner GL (1975) Leghemoglobin and the supply of O 2 to nitrogen-fixing
root nodule bacteroids: presence of two oxidase systems and ATP production at
low free O 2 concentration. J Gen Microbiol91 :345-354
324 A. QUISPEL:

Bergersen FJ, Turner GL (1978) Activity of nitrogenase and glutamine synthetase in

relation to availability of oxygen in continuous cultures of a strain of cowpea Rhizo-
bium sp. supplied with excess ammonium. Biochim Biophys Acta 538:406-416
Bergey's Manual of determinative bacteriology (1974) 8th edn. Williams and Wilkins,
Baltimore
Beringer JE, Beynon JL, Buchanan-Wollaston AV, Johnston AWB (1978) Transfer of
the drug-resistance transposon Tn5 to Rhizobium. Nature 276:633-634
Bethlenfalvay GJ, Phillips DA (1978) Interactions between symbiotic nitrogen fixation,
combined-N application and photosynthesis in Pisum sativum. Physiol Plant
42:119--123
Bishop PE, Guevara JG, Engelke JA, Evans HJ (1976) Relation between glutamine
synthetase and nitrogenase activities in the symbiotic association between Rhizobium
japonicum and-Glycine max. Plant PhysioI57:542-546
Bisseling T, Bos RC v d, Kammen A v (1978) The effect of ammonium nitrate on
the synthesis of nitrogenase and the concentration of leghemoglobin in pea root
nodules induced by Rhizobium leguminosarum. Biochim Biophys Acta 539: 1-11
Bisseling T, Bos RC v d, Weststrate MW, Hakkaart MJJ, Kammen A v (1979) Develop-
ment of the nitrogen-fixing and protein-synthesizing apparatus of bacteroids in pea
root nodules. Biochim Biophys Acta 562:515-526
Blom J (1980) Utilization of fatty acids and NHt by Frankia Avcl1. FEMS Microbiol
Lett 10:143-145
Bogers RJ, Hooijmans-Klappe HTM, Libbenga KR (1976) The first cell cycle in explants
from the mature root cortex of 7-day-old Pisum sativum. Plant Sci Lett 6:43-48
Bowles DJ, Kauss H (1976) Characterization, enZymatic and lectin properties of isolated
membranes of Phaseolus aureus. Biochim Biophys Acta 443:360-376
Brown CM, Dilworth MJ (1975) Ammonia assimilation by Rhizobium cultures and bac-
teroids. J Gen Microbiol 86: 39--48
Brussel van AAN, Planque K, Quispel A (1977) The wall of Rhizobium leguminosarum
bacteroid and free-living form. J Gen Microbioll0l :51-56
Brussel van AAN, Costerton JW, Irvin RF, Korevaar CGN (1978) Localization studies
in Rhizobium leguminosarum Abstract 115. In: Proc Steenbock-Kettering Int Symp
Nitrogen Fixation, Madison
Brussel van AAN, Costeron JW, Child JJ (1979) Nitrogen fixation by Rhizobium sp.
32 Hi a morphological and ultrastructural comparison of asymbiotic and symbiotic
nitrogen-fixing forms. Can J MicrobioI25:352-361
Buchanan-Wollaston V (1979) Generalized transduction in Rhizobium leguminosarum.
J Gen Microbiol112: 135-142
Carlson RW, Sanders RE, Napoli C, Albersheim P (1978) Host-symbiont interactions.
III Purification and partial characterization of Rhizobium lipopolysaccharides. Plant
Physiol 62: 912-917
Chandler MR (1978) Some observations on infection of Arachis hypogaea L. by Rhizo-
bium. J Exp Bot 29: 749--755
Chen AP, Phillips D (1976) Attachment of Rhizobium to legume roots as the basis for
specific interactions. Physiol Plant 38: 83-88
Child JJ, Kurz WGW (1978) Inducing effect of plant cells on nitrogenase activity by
Spirillum and Rhizobium in vitro. Can J MicrobioI24:143-148
Criswell JG, Havelka VD, Quebedoux B, Hardy RWF (1976) Adaptation of nitrogen
fixation by intact soybean nodules to altered rhizosphere p02' Plant Physiol
58:622-625
Dandekar AM, Vasavadi HA, Modi VV (1978) Partial purification of' a competence
factor from Rhizobiumjaponicum. Arch Microbiol118 :253-256
Dazzo FB, Brill WJ (1979) Bacterial polysaccharide which binds Rhizobium trifolii to
clover root hairs. J Bacteriol127: 1362-1373 -
Dazzo B, Hubbell DH (1975) Cross-reactive antigens and lectin as determinants of sym-
biotic specificity in the Rhizobium clover association. Appl Microbiol 30: 1017-1033
Dazzo FB, Napoli CA, Hubbell DH (1976) Adsorption of bacteria to roots as related
to host specificity in the Rhizobium-clover syinbiosis. Appl Environm Microbiol
32:166-171
II.2 Dinitrogen-Fixing Symbioses 325

Dijk C van (1978) Spore formation and endophyte diversity in root nodules of Alnus
glutinosa (L.) Vill. New Phytol 81: 601-615
Dijk C van, Merkus E (1976) A microscopical study of the development of a spore-like
stage in the life cycle of the root-nodule endophyte of Alnus glutinosa (L.) Gaertn.
New Phytol 77:73-91
Dobereiner J, Burris RH, Hollaender A (1977) Limitations and potentials for biological
nitrogen fixation in the Tropics. Plenum, New York London
Dullaart J (1970) The bioproduction of indole-3-acetic acid and related compounds in
root nodules and roots of Lupinus luteus L. and by its rhizobial symbiont. Acta
Bot Neerl19: 573-615
Dullaart J, Duba LI (1970) Presence of gibberellin-like substances and their possible
role in auxin bioproduction in root nodules and roots of Lupinus luteus L. Acta
Bot Neerl19: 877-883·
Egeraat AWSM v (1972) Pea root exudates and their effects on root nodule bacteria.
Meded Landbouwhogesch Wageningen 72-127
Evans WR, Keister DL (1976) Reduction of acetylene by stationary cultures of free-living
Rhizobium sp. under atmospheric oxygen levels. Can J MicrobioI22:949-952
Fogg GE, Stewart WOP, Fay P, Walsby AE (1973) The blue-green algae. Academic
Press London New York
Fred EB, Baldwin IL, McCoy E (1932) Root nodule bacteria and leguminous plants.
U niv Wisconsin Press, Madison
Gibson AH, Newton WE (1981) Current perspectives in nitrogen fixation. Aust Acad,
Sci Canberra
Gibson AH, Scowcroft WR, Child 11, Pagan JD (1976) Nitrogenase activity in cultured
Rhizobium sp. strain 32 H. Nutritional and physical considerations. Arch Microbiol
108:45-54
Glenn AR, Dilworth MJ (1979) An examination of Rhizobium leguminosarum for the
production of extracellular and periplasmic proteins. J Gen MicrobioI112:405-409
Gordon J, Wheeler CT, Perry DA (eds) (1979) Symbiotic nitrogen fixation in the manage-
ment of temperate forests. For Res Lab, Oregon State Univ
Greshoff PM, Skotnicki ML, Eady JF, Rolfe BG (1978) Viability of Rhizobium trifolii
bacteroids from clover root nodules. Plant Sci Lett 10: 299-304
Hanus FJ, Maier RJ, Evans H (1979) Autotrophic growth of H 2 -uptake positive strains
of Rhizobium japonicum in an atmosphere supplied with hydrogen gas. Proc Natl
Acad Sci USA 76:1788-1792
Hardy RWF, Gibson AH (1977) A treatise on dinitrogen fixation, vol IV. Wiley and
Sons, New York
Hardy RWF, Silver WS (1977) A treatise on dinitrogen fixation, vol III. Wiley and
Sons, New York
Henson LE, Wheeler CT (1977 a) Hormones in plants bearing nitrogen-fixing root nod-
ules, gibberellin-like substances in Alnus glutinosa (L) Gaertn. New Phytol 78: 373-382
Henson LE, Wheeler CT (1977b) Hormones in plants bearing nitrogen-fixing root nod-
ules, distribution and seasonal changes in levels of cytokinins in Alnus glutinosa (L)
Gaertn. J Exp Bot 28:205-214
Hollaender A (ed) (1977) Genetic engineering for nitrogen fixation. Plenum, New York
London
Hollander JA de, Stouthamer AH (1979) Multicarbon-substrate growth of Rhizobium
trifoW. FEMS Microbiol Lett 6: 57-59
Houwaard F (1978) Influence of ammonium chloride on the nitrogenase activity ofnodu-
lated pea plants (Pisum sativum). Appl Environm MicrobioI35:1061-1065
Hubbell DH, Morales VM, Umali-Garcia M (1978) Pectolytic enzymes in Rhizobium.
Appl Environm MicrobioI35:21Q-213
Johnston AWB, Beynon JL, Buchanan-Wollaston AV, Setchell SM, Hirsch PR, Beringer
JE (1978) High frequency transfer of nodulating ability between strains and species
of Rhizobium. Nature 276:634--636
Keister DL, Evans WR (1976) Oxygen requirements for acetylene reduction by pure
cultures of rhizobia. J BacterioI129:149-153
Kennedy IR (1979) Assimilating fixed nitrogen. Proc Austr Biochem Sci 12:Q 15
326 A. QUISPEL:

Kijne JW (1975a) The fine structure of pea root nodules 1. Vacuolar changes after
endocytotic host cell infection by Rhizobium leguminosarum. Physiol Plant Pathol
5:75--79
Kijne JW (1975b) The fine structure of pea root nodules 2. Senescence and disintegration
of the bacteroid tissue. Physiol Plant Pathol 7: 17-21
Kijne JW, Planque K (1979) Ultrastructural study of the endomembrane system in in-
fected cells of pea and soybean root nodules. Physiol Plant PathoI14:339-345
Kijne JW, Schaal CAM van der, Vries GE de (1980) Pea lectins and the recognition
of Rhizobium leguminosarum. Plant Sci Lett 18: 65--74
Kurz WGW, La Rue TA (1977) Citric acid cycle enzymes and nitrogenase in nodules
of Pisum sativum. Can J Microbiol23: 1197-1200
Laane C, Haaker H, Veeger C (1978) Involvement of the cytoplasmic membrane in
nitrogen fixation by Rhizobium leguminosarum bacteroids. Eur J Biochem 87: 147-153
Laane C, Krone W, Konings WN, Haaker H, Veeger C (1979) The involvement of
the membrane potential in nitrogen fixation by bacteroids. FEBS Lett 103: 328--332
Laing WA, Christeller JT, Sutton WD (1979) Carbon dioxide fixation by lupin root
nodules. II. Studies with 14C labeled glucose, the pathway of glucose catabolism
and the effects of some treatments that inhibit nitrogen fixation. Plant Physiol
63:450-454
Lalonde M, Devoe IW (1976) Origin of the membrane envelope enclosing the Alnus
crispa var. mollis Fern. root nodule endophyte as revealed by freeze-etching microsco-
py. Physiol Plant Pathol 8: 123-129
Lawrie AC, Wheeler CT (1975) Nitrogen fixation in the root nodules of Vicia faba
L. in relation to the assimilation of carbon. New Phytol 74:429-445
Legocki RP, Verma DP (1979) A nodule-specific plant protein (Nodulin-35) from
soybean. Science 205:190-193
Lepidi AA, Nuti MP, Bernacchi G, Neglia R (1976) Physiology of the Rhizobium-legume
association. I The presence of bacterial DNA within the plant root. Plant Soil
45: 555--569
Lim ST, Shanmugam KT (1979) Regulation of hydrogen utilization in Rhizobiumjaponi-
cum by cyclic AMP. Biochim Biophys Acta 584:479-492
Lim ST, Hennecke H, Scott DB (1979) Effect of guanosine-3'-5'-cyclic monophosphate
on nitrogen fixation in Rhizobium japonicum. J Bacteriol 139: 256-263
Ljunggren H (1969) Mechanism and pattern of Rhizobium invasion into leguminous
root hairs. Physiol Plant Suppl 5
Ludwig RA (1978) Control of ammonium assimilation in Rhizobium 32 H1. J Bacteriol
135:114-123
Maier RJ, Brill WJ (1978) Mutant strains of Rhizobiumjaponicum with increased ability
to fix nitrogen for soybean. Science 201 : 448--449
Mennes AM (1973) The indole-3-acetic acid oxidase of Lupinus luteus L. Acta Bot Neerl
22:694-729
Miflin BJ, Lea PJ (1977) Amino acid metabolism. Annu Rev Plant PhysioI28:299-329
Nadler KD, Avissar YJ (1977) Heme synthesis in soybean root nodules. I. On the
role of bacteroid {) aminolevulinic acid synthase and aminolevulinic dehydrase in
the synthesis of the heme of leghemoglobin. Plant Physiol 60: 433-436
Nash DT, Schulman HM (1976) Leghemoglobin and nitrogenase activity during soybean
root nodule development. Can J Bot 54:2790-2797
Newcomb W (1976) A correlated light and electron microscopic study of symbiotic growth
and differentiation in Pisum sativum root nodules. Can J Bot 54:2163-2186
Newcomb W, Syono K, Torrey JG (1977) Development of ineffective pea root nodule:
morphogenesis, fine structure and cytokinin biosynthesis. Can J Bot 55: 1891-1907
Newton WE, Nyman CJ (eds) (1976) Proc 1st Int Symp Nitrogen Fixation. Wash State
Univ Press
Newton WE, Orme-Johnson NR (eds) (1980) 3rd Int Conf Nitrogen Fixation. Univ
Park Press, Baltimore
Newton WE, Postgate JR, Rodriguez-Barrueco C (1977) Recent developments in nitro-
gen fixation. Academic Press, London New York
Nuti MP, Lepidi AA, Prakash RK, Schilperoort PA, Cannon FC (1979) Evidence for
IL2 Dinitrogen-Fixing Symbioses 327

nitrogen fixation (nif) genes on naturally occurring plasmids in Rhizobium. Nature

282:533--535
Nutman PS (ed) (1976) Symbiotic nitrogen fixation. Int BioI Programm, vol VII. Cam-
bridge Univ Press, Cambridge
O'Gara F, Shanmugam KT (1976) Regulation of nitrogen fixation by rhizobia, export
of fixed N2 as NHt. Biochim Biophys Acta 437:313--332
Oostrom H, Treurniet FE, Mennes AM (1975) Cytochemical localization of peroxidase
during the development of root nodules of Pisum sativum L. Z Pflanzenphysiol
74:451-463
Paau AS, Oro J, Cowles JR (1979) DNA content of free-living rhizobia and bacteroids
of various Rhizobium-legume associations. Plant Physiol 69: 402-405
Pankhurst CE, Craig AS (1978) Effect of oxygen concentration, temperature, and com-
bined nitrogen on the morphology and nitrogenase activity of Rhizobium spp. strain
32H1, in agar culture. J Gen MicrobioI106:207-219
Pate JS, Layzell DB, MacNeil DL (1979) Modelling the transport and utilization of
carbon and nitrogen in a nodulated legume. Plant Physiol 63: 730-739
Phillips DA, Daniel RM, Appleby CA, Evans HJ (1973) Isolation from Rhizobium of
factors which transfer electrons to soybean nitrogenase. Plant Physiol51: 136-138
Planque K, Vries GE de, Kijne JW (1978) The relation between nitrogenase and glutamine
synthetase levels in bacteroids of Rhizobium leguminosarum of various age. J Gen
Microbiol 106: 173--178
Planque K, Nierop JJ van, Burgers A, Wilkinson SG (1979) The lipopolysaccharide
of free-living and bacteroid forms of Rhizobium leguminosarum. J Gen Microbiol
110:151-159
Puppo A, Rigaud J (1978) Cytokinins and morphological aspects of French-bean roots
in the presence of Rhizobium. Physiol Plant 42: 202-206
Quispel A (ed) (1974) The biology of nitrogen fixation. Elsevier, North-Holland Amster-
dam Oxford New York
Quispel A (1976) Symbiotic nitrogen fixation in non-leguminous plants V. The growth
requirements of the endophyte of Alnus glutinosa. Acta Bot Neerl9:380-396
Quispel A, Tak T (1978) Studies on the growth of the endophyte of Alnus glutinosa
(L.). Vill in nutrient solutions. New Phytol 81 :587-600
Ranga Rao V, Darrow RA, Keister DL (1978) Effect of oxygen tension on nitrogenase
and glutamine synthetases I and II in Rhizobiumjaponicum 61 A 76. Biochem Biophys
Res Commun 81 :224--231
Robertson JG, Warburton MP, Farnden KJF, Banks JM (1975) Induction of glutamine
synthetase during nodule development in lupin. Austr J Plant Physiol 2: 265--272
Robertson JG, Lyttleton P, Bullivant S, Grayston GF (1978a) Membranes in lupin
root nodules. L The role of Golgi bodies in the biogenesis of infection threads and
peribacteroid membranes. J Cell Sci 30: 129-149
Robertson JG, Warburton MP, Lyttleton P, Fordyce AM, Bullivant S (1978b) Mem-
branes in lupin root nodules. II Preparation and properties of peribacteroid mem-
branes and bacteroid envelope inner membranes from developing lupin nodules. J
Cell Sci 30:151-174
Ronson CW, Primrose SB (1979) Carbohydrate metabolism in Rhizobium trifolii: identifi-
cation and symbiotic properties of mutants. J Gen MicrobioI112:77-88
Ryle GJA, Powell CE, Gordon AJ (1979) The respiratory costs of nitrogen fixation
in soybean, cowpea and white clover. J Exp Bot 30:135--153
Schaede R (1976) Die pflanzlichen Symbiosen. Fischer, Stuttgart
Scott DB, Farnden KJF, Robertson JGF (1976) Ammonia assimilation in lupin nodules.
Nature 263: 703--706
Shaw BD, Sutton WD (1977) Polypeptide synthesis by Rhizobium bacteroids and bacteria.
Biochim Biophys Acta 563:216-226
Silsbury JH (1977) Energy requirements of symbiotic nitrogen fixation. Nature
267:149-150
Skotnicki ML, Rolfe BG (1978) Differential stimulation and inhibitional growth of Rhizo-
bium trifolii strain T1 and other Rhizobium species by various carbon sources. Micro-
biology 20: 15--28
328 A. QUISPEL:

Solheim B, Raa J (1977) Cell wall biochemistry related to specificity in host-plant patho-
gen interactions. Univ Forlaget, Oslo 1977
Stewart WDP, Rodgers GA (1977) The cyanophyte-hepatic symbiosis. 2. Nitrogen fixa-
tion and the interchange of nitrogen and carbon. New Phytol 78:45(}-471
Stovall I, Cole M (1978) Organic acid metabolism by isolated Rhizobiumjaponicum bacter-
oids. Plant Physiol 61 : 787-790
Strand R, Laetsch WM (1977) Cell and endophyte structure of the nitrogen fixing root
nodules of Ceanothus integerrimus H and A. Protoplasma 93:165-190
StripfR, Werner D (1978) Differentiation of Rhizobiumjaponicum II. Enzymatic activities
in bacteroids and plant cytoplasm during the development of nodules of Glycine
max. Z Naturforsch 33c:373--381
Sutton WD, Bos van den RC, Bisseling T (1978) The DNA content of Rhizobium (Strain
NZP 2257) bacteroids and bacteria. Plant Sci Lett 12: 145-149
Syono K, Newcomb W, Torrey JG (1976) Cytokinin production in relation to the develop-
ment of pea root nodules. Can J Bot 54:2155-2162
Te May Ching, Hedtke S, Newcomb W (1977) Isolation of bacteria, transforming bacteria
and bacteroids from soybean nodules. Plant Physiol 60: 771-777
Tjepkema JD, Evans HJ (1975) Nitrogen fixation by free-living Rhizobium in a defined
liquid medium. Biochem Biophys Res Commun 65: 625-628 '
Tjepkema JD, Yocum CS (1974) Measurement of oxygen partial pressure within soybean
nodules by oxygen micro electrodes. Planta 119: 351-360
Tjepkema JD, Ormerod W, Torrey JG (1980) On vesicle formation and in vitro acetylene
reduction by Frankia. Nature 287: 633--635
Torrey JG, Tjepkema JD (eds) (1979) Symbiotic nitrogen fixation in actinomycete-nodu-
lated plants. Bot Gaz Suppl140
Trinick MJ (1973) Symbiosis between Rhizobium and the non-legume Trema aspera.
Nature 244: 45(}-460
Trinick MJ (1979) Structure of nitrogen-fixing nodules formed by Rhizobium on roots
of Parasponia andersonii Planck. Can J Microbiol 25: 565- 578
Trinick MJ, Galbraith J (1976) Structure of root nodules formed by Rhizobium on the
non-legume Trema cannabina var. scabra. Arch Microbiol 108: 15(}-166
Tu TC (1979) Alterations in the membranes of bacteroidal cells in soybean root nodules
as revealed by freeze-fracturing. Physiol Plant PathoI15:35-41
Veeger C, Laane C, Scherings C, Zeeland-Wolbers L van (1978) Membrane energization
in relation with nitrogen fixation in Azotobacter vinelandii and Rhizobium legumino-
sarum. Biochimie 60: 237-243
Verma DPS (1980) Expression of host genes during symbiotic nitrogen fixation. In:
Leaver CJ (ed) Genome organization and expression in plants. Plenum, New York
London
Verma DPS, Bal AK (1976) Intercellular site of synthesis and localization of leghemoglo-
bin in root nodules. Proc Nat! Acad Sci USA 73:3843--3847
Vincent JM (1970) A manual for the practical study of root nodule bacteria. IBP Hand-
book vol 15. Blackwell, Oxford Edinburgh
Virtanen AI (1947) The biology and chemistry of nitrogen fixation by legume bacteria.
BioI Rev 22: 23(}-269
Visser R de, Lambers H (1979) Root respiration of Pisum sativum L. Activity of the
alternate pathway and relation to N 2-fixation. Plant Physiol 63 (Suppl): 482
Vries GE de (1980) Transport of sugars and organic acids in Rhizobium leguminosarum
and their role in symbiosis. Thesis, Univ Leiden
Werner D (1978) Differentiation of Rhizobium japonicum III. Inhibition of nitrogenase
derepression by chloramphenicol and rifampicin concentrations not inhibiting growth.
Z Naturforsch 33c:85(}-862
Werner D, Morsche1 E (1978) Differentiation of nodules of Glycine max. Ultrastructural
studies of plant cells and bacteroids. Planta 141 : 16(}-177
Werner D, StripfR (1978) Differentiation of Rhizobiumjaponicum. 1. Enzymatic compari-
son of nitrogenase repressed and derepressed living cells and of bacteroids. Z Natur-
forsch 33c:245-252
11.2 Dinitrogen-Fixing Symbioses 329

Werner D, Morschel E, Stripf R, Winchenbach B (1980) Development of nodules of

Glycine max infected with an ineffective strain of Rhizobium japonicum. Planta
147:320-329
Whiting MJ, Dilworth MJ (1974) Legume root nodule nitrogenase purification, properties
and studies on its genetic control. Biochim Biophys Acta 371: 337-351
Wilcockson J, Werner D (1979) Organic acids and prolonged nitrogenase activity by
non-growing free-living Rhizobiumjaponicum. Arch Microbiol122: 153-160
Wilson PW (1940) The biochemistry of symbiotic nitrogen fixation. Dniv Wisconsin
Press, Madison
Wooi KC, Broughton WJ (1979) Isolation and metabolism of Vigna unguiculata root
nodule protoplasts. Planta 145:487-495
Yao PY, Vincent JM (1976) Factors responsible for the curling and branching of clover
root hairs by Rhizobium. Plant Soil 45: 1-16
ll.3 Dinitrogen Fixation in Rhizosphere
and Phyllosphere Associations
J. DOBEREINER

1 Introduction

Among the agricultural systems which contribute biologically fixed nitrogen,

the symbiotic legume associations described in Chapter 11.2 are the most impor-
tant. With the steadily increasing prices of nitrogen fertilizer however, other
less perfect N 2-fixing systems are becoming more and more important. Members
of the family Gramineae are here of major interest because cereals are the
most important food basis all over the world, especially in developing countries.
Fortunately it is in the tropics and subtropics, where most poor people live,
that chances are best for supplying large parts of the nitrogen needs for produc-
tive agricultural systems through biological fixation. Not only are soil tempera-
tures during most of the year suitable for microbial activities and plant growth,
hqt also many tropical plants have developed a specific photosynthetic pathway
(C 4 -dicarboxylic acid pathway) which enables them to convert sun energy twice
as effectively as temperate species. Such plants can more easily devote part
of their photosynthates to nitrogen fixation. The major tropical forage grasses
and cereals, except rice, are in this group.
Nitrogen fixation has been shown in many forage grasses and grains all
over the world, but substantial amounts were found under tropical conditions
only. Table 1 summarizes a few relevant data obtained from nitrogen balance
studies. Many other results reported in the literature (e.g. DOBEREINER 1978,
NEYRA and DOBEREINER 1977) were obtained by the indirect acetylene reduction
assay and are in the same range. Except for sugar cane (see Sect. 2.1) very
little or no nitrogen fixation has been found in soil without roots (DOBEREINER
et al. 1972b; LEE et al. 1977a).

2 Characterization of Rhizocoenoses

The identification of an association between N 2-fixing bacteria and plant roots,

termed rhizocoenosis (Intern. Workshop Associative N2 Fixation, Piracicaba,
1979) requires demonstration of plant-bacteria interactions. The simple isolation
of one or several species of N 2-fixing bacteria from the rhizosphere is insufficient
even if numbers of bacteria are high. The description of such systems in this
chapter has therefore been limited to associations where plant-bacteria interac-
tions have been shown.
11.3 Dinitrogen Fixation in Rhizosphere and Phyllosphere Associations 331

Table 1. Nitrogen gains in grass systems as estimated by nitrogen balance studies

Country Plant species Period N gain Observation Reference

or regions (kg ha- 1)

Temperate regions
U.S.A. Grassland 82 years 3,240 Field expo WHITE et al.
(pennsylvania) 1945
England Continuous 72 years 2,788 Broadbalk JENKINSON 1973
wheat field expo

Subtropical regions
U.S.A. (Texas) Grassland 1 year 22-34 Field expo SMITH et al. 1954
U.S.A. Zoysia grass 2 years 147 Pots with GIDDENS 1977
(Georgia) (z. matrella) 9 kg soil
U.S.A. Bahia grass 2 years 183 Pots with GIDDENS 1977
(Georgia) (Paspalum 9 kg soil
notatum)
Australia Rye grass 3 years 185 Field expo PARKER 1957
(Lolium
rigidum)

Tropical regions
Nigeria Stargrass 3 years 269 Field expo JAIYEBO and
(Cynodon plecto- MOORE 1963
stachtris)
Nigeria Taurbagrass 4 months 112-146 Pots with MOORE 1963
(Eleusine 2.5 kg soil
caracana)
Rhodesia Varsey grass 4 months 202 Drums with PuRCHASE 1978
(Paspalum 100 kg soil
urvillei)
Philippines Rice 6 crops' 32 Pots with App et al. 1980
(Oryza sativa) 10 kg soil
Philippines Rice 4 crops' 36 Pots with App et al. 1980
(Oryza sativa) 10 kg soil

• Bacterial N2 fixation only, algae were eliminated with black cloth

2.1 Sugar Cane - Beijerinckia

Many years ago, rhizosphere and vegetation effects on Beijerinckia spp. were
shown for sugar cane (DOBERElNER 1961). Almost all (95%) of the soils coIlected
in continuously planted sugar cane fields contained an abundant Beijerinckia
population, while only 62% of soils collected under a number of other types
of vegetations contained this organism and in much lower numbers. Fields
planted with sugar cane for the first time showed steady increases in Beijerinckia
numbers and the soil on the root surface contained 2 times more of these
bacteria than the rhizosphere soil, and 56 times more than the soil between
332 J. DOBEREINER:

rows. Assays of nitrogenase activity in sugar cane fields confirmed that the
major contribution was in the rhizosphere and root surface soil and not in
the roots themselves (DOBEREINER et al. 1972b, RUSCHEL et al. 1978). In addition
rain-water can carry leaf exudates into the soil which enhance Beijerinckia
growth and nitrogenase activities (DOBEREINER 1961, DOBEREINER et al. 1972b).
Sugar cane seedlings exposed to 15N2 showed fixation, incorporation and trans-
location of nitrogen into the leaves (RUSCHEL et al. 1978); stem cuttings are
infected by nitrogen-fixing bacteria which extend into xylem vessels and intercel-
lular spaces of roots of field-grown cane plants (PATRIQUIN et al. 1980). These
organisms have not been identified and their role in sugar cane nitrogen fixation
is not yet known.

2.2 Paspalum notatum - Azotobacter paspali

The association of Paspalum notatum with Azotobacter paspa/i (DOBEREINER

1966) seems unique among grass rhizocoenoses because of its extreme specificity,
comparable with the most specific Rhizobium symbioses. Among 31 cultivars
or ecotypes of this species only the cv. 'batatais' associates with A. paspa/i
and 98% of a large number (252) of root surface soil samples from this cultivar
collected all over Brazil and in several other countries contained the bacteria
(DOBEREINER 1970). Only 3% of the samples from the remaining cultivars and
none of 200 samples of other grasses and legumes contained the bacteria. Due
to the nature of Paspalum as a useless invasion grass, this most interesting
rhizocoenosis has unfortunately not been sufficiently studied. From the informa-
tion available on host plant specificity and also some physiological data (DOBER-
EINER and CAMPELO 1971, DOBERElNER and DAY 1975), it must be concluded
that the plant-bacteria interaction is a very close one. Strong washing of the
roots did not reduce nitrogenase activity, but surface sterilization eliminated
the bacteria. The site of the bacteria was therefore suggested to be the mucigel
layer (DOBEREINER et al. 1972a). A. paspa/i is an aerobic fast-growing bacterium
which can be easily isolated from batatais roots (DOBEREINER 1966), where it
occurs in very high numbers. Although the organism has a very narrow pH
requirement in culture (MACHADO and DOBEREINER 1969), it does occur on
P. notatum roots grown in soil with pH 4.9 (DOBEREINER 1966). Root exudates
from P. notatum stimulate the growth of A. paspa/i in culture, and all A. paspa/i
strains tested so far are unable to reduce nitrate (PENA and DOBEREINER 1974).
Nitrogenase activity in the organism was not inhibited by concentrations of
20 mM NO; and the derepression of nitrogenase in NH; -grown cells was
the same in the presence and absence of 10 mM NO; (DOBEREINER and DAY
1975).

2.3 Azospirillum Rhizocoenoses

Azospirillum associations seem very common in tropical regions. High numbers

of these bacteria were found in rhizosphere soil and roots of many forage
11.3 Dinitrogen Fixation in Rhizosphere and Phyllosphere Associations 333

grasses (e.g. Panicum maximum, Brachiaria spp., Pennisetum purpureum, Digi-

taria spp.), and in maize, millets, sorghum, rice, wheat and rye in the areas
of Central and South Brazil (DOBEREINER et al. 1976; DA SILVA and DOBEREINER
1978, NERY et al. 1977). A survey in India revealed Azospirillum spp. in roots
of rice, sorghum, maize, Panicum and a number of other non-Gtamineae
(LAKSHMI-KURAMI et al. 1976). The organism was also isolated from various
grass roots in Mexico (QUINTERO and GARZA 1978), Australia (BERGERSEN per-
sonal communication), Belgium (REYNDERS and VLASSAK 1976) and the U.S.A.
(SLOGER and OWENS 1976).
The role of Azospirillum in N2 fixation of Digitaria (DoBEREINER and DAY
1976) and maize (VON BULLOW and DOBEREINER 1975) was confirmed by highly
significant correlations of root segment nitrogenase activities with Azospirillum
enrichment culture activities obtained with the same root segments. It is also
supported by the correlation of numbers of Azospirillum within roots with the
nitrogenase activity of maize (MAGALHAES et al. 1979) and rice rwATANABE and
CHOLITKUL 1979) roots during the growth cycle.
Koch's postulates have been established in that sterile maize plants grown
in agar cultures when inoculated with Azospirillum brasilense showed nitrogenase
activity and the organism could be reisolated and reinoculated with the same
results (BURRIS 1977). Similarly, Brachiaria, sorghum and wheat seedlings grown
in sterile vermiculite inoculated with A. brasilense or A. /ipoferum reduced C 2H 2
in intact systems, but only when the partial oxygen pressure on top of the
vermiculite cultures was reduced to less than 0.02 atm. (FREITAS and DOBEREINER
1978). In none of these experiments, however, was the incorporation of fixed
nitrogen into the plant observed.

2.3.1 Taxonomy of Azospirillum spp.

BEIJERINCK (1925) described as Spirillum /ipoferum, N 2-fixing organisms which
he had isolated from muds and soils in Holland, but he was unable to demon-
strate N2 fixation in pure culture. BECKING (1963) showed 15N2 incorporation
with a "Vibrio or Spirillum" which was most probably the same organism.
A large number of strains have now been isolated from all over the world,
two (OKON et al. 1976a) or possibly three groups being identified (SAMPAIO
et al. 1978) on morphological and physiological characteristics. Based on DNA
homology studies with 60 strains, TARRAND et al. (1978) described a new genus,
Azospirillum, having two species, A. /ipoferum, (probably identical with Beijer-
inck's S. /ipoferum), and A. brasilense. The major differential characteristics
are absolute biotin requirement and use of glucose and a-ketoglutarate as sole
carbon sources, and typical large involution forms in older, especially alkaline
cultures in A. /ipoferum.
Both species contain two groups, one which denitrifies (nir+) and another
which does not (nir-). The major characteristics of the two Azospirillum spp.
are summarized in Table 2.
Differentiation between the two species and between the nir groups was
also possible by fluorescent antibody techniques (DE-POLLI et al. 1980).
334 J. DOBEREINER:

Table 2. Characteristics of Azospirillum spp. (TARRAND et al. 1978)

A. /ipoferum A. brasilense

DNA base composition (mol % G+C) 70 70

DNA homology with Sp7 (%) 35.3±5 83±9
DNA homology with Sp59 (%) 74.8±7 42±6
Use of glucose mannitol and ex-ketoglutaric acid as Yes No
C-source
Acid from glucose, fructose, ribose, mannitol Yes No
Pleomorph cells in older especially alkaline Yes No
cultures
Biotin required for growth Yes No
NO; to NO; (nr) Yes Yes
NO; to N 20 (nir) Yes/no Yes/no
N 20 to N2 Yes Yes
Plants found to be infected C 4 plants C 3 plants

2.3.2 Root Infection

In a symbiotic association like that of legumes the first step is selective multipli-
cation and attachment of the bacteria followed by root hair infection and multi-
plication of the infecting bacteria within the roots (see Chap. II.2). Adsorption
of A. brasilense to root hairs and undifferentiated epidermal cells was observed
with sterile Panicum maximum seedlings (UMALl-GARClA et al. 1978). Trypto-
phan is converted by Azospirillum to indole acetic acid (VLASSAK and REYNDERS
1978b), a growth substance involved in root infection of legumes. Modification
of the bacterial form to large cyst-like structures was observed in monoxenic
maize and millet cultures inoculated with some Azospirillum strains and not
with others (RuscoE et al. 1978, UMALl-GARClA et al. 1980). Similar "C-forms"
were reported to be formed as a response to excessive oxygen concentrations
in sugar cane tissue cultures supporting Azospirillum growth (BERG et al. 1980)
(Fig. 1).
Infection of sterile seedlings of maize (BURRIS 1977), Panicum maximum
(UMALl-GARCIA et al. 1978), rice and alfalfa (SUBBA RAo 1981) and Pennisetum
typhoideae callus tissue (VASIL et al. 1977) in intercellular and intracellular spaces
by Azospirillum spp. has been demonstrated. Field-grown plants of Digitaria
decumbens (DOBEREINER and DAY 1976), maize and Panicum maximum (PATRI-
QUIN and DOBEREINER 1978) show very similar tetrazolium-reducing bacteria
localized in intracellular and intercellular spaces of the cortex and within the
stele. Numbers of Azospirillum in surface-sterilized roots increased from 105
in young plants to 10 7 in flowering plants, while the numbers in non-sterilized
roots were high during the whole growth cycle (MAGALHAES et al. 1979). Pro-
nounced variations of nitrogenase activity during the growth cycle, with peak
activities at flowering in maize, rice, sorghum and wheat have been shown
by many authors (VON HOLLOW and DOBEREINER 1975, LEE et al. 1977b, NERY
II.3 Dinitrogen Fixation in Rhizosphere and Phyllosphere Associations 335

Fig. lA, B. Infection of wheat seedlings with Azospirillum brasilense nir-. Plants were
grown in 50-ml test tubes with sterilized vermiculite and basic fertilization with low
nitrogen (20 ppm N as NO;). Inoculant strain SpBr14 was isolated originally from wheat
soil. Nitrogenase activity of 1-month-old plants was 60 nmol C 2 H 4 h - 1 per tube. Hand-
cut sections were mounted in glycerol after 15 h incubation of the intact plant-vermiculite
systems under 0.15% tetrazolium chloride in 0.05 M phosphate buffer. Magnification
A x 250 ; B x 500. Dark stained areas in the intercellular spaces represent A. brasi(ense
cells with formazan crystals due to reduction of the tetrazolium salt

et al. 1977). Surface-sterilized roots of maize (VON BULLOW and DOBERElNER

1975), wheat and rice (BALDANI and DOBEREINER 1980) yielded almost pure
Azospirillum enrichment cultures while non-sterilized roots contained more
mixed populations. Crushing of surface-sterilized maize roots yielded 10 3-10 4
times as many Azospirillum cells as intact roots (OKON et al. 1977 a). All these
336 J. DOBEREINER:

observations confirm the internal nature of these rhizocoenoses (PATRIQUIN and

DOBEREINER 1978). Substrate availability and oxygen conditions would suggest
that N2 fixation within roots is more important and efficient than that ofbacte-
ria situated on the root surface or in soil (BURRIS et al. 1978). In maize (MAGAL-
HAES et al. 1979), wheat (KAVIMANDAN et al. 1978) and rice (WATANABE et al.
1979), the ste,ms as well as the roots were found to be infected with Azospirillum.

2.3.3 Host Plant Specificity

Pathogenic or symbiotic plant bacteria associations which involve root infection
show all degrees of specificity. Cross-inoculation groups in the legume/Rhizo-
bium symbiosis have been discussed in Chapter. II.2. It is therefore not surprising
that a certain degree of host plant specificity has been found in Azospirillum
rhizocoenoses as well (BALDANI and DOBEREINER 1980, DOBEREINER and DE-
POLL! 1980). From a large number of isolates obtained from various grasses
grown in soil containing both species of Azospirillum, it could be concluded
that maize, sorghum, several other C 4 forage grasses and one C 4 Cyperacea
are infected by A. /ipoferum, while the C 3 cereals wheat, barley, oat, rye and
rice are almost exclusively infected by A. brasilense. The only exception is sugar
cane (Table 3). Furthermore most root-infecting Azospirillum strains were not
denitrifying (nir-) (DOBEREINER and DE-POLL! 1980). This discrimination by
the plant between species and nir groups was confirmed by differential immuno-

Table 3. Host plant specificity in the infection of grass roots by Azospirillum spp. in
field- and pot-grown plants. (BALDANI and DOBEREINER 1980, BALDANI et al. 1981,
BODDEY 1980)

Sample Type of No. of % of root b isolates identified as

photo- isolates
synthesis A. /ipoferum A. brasilense
nir+ nir- nir+ nir-

Maize C4 49 55 41 17 3
Sorghum C4 19 20 80 0 0
Forage grasses c C4 45 33 70 0 7
Sugar cane C4 24 0 0 0 100
Cyperus rotundus C4 7 0 100 0 0
Wheat C3 50 0 0 7 93
Paddy rice 8 C3 56 0 0 7 93
Dry rice C3 22 7 13 80
Oat C3 5 20 0 80
Rye C3 22 18 0 82
Barley C3 11 18 0 82

8 Includes 34 samples from Trinidad

b Roots sterilized in chloramine T for 0.5 to 60 min, depending on the root structure
of each species. Soil samples in almost all harvests contained all three Azospirillum
groups
C Panicum maximum, Cynodon dactylon, Pennisetum purpureum, Digitaria decumbens,
Hemarthria altissima
II.3 Dinitrogen Fixation in Rhizosphere and Phyllosphere Associations 337

fluorescence reactions (DE-POLLI et al. 1980). Azospirillum strains isolated from

rice roots in Trinidad (Table 3) and Senegal (RINAUDO personal communication)
were also almost exclusively A. braisilense nir-, while the strains isolated from
maize roots in Belgium seem from the description to be A. lipoferum (VLASSAK
and REYNDERS 1978a). These authors also observed a certain specificity offunc-
tion in that some Azospirillum strains (species not mentioned) induced higher
nitrogenase activity in maize, and others in Chicorium intybus. In contrast, infec-
tion and nitrogenase activity in monoxenic grass seedlings seems non-specific
(UMALI-GARCIA et al. 1978, BURRIS 1977 and unpublished data from our labora-
tory). Monoxenic wheat roots adsorbed 109 Rhizobium cells per cm of root
(SHIMSHICK and HERBERT 1978) confirming lack of specificity in the absence
of the soil micro flora, also in Rhizobium attachment.
When specific Azospirillum strains were genetically marked with streptomy-
cin resistance (20 ~g ml- 1 ) and then inoculated into field-grown maize and
wheat, more than 80% of the rhizosphere and root surface population was
from the inoculated strains (BALDANI and DOBEREINER 1980). The identification
of the inoculated strains within surface-sterilized roots was complicated by the
occurrence of high proportions of low-level streptomycin-tolerant Azospirillum
within maize and wheat roots. Non-homologous strains also established in the
rhizosphere but not within roots.
Proportions of streptomycin-resistant bacteria higher than in soil were ob-
served in legumes and vegetables and in wheat, maize and sorghum (BROWN
1961, DOBEREINER and BALDANI 1980). Increases of numbers of actinomycetes
in the rhizosphere (RoVIRA 1965) and selective assimilation of streptomycin
by plant cells (PRAMER, 1955) have been reported. It seems possible, therefore,
that cereals can regulate bacterial infection through requirements for antibiotic
resistance.

2.3.4 Physiology of Azospirillum

Better understanding of the physiology of Azospirillum spp. in culture media
will help in understanding the more complicated nature of the root-bacteria
ensemble.
Azospirillum is an obligate aerobe and is unable to grow in the absence
of air except in the presence of NO; and an organic N source (NEYRA and
VAN BERKuM 1977). Rapid aerobic growth occurs with NH; as nitrogen source
but growth on N2 is slower (generation time 1 and 5.3-7 h respectively) (OKON
et al. 1977a, VOLPON et al. 1981) and does not proceed under air. Oxygen re-
quirements are very narrow, and a dissolved p02 in the order of 0.003 atm
is necessary for a functional nitrogenase (VARGAS 1977, NELSON and KNOWLES
1978). On the other hand, rapidly growing continuous or batch cultures consume
so much O 2, that once an optical density of about 0.3 is reached (560 nm)
Azospirillum can be grown under air with rapid stirring and sparging (DOBER-
EINER 1977, VOLPON et al. 1981) as long as the input of O 2 corresponds to
the O 2 uptake and the dissolved p02 remains below the above-mentioned limit.
Cultivation of Azospirillum in semi-solid media avoids this problem due to
the O 2 diffusion gradient. The very motile organisms move to the zone with
338 J. DOBEREINER:

the required p02' where they multiply. Once O 2 demands increase, the bacterial
pellicle moves closer to the surface.
The oxygen tolerance of Azospirillum changes with the age of the culture,
and late log phase organisms are more tolerant to O 2 than early log phase
cells (STEPHAN et al. 1981, VOLPON et al. 1981). The increased O 2 tolerance
was more pronounced in A. lipoferum and occurred when the glucose or lactate
concentration in the medium had dropped to less than 0.05% (STEPHAN et al.
1981, VOLPON et al. 1981). This is in contrast to observations with Azotobacter
(HILL et al. 1972) and stresses the weak oxygen protection mechanism for nitro-
genase through carbon respiration, in Azospirillum. The increased O 2 tolerance
was attributed to a threefold increase in uptake hydrogenase (HUP) activity
which reached much higher levels than those observed in Rhizobium bacteroids
(MAIER et al. 1979) (see Chap. II.1 for explanation of the role of HUP). Hydro-
gen recycling seems complete in Azospirillum spp. and no traces of H2 evolution
were observed in the above-mentioned experiments. How far H2 oxidation is
responsible for ATP generation and oxygen consumption, as suggested for Rhi-
zobium bacteroids (RUIZ-ARGUEso et al. 1979), is not yet clear.
Efficiency of nitrogen fixation also changes during the growth cycle. In-
creases from 20 mg N g - 1 glucose during early log phase to 48 mg during
late log phase and beginning stationary phase in A. lipoferum and from 15 mg
N g-l lactate to 90 mg in A. brasilense were observed (VOLPON et al. 1980,
STEPHAN et al. 1981). These changes in efficiency were accompanied by changes
in specific activity and coincided with the pattern of hydrogenase activity and
oxygen tolerance. The high efficiencies confirm earlier reports from our labora-
tory (DAY and DOBEREINER 1976) and the changes during the growth cycle
explain the discrepancy in the data from various laboratories (OKON et al. 1976b,
VARGAS 1977). There was an increase of total PHB (from 24 to 68 J.lg ml- 1 )
during growth, although the percentage decreased (VOLPON et al. 1981) and
therefore the given efficiencies cannot be attributed to the use of PHB accumu-
lated earlier in the growth cycle. Whether the high efficiencies operate in bacteria
within plant roots needs to be shown. In searching for rhizocoenoses which
make substantial contributions to the N economy of the plant, bacteria which
fix N2 actively during restricted growth are the kind to look for.
In batch cultures with A. brasilense (semi-solid medium) NH: is excreted
during early growth, accumulating to concentrations of 0.3 mM (ELMERICH et al.
1978). Later this NH: seems to be assimilated. A. /ipoferum in Orlimited batch
cultures with liquid medium also excretes up to 25% of the fixed nitrogen.
Cell-free extracts containing active nitrogenase were isolated from A. brasi-
lense (OKON et al. 1977b), and showed requirements similar to other nitrogen-
ases. The activation factor necessary for a functional Rhodospirillum rubrum
nitrogenase was necessary for A. brasilense (LUDDEN et al. 1978, OKON et al.
1977b). NH: incorporation and regulation of nitrogenase seems similar to that
in other N 2-fixing organisms and N 2-grown cells showed higher glutamine syn-
thetase (GS) and glutamate-oxo-glutarate amino-transferase (GOGAT) activi-
ties and lower glutamate dehydrogenase (GDH) activities when compared with
NH: -grown cells (OKON et al. 1976b).
11.3 Dinitrogen Fixation in Rhizosphere and Phyllosphere Associations 339

02 02 02
Access >Uptake Access IS; Uptake Access Traces

No N2 Fixation N2 Fixation in the N2 Fixation With-

Absence of NH3. out Growth. NOli
N03" or N02 and N03 do not
Inhibit N2 Fixati-
on

Fig. 2. Effect of oxygen on nitrogen metabolism by Azospirillum spp. When O 2 access

is faster than O 2 uptake by the cells (left square) NO; or NH3 are assimilated and
there is no N2 fixation. When O 2 access is in equilibrium with respiration (middle square)
N 2 fixation occurs in the absence of mineral nitrogen. When traces or no O 2 has access
to the cells (right square) there is nitrate-dependent nitrogenase activity and denitrification.
Notice that nir+ Azospirillum strains are able to participate in all steps of the nitrogen
cycle except nitrification (for details see text)

Both Azospirillum spp. were shown to be capable of autotrophic growth

on Hz and CH 4 with N z as sole nitrogen source (SAMPAIO et al. 1981). Figure 2
summarizes the various steps in nitrate metabolism of Azospirillum spp. and
their interactions with Oz. All strains from both species isolated so far are
able to assimilate NO~ and NH; and there is no N z fixation with more than
1 mM NO~ or NH; (DAY and DOBEREINER 1976), under aerobic conditions.
Under Oz-limiting conditions NO~ can be dissimilated to NO; which accumu-
lates in the medium in some strains (nir-) (MAGALHAES et al. 1978, NEYRA
et al. 1977). About two thirds of the strains of A. brasilense (TARRAND et al.
1978) and most strains of A. lipoferum (except root isolates) (DOBEREINER and
BALDANI 1980) can dissimilate NO; to gas (NEYRA et al. 1977). Mutants (nr-,
nir- or both) were isolated by selection for CIO~ resistance from both species
which in the presence of 10 mM NO~ fix as much N z as without (MAGALHAES
et al. 1979). Monoxenic plants inoculated with nr- mutants, in the presence
of 1 mM KN0 3 , had twice as much nitrogenase activity than the parent strains
(DoBEREINER and DE-PoLLI 1980). Mutants with nr+ but nir- do not fix N z
in the presence of NO~ , indicating that the accumulation of NO; and not
NO~ assimilation is responsible for nitrogenase inhibition. The inhibition by
NO; of nitrogenase activity in cell-free extracts (P.B.W. VAN BERKuM personal
communication) and that ofnr- as well as nir- mutants supports this (MAGAL-
HAES et al. 1978). Sulphate, iron or fumarate cannot replace NO~ as terminal
respiratory electron acceptor (BOUIE et al. 1981).
Under anaerobic conditions NO~ can supply energy for N z fixation but
without supporting growth. This process however was suggested to be transitory
only until the assimilatory N0 3 reductase is synthesized (BOTHE et al. 1981)
340 J. DOBEREINER:

when anaerobic growth proceeds with NO; as N source. Anaerobic NO;-

dependent nitrogenase activity (10-20 mM NO;) in rates similar to those ob-
served with O 2 was reported by NEYRA and BERKUM (1977) and SCOTT et al.
(1979), in both Azospirillum species. The activities were higher than those ob-
served of NO; -dependent nitrogenase activity in Rhizobium bacteroids (ZABLO-
TOWICZ and FOCHT 1979). Mutants (nr-) showed no anaerobic nitrogenase
activity in the presence of NO; . The NO; -dependent nitrogenase activity was
also observed in nr+ nir- strains of A. brasilense (SCOTT et al. 1979). This
shows that there is no need for NO;: dissimilation to gaseous forms in order
to replace O 2 for respiration and nitrogenase activity. The same was observed
with Rhizobium bacteroids (ZABLOTOWICZ et al. 1978). In fact it has not been
shown yet that this second step in nitrate dissimilation can be used by Azospiril-
fum to drive nitrogenase activity. This apparent contradiction of a nitrogen-
fixing organism denitrifying has now also been shown with Rhizobium bacteroids
(ZABLOTOWICZ and FOCHT 1979). These authors suggest a possible role in avoid-
ing NO;: inhibition of N 2 -ase or as alternative e- acceptors under limited O 2
access. This would, however, not improve the efficiency of these systems because
NO; reduction compares unfavorably with O 2 in terms of voltage couples
(421 and 820 mV respectively) (ZABLOTOWICZ et al. 1978), and 100 mol NO;
were reduced for 1 mol N2 fixed (SCOTT et al. 1979). The importance of these
characteristics as a basis for studies of complementation of N 2 fixation with
mineral fertilizer N is obvious.

2.4 Associations with Other N2-Fixing Bacteria

A number of less well-defined associations have been described. Under temper-

ate conditions a Bacillus sp., probably B. pofymixa, has been shown to multiply
selectively in the rhizosphere of certain wheat lines (NEAL and LARSON 1976)
and inoculation of wheat in small pots resulted in increases of plant nitrogen
(RENNIE and LARSON 1981). The wheat lines which stimulate Bacillus growth
are non-leaky lines, that is they accumulate rather than excrete photosynthates
and therefore support a localization of the bacteria within the roots. NEAL
and LARSON (1976) showed infection of intercellular spaces and middle lamellae
of roots.
Many reports are available on paddy rice associations with diazotrophs
and there are certainly several organisms involved. Azospirillum brasilense has
been isolated from surface sterilized roots in large numbers, in many places
(BALDANI and DOBEREINER 1980, BODDEY 1980, RAJARAMAMOHAN et al. 1978,
WATANABE et al. 1979) and highest numbers were observed during the peak
of nitrogenase activity at flowering; in contrast total numbers of bacteria were
high all through the growth cycle of the plant (WATANABE and CHOLITKUL
1979). Generally in the rhizosphere of paddy rice, aerobic N 2 -fixing bacteria
were more numerous than anaerobes, and Beijerinckia spp. and Enterobacter
cloaceae were found in high numbers (YOSHIDA and ANCAJAS 1971, BALANDREAU
1975). Methane-oxidizing diazo trophic bacteria were also found in high
numbers, and the large amount of CH 4 which can accumulate in rice paddies
11.3 Dinitrogen Fixation in Rhizosphere and Phyllosphere Associations 341

should not be overlooked as potential energy source for N2 fixation (DE-BoNT

et al. 1978). Recently WATANABE and BARRAQUIO (1979) reported the occurrence
of glucose-utilizing aerobic semi-diazotroph bacteria within rice roots which
comprised 81 % of the total bacteria flora and which were tentatively classified
as Achromobacter. Proportions of these bacteria in the stem and rhizosphere
soil were 37.5% and 2.4% respectively. Roots of several other water plants
contained similar organisms but none of dryland plants did so. Diazotrophic
Spirillum-like organisms were isolated from roots of Potamogetonfiliformis (SYL-
VESTER-BRADLEY 1976) and Spartina alterniflora, a C4 marsh grass and were
identified as Compylobacter sp. (PATRIQUIN et al. 1981). In Spartina, root nitro-
genase activity was correlated with root sugar concentrations and was CO 2 -
dependent (PATRIQUIN et al. 1981). Additions of sugar or malate did not substan-
tially increase the internal nitrogenase activity, indicating a large carbon pool
(BOYLE 1978). Diazotrophic Enterobacteriaceae and Bacillus sp. have been found
in high numbers in several other plants, but plant-bacteria interactions have
not been shown.

3 Agronomic Aspects

Striking implications of manipulating diazo trophic biocoenoses in increased

crop yields or even in fertilizer economy, if ever, can only be expected after
the nature of the various systems is better understood, and much has to be
done in order to achieve this. There is, however, the possibility of more empirical
identification of limiting factors and the search for economically feasible mea-
sures to minimize them.

3.1 Plant Genotype Effects

The first priority for improved crop production is usually plant breeding. Signifi-
cant plant genotype differences in nitrogenase activity in intact soil plant systems
have been shown in rice (BALANDREAU et al. 1978, LEE et al. 1977b), wheat
(NERY et al. 1977, SUBBA RAo 1981), sorghum (NERY 1978, SUBBA RAo 1981)
and Brachiaria (PEREIRA and DOBEREINER 1981). There are cereal breeding pro-
grams in progress in Brazil (I.E. MARRIEL 1977, personal communication,
RUSCHEL and RUSCHEL 1981) and in India (SUBBA RAo 1981).
So far these breeding efforts envisage higher and more stable nitrogenase
activity, and there has been some preliminary information on inheritance mecha-
nisms (VON HULLOW 1978, RUSCHEL and DA SILVA 1981). These studies however
are made difficult by the lack of information about the relevant plant character-
istics to be selected for. Although initial screenings were aimed at selecting
"leaky" plants which could feed the bacteria in the rhizosphere (BRILL 1978)
there is now increasing evidence that the opposite feature is what is wanted.
The wheat lines which stimulate Bacillus sp. are all root-rot resistant, that is,
non-leaky types (RENNIE and LARSON 1981) and PATRIQUIN et al. (1981) have
342 J. DOBEREINER:

found that plants which accumulate large sugar pools are better Nz-fixers than
leaky plants.
Another important question is the possibility of breeding for positive or
at least non-negative interactions between nitrogen fixation and nitrate assimila-
tion, because even if part of the nitrogen requirements are provided by biological
fixation, the amount is unlikely to be sufficient for high yields. Maize genotypes
showed negative correlations of nitrogenase activity in excised roots with the
nitrate reductase in leaves when high nitrogen fertilizer (80 kg N ha- 1 ) levels
were applied, but no correlation with 20 kg N ha -1. An increase of this limit
would permit simultaneously efficient nitrogenase activity and nitrate assimila-
tion. Maize lines selected for high nitrogenase activity at reproductive growth
stage reduced as much CzH z with 80 as with 20 kg N ha -1 (BALDANI et al.
1979). Similar observations were made by TROLLDENIER (1977) with rice. Fur-
thermore application of 20 kg N ha - 1 increased nitrogenase activity of one
out of five Brachiaria genotypes (intact soil plant systems) (PEREIRA and DOBER-
EINER 1981), while the effect on the others was negative.

3.2 Environmental Effects

The identification of environmental factors which limit nitrogenase activity

under field conditions is essential for any attempt to find agriculturally viable
practices which may increase biological N z fixation in grasses. A general ap-
proach has been presented by BALANDREAU (1975), who monitored weekly dur-
ing one growing season soil and air temperature, light intensity, photosynthesis,
air humidity, and soil nitrogen content with in situ assay of nitrogenase activity
in soil plant cores of maize and forage grasses. Weekly variations in nitrogenase
activity were mostly dependent on soil N (mineral) and humidity, while oscilla-
tions during the day could be attributed to soil temperature and light energy
variations. Computer analysis of the interference of minimal and maximal soil
and air temperatures, nitrogen content, and soil humidity in Digitaria nitrogen-
ase activity confirmed minimum temperatures and NH; content in soil as the
major limiting factors (ABRANTEs et al. 1976).

3.3 Inoculation

The urgency to obtain cereals and grasses which can cover part of their nitrogen
needs by biological fixation has precipitated attempts of solving all problems
by inoculating with the first strains of Azotobacter (RUBENCHIK 1963) or Azospir-
ilium (BOUTON et al. 1979 and SMITH et al. 1977) at hand, irrespective of their
origin or efficiency. Even in the well-established legume symbiosis, inoculation
with specifically selected strains seldom brings about spectacular yield increases.
The recent advances discussed in this chapter show clearly the need for better
understanding of host strain interactions before it is known which characteristics
of Azosp irillum , Bacillus or other diazo trophic bacteria are of advantage in
the various biocoenoses. Plant roots seem to select for nitrite reductase negative
II.3 Dinitrogen Fixation in Rhizosphere and Phyllosphere Associations 343

(nir-) strains (DOBEREINER and BALDANI 1980; DOBERElNER and DE-POLLI 1980)
but it is not yet known whether the nitrate reductase (ur) is of advantage to
the bacteria, the plant or both. Azospirillum strains seem to interact in nitrate
assimilation by the plant (DE FREITAS et al. 1981 and VILLAS BOAS and DOBER-
EINER 1981). Active denitrification occurs in nir+ Azospirillum cultures, but no
information as yet is available on the role of such strains in soil, or in the
rhizosphere. As much as 7% of the applied NO; fertilizer was lost in 3 days
as N 20 or N2 in Brachiaria cores (PEREIRA and DOBEREINER 1981).
Field inoculation of wheat with low level streptomycin-resistant Azospirillum
brasilense nir-, isolated from sterilized wheat roots, brought about 30% plant
nitrogen increases (DOBEREINER and DE-POLL! 1980). Inoculation with Azospiril-
tum strains isolated from maize gave total plant nitrogen increases in the order
of 40 kg N ha - 1, equivalent to those obtained with 60 kg ha -1 of fertilizer
N (DOBERElNER and BALDANI 1982). Inoculation effects with empirically selected
Azospirillum strains in several field experiments in India and Israel corresponded
to grain or forage yield increases obtained with 20 to 80 kg of fertilizer N
(SUBBA RAo 1981, KAPULNIK et al. 1981) altho-q.gh many other attempts failed.
In all these experiments, the inoculation effects increased with low nitrogen
fertilizer levels, and a role of Azospirillum spp. in nitrate assimilation cannot
be ruled out.

4 Phyllosphere Associations

According to the newly proposed nomenclature (International Workshop on

Associative N2 Fixation, Piracicaba 1979), associations of N 2-fixing bacteria
with plant leaves were termed phyllocoenoses. RUINEN (1956) reported the very
common occurrence of nitrogen-fixing bacteria on leaves of tropical plants and
proposed the name phyllosphere for this habitat. To what extent the occurrence
of N 2-fixing microorganisms in the phyllosphere can be considered an associa-
tion or a phyllocoenosis is not yet clear. So far no specificity in plant-bacteria
interactions has been demonstrated, although the high frequency seems to indi-
cate selective environments on tropical forest leaves. RUINEN (1956) was able
to isolate Beijerinckia spp. from 192 out of 198 randomly collected leaves of
shrubs and trees.
The phyllosphere as characterized by RUINEN includes the leaf surfaces, and
leaf sheaths which especially in grasses represent a suitable habitat for microor-
ganisms (RUINEN 1970). Substantial amounts of minerals and photosynthates
accumulate on the leaf surface and in leaf sheaths, and dew and liwt rains
which are intercepted by the canopy can transform these substances into a
rich culture medium for microorganisms. Sugar concentrations in rainwater
collected from broad forest leaves in West Africa were in the order of 10 to
200 ppm (RUINEN 1965). In the sheath water, even higher quantities of sugar
can accumulate and 4,000-7,000 ppm were found in young leaves of C 4 forage
grasses in Indonesia and'West Africa (BESSEMS 1973; RUINEN 1970).
344 J. DOBERElNER:

4.1 Microorganisms in the Phyllosphere

A large variety of N 2 fixing bacteria have been isolated from leaves of tropical
shrubs, trees and grasses. Beijerinckia spp. and Azotobacter spp. seem most
common, but this could rather reflect the easiness of their isolation and identifi-
cation (RUINEN 1956, 1961; VASANTARIAN and BHAT 1968, HARRELSON 1969,
BHAT et al. 1971). In the damp warm evening air near tropical forests a very
characteristic smell of Beijerinckia spp. was recognised repeatedly (DOBERElNER
unpublished). Rice sheaths in Java contained predominantly Azospirillum and
Beijerinckia (RUINEN 1974), and the leaf sheaths of mature Tripsacum laxum,
a tall tropical grass, contained 10 8 cells of Azospirillum and as many of Klebsiella
sp. when the sugar concentration reached maximal values of 4,000 ppm (RUINEN
1975). This author suggested that N 2-fixing bacteria are washed into the soil
with sugar-containing leaf and stem washings. We refer to Section 2.3.2 where
almost equally high numbers of Azospirillum were reported, within roots and
stems of maize.
Various Cyanobacteria were also found on forest leaves in Puerto Rico (HAR-
RELSON 1969) and Java (RUINEN 1974).
A few reports also indicate occurrence of N 2-fixing bacteria in the phyllo-
sphere of temperate plants. BESSEMS (1973) found Klebsiella spp. in the maize
leaf sheaths in Holland, and JONES (1970) found considerable nitrogenase activity
(estimated by the acetylene reduction assay) on Douglas fir leaves (Pseudotsuga
douglash) but no attempts were reported to identify the responsible organisms.

4.2 Nitrogen Fixation in the Phyllosphere

Estimates of nitrogen fixation in the phyllosphere, as in roots, are very variable

and range from traces to 70 g N ha -1 per day (RUINEN 1974) or 10% of the
total nitrogen incorporated into the leaves (HARRELSON 1969). These values
may seem small in relation to the large amounts of N which are recycled in
a tropical forest where 2 t N ha -1 are locked up in the plant system (GREENLAND
and NYE 1959). Losses, however, are small in equilibrium systems where nitrogen
rarely will remain the limiting factor. In regenerating forests on the other hand
40 to 100 kg N ha -1 have been found to be incorporated in ecosystems practi-
cally devoid of legume species in Ghana and West Africa (GREENLAND and
KOWAL 1960). There it is possible that N2 fixation in rhizocoenoses and in
the humus-rich soil is complemented by contributions from the phyllosphere.

5 General Conclusion

Nitrogen fixation associated with Gramineae roots (rhizocoenoses) and leaves

(phyllocoenoses) is very variable but can reach economically important propor-
tions especially in the tropics. In low nitrogen input agricultural systems this
11.3 Dinitrogen Fixation in Rhizosphere and Phyllosphere Associations 345

contribution will be most important. Also in regenerating forests and savannas

biological nitrogen fixation is the only nitrogen input to replace constant losses
through leaching and denitrification. In such environments legumes and nitrogen
fixing grasses have competitive advantages and it is there that highest rates
of nitrogen fixation should be expected until another element, usually phospho-
rus, becomes limiting. In contrast, equilibrium forests or savannas will never
be nitrogen-limited because there is no other way for nature to add plant nu-
trients to the ecosystem except by nitrogen. Legumes are seldom nodulated
once the climax is reached and these systems will not give us a good measure
of the potential of nitrogen fixation by either system.

References

Abrantes GTV, Day JM, Cruz VC, D6bereiner J (1976) Fatores limitantes da fixas;ao
de nitrogenio em campo de Digitaria decumbens cv. Transvala. Art XV Congr Bras
Cie Solo
App AA, Watanabe I, Alexander M, Ventura W, Daez C, Santiago T, De Datta SK
(1980) Nonsymbiotic nitrogen fixation associated with the rice plant in flooded soils.
Soil Sci 130 5: 283-289
Balandreau J (1975) Activite nitrogenasique dans la rhizosphere de quelques Graminees.
PhD thesis, Univ Nancy, France
Balandreau J, Ducerf P, Hamad-Fares I, Weinhard P, Rinaudo G, MiIIier C, Dom-
mergues Y (1978) Limiting factors in grass nitrogen fixation. In: D6bereiner J, Burris
RH, Hollaender A (eds) Limitations and potentials for biological nitrogen fixation
in the tropics. Basic Life Sciences vol X. Plenum, New York London
Baldani 11, BIafia RAG, D6bereiner J (1979) Efeito do genotipo do milho na atividade
da nitrogenase e da nitrato-reductase. Pesqu Agropecu Bras 14: 165-173
Baldani 11, Pereira PAA, da Rocha REM, D6bereiner J (1981) Especificidade na infecs;ao
de raizes por Azospirillum spp. em plantas com via fotossintetica C 3 e C 4 • Pesqu
Agropecu Bras 16:325-330
Baldani VLD, D6bereiner J (1980) Host plant specificity in the infection of cereals with
Azospirillum spp. Soil BioI Biochem 12:433-439
Becking JH (1963) Fixation of molecular nitrogen by an aerobic Vibrio or Spirillum
sp. Antonie van Leeuwenhoek J Microbiol Serol 29:326
Beijerinck MW (1925) Uber ein Spirillum, welches freien Stickstoff binden kann? Zen-
tralbl Bakteriol Parasitkde Abt II 63: 353-359
Berg RH, Tyler ME, Novick NJ, Vasil V, Vasil IK (1980) Biology of Azospirillum-
sugarcane association: enhancement of nitrogenase activity. Appl Environ Microbiol
29:642-649
Bessems EPM (1973) Nitrogen fixation in the phyllosphere of Gramineae. Agric Ses
Rep 786: 1-68
Bhat JV, Limaye KS, Vasantharajan VN (1971) The effect of the leaf surface microflora
on the growth and root exudation of plants. In: Preece TF, Dickinson CH (eds)
Ecology of leaf surface microorganisms. Academic Press London New York
Boddey RM (1980) Biological nitrogen fixation in the rhizosphere of lowland rice. PhD
thesis, Univ West Indies, Trinidad
Bothe H, Klein B, Stephan MP, D6bereiner J (1981) Transformations of inorganic nitro-
gen by Azospirillum spp. Arch Microbiol 130:96-100
Bouton JH, Smith RL, Schank SC, Burton GW, Tyler ME, Littell RC, Gallaher RN,
Quesenberry KH (1979) Response of pearl millet inbreds and hybrids to inoculation
with Azospirillum brasilense. Crop Sci 19: 12-16
346 1. DOBERElNER:

Boyle CD (1978) Some characteristics of nitrogenase activity associated with diazotrophs

in and around Spartina alterniflora roots. MS thesis, Dalhousie Univ, Can
Brill WI (1978) Genetics and regulation of nitrogen fIxation. In: Dobereiner 1, Burris
RH, Hollaender A (eds) Limitations and potentials for biological nitrogen fIxation
in the tropics. Basic Life Sciences, vol X. Plenum, New York London
Brown ME (1961) Stimulation of streptomycin-resistant bacteria in the rhizosphere of
leguminous plants. 1 Gen MicrobioI24:369-377
von Biillow IFW (1978) Plant influence in symbiotic nitrogen fIxation. In: Dobereiner
1, Burris RH, Hollaender A (eds) Limitations and potentials for biological nitrogen
fIxation in the tropics. Basic life sciences, vol X. Plenum, New York London
von Biillow IFW, Dobereiner 1 (1975) Potential for nitrogen fIxation in maize genotypes
in Brazil. Proc Nat! Acad Sci USA 72:2389-2393
Burris RH (1977) A synthesis paper on nonleguminous N 2-fIxing systems. In: Newton
W, Postgate lR, Rodriguez-Barrueco C (eds) Recent developments in nitrogen fIxa-
tion. Academic Press, London New York
Burris RH, Albrecht SL, Okon Y (1978) Physiology and biochemistry of Spirillum /ipo-
ferum. In: Dobereiner 1, Burris RH, Hollaender A (eds) Limitations and potentials
for biological nitrogen fIxation in the tropics. Basic Life Sciences, vol X. Plenum,
New York London
Day 1M, Dobereiner 1 (1976) Physiological aspects of N 2-fIxation by a Spirillum from
Digitaria roots. Soil BioI Biochem 8:45-50
De-Bont lAM, Lee KK, Bouldin DF (1978) Bacterial oxidation of methane in a rice
paddy. In: Environmental role of nitrogen-fIxing blue-green algae and asymbiotic
bacteria. Ecol Bull (Stockholm) 26: 91-96
De-Polli H, Bohlool BB, Dobereiner 1 (1980) Serological differentiation of Azospirillum
species belonging to different. host-plant specifIcity groups. Arch Microbiol
126: 217-222
Dobereiner 1 (1961) Nitrogen-fIxing bacteria of the genus Beijerinckia Derx in the rhizo-
sphere of sugar cane. Plant Soil 15: 211-216
Dobereiner 1 (1966) Azotobacter paspa/i sp. n., uma bacteria fIxadora de nitrogenio
na rizosfera de Paspalum. Pesqu Agropecu Bras 1: 357-365
Dobereiner 1 (1970) Further research on Azotobacter paspa/i and its variety specifIcity
occurrence in the rhizosphere of Paspalum notatum Fliigge. Zentralbl Bakterid Para-
sitenkde II 124:224-230
Dobereiner 1 (1977) Physiological aspects of N2 fIxation in grass-bacteria associations.
In: Newton W, Postgate lR, Rodriguez-Barrueco (eds) Recent developments in nitro-
gen fIxation. Academic Press, London New York
Dobereiner 1 (1978) Nitrogen fIxation in grass-bacteria associations in the tropics. Proc
Isot Bioi Dinitrogen Fix. IAEA, Vienna
Dobereiner 1, Baldani VLD (1980) Selective infection of maize roots by streptomycin-
resistant Azospirillum /ipoferum and other bacteria. Can 1 Microbiol 25: 1264-1269
D6bereiner 1, Baldani 11 (1982) Bases cientpificas para uma agricultura biol6gica. Cienc
Cult (Sao Paulo) 37: 869-881
Dobereiner 1, Campe10 AB (1971) Non-symbiotic nitrogen-fIxing bacteria in tropical
soils. Plant Soil Spec Vol 457-470
Dobereiner 1, Day 1M (1975) Nitrogen fIxation in the rhizosphere of tropical grasses.
In: Steward WDP (ed) Nitrogen fIxation by free-living micro-organisms. Cambridge
U niv Press, London
Dobereiner 1, Day 1M (1976) Associative symbioses in tropical grasses: Characterization
of microorganisms and dinitrogen-fIxing sites. In: Newton WE, Nyman Cl (eds)
Int Symp N2 Fix. Washington State Univ Press, Pullman
Dobereiner 1, De-Polli H (1980) Diazotrophic rhizocoenoses. In: Stewart WDP, Gallon
lR (eds) Nitrogen fIxation. Academic Press, London New York
Dobereiner 1, Day 1M, Dart PI (1972a) Nitrogenase activity of the Paspalum notatum-
Azotobacter paspa/i association and oxygen sensitivity. 1 Gen Microbiol 71: 103-106
Dobereiner 1, Day 1M, Dart PI (1972b) Nitrogenase activity in the rhizosphere of sugar
cane and some other tropical grasses. Plant Soi137:191-196
11.3 Dinitrogen Fixation in Rhizosphere and Phyllosphere Associations 347

D6bereiner J, Marriel IE, Nery M (1976) Ecological distribution of Spirillum lipoferum

Beijerinck. Can J Microbiol 22: 1464-1473
Elmerich C, Gauthier D, Houward J (1978) Regulation of nitrogenase biosynthesis in
Azospirillum brasilense. Proc Steenbock-Kettering Int Symp N2 Fix, Madison
Freitas de JLM, D6bereiner J (1978) Effects of oxygen on nitrogenase activity in a
sterilized intact grass system inoculated with Azospirillum spp. In: D6bereiner J,
Burris, RH, Hollaender A (eds) Limitations and potentials for biological nitrogen
fixation in the tropics. Basic Life Sciences, vol X. Plenum, New York London
Freitas de JLM, Pereira PAA, D6bereiner J (1981) Effect of organic matter and Azospiril-
tum spp. strains in the metabolism of nitrogen in Sorghum vulgare. In: Vose PB,
Ruschel AP (eds) Associative Nrfixation, vol I. CRC Press, Boca Raton
Giddens J (1977) Nitrogen accumulation by grasses. Ga Agr Res 19: 12~13
Greenland DJ, Kowal JLM (1960) Nutrient content of the moist tropical forest of Ghana.
Plant Soil 12: 154-174
Greenland DJ, Nye PH (1959) Increases in the carbon and nitrogen contents of tropical
soils under natural fallows. J Soil Sci 9: 284-299
Harrelson MA (1969) Tropical epiphyllae and nitrogen fixation. PhD thesis, Univ Georgia
Hill S, Drozd JW, Postgate JR (1972) Environmental effects on the growth of nitrogen-
fixing bacteria. J Appl Chern BiotechnoI22:541~558
Jaiyebo EO, Moore AW (1963) Soil nitrogen accretion under different covers in a tropical
rain-forest environment. Nature 197:317~318
Jenkinson DS (1973) Organic matter and nitrogen in soils of the Rothamsted classical
experiments. J Sci Food Agric 24: 1149-1150
Jones K (1970) Nitrogen in the Douglas fir, Pseudotsugadouglasii. Ann Bot 34:239-244
Kapulnik Y, Sarig S, Nur I, Okon Y, Kigel J, Henis Y (1981) Yield increases in summer
cereal crops in Israel fields inoculated with Azospirillum. Exp Agric 17: 179-187
Kavimandan SK, Subba Rao NS, Mohrir AV (1978) Isolation of Spirillum /ipoferum
from the stems of wheat and nitrogen fixation in enrichment cultures. Curr Sci
47:96-98
Lakshmi-Kurami M, Kavimandan SK, Subba Rao NS (1976) Occurrence of nitrogen
fixing Spirillum in roots of rice, sorghum, maize and other plants. Indian J Exp
Bioi 14 : 638-639
Lee KK, Alimagno B, Yoshida T (1977a) Field technique using the acetylene reduction
method to assay nitrogenase activity and its association with the rice rhizosphere.
Plant Soil 47 : 519-526
Lee KK, Castro T, Yoshida T (1977b) Nitrogen fixation throughout growth and varietal
differences in nitrogen fixation by the rhizosphere of rice planted in pots. Plant Soil
48:613-619
Ludden PW, Okon Y, Burris RH (1978) The nitrogenase system of Spirillum /ipoferum.
Biochem J 173: 1001~ 1003
Machado WC, D6bereiner J (1969) Estudos complementares sobre a fisiologia de Azoto-
bacter paspali e sua dependencia da planta (Paspalum notatum). Pesqu Agropecu
Bras 4: 53- 58
Magalhaes FMM, Patriquin D, D6bereiner J (1979) Infection of field-grown maize with
Azospirillum spp. Rev Bras Bioi 39: 587~596
Magalhaes LMS, Neyra CA, D6bereiner J (1978) Nitrate and nitrite reductase negative
mutants ofN 2 -fixing Azospirillum spp. Arch MicrobioI117:247~252
Maier RJ, Hanus FJ, Evans HJ (1979) Regulation of hydrogenase in Rhizobium japoni-
cum. J Bacteriol137: 824-829
Moore AW (1963) Nitrogen fixation in latosolic soil under grass. Plant Soil 19: 127~138
Neal JL Jr, Larson RI (1976) Acetylene reduction by bacteria isolated from the rhizo-
sphere of wheat. Soil Bioi Biochem 8:151~155
Nelson LM, Knowles R (1978) Effect of oxygen and nitrate on nitrogen fixation and
denitrification by Azospirillum brasilense grown in continuous culture. Can J Micro-
bioi 24: 1395-1403
Nery M (1978) Interaction of nitrogen fertilizer with nitrogen fixation (C 2 H 2 ) in sorghum.
In: D6bereiner J, Burris RH, Hollaender A (eds) Limitations and potentials for bio-
348 J. DOBERHINER:

logical nitrogen fIxation in the tropics. Basic life sciences, vol X. Plenum, New York
London
Nery M, Abrantes GTV, Santos dos 0, Dobereiner J (1977) Fixayao de nitrogenio em
trigo. Rev Bras Cie Solo 1: 15-20
Neyra CA, Berkum Van P (1977) Nitrate reduction and nitrogenase activity in Spirillum
/ipoferum. Can J Microbio123:306-310
Neyra CA, Dobereiner J (1977) Nitrogen fIxation in grasses. Adv Agron 29: 1-38
Neyra CA, Dobereiner J, Lalande R, Knowles R (1977) DenitrifIcation by N 2 -fIxing
Spirillum /ipoferum. Can J Microbiol 23: 30(}-305
Okon Y, Albrecht SL, Burris RH (1976a) Factors affecting growth and nitrogen fIxation
of Spirillum /ipoferum. J BacteriolI25:1248-1254
Okon Y, Albrecht SL, Burris RH (1976b) Carbon and ammonia metabolism of Spirillum
/ipoferum. J BacteriolI28:592-597
Okon Y, Albrecht SL, Burris RH (1977 a) Methods for growing Spirillum /ipoferum
and for counting it in pure culture and in association with plants. Appl Environ
Microbiol 33: 85-88
Okon Y, Houchins JP, Albrecht SL, Burris RH (1977b) Growth of Spirillum /ipoferum
at constant partial pressure of oxygen, and the properties of its nitrogenase in cell-free
extracts. J Gen Microbio198:87-93
Parker CA (1957) Non-symbiotic nitrogen-flXing bacteria in soil. III. Total nitrogen
changes in a fIeld soil. J Soil Sci 8: 48- 59
Patriquin DG, Dobereiner J (1978) Light microscopy observations of tetrazolium-reduc-
ing bacteria in the endorhizosphere of maize and other grasses in Brazil. Can J Micro-
bioi 24:734-742
Patriquin DG, Gracioli LA, Ruschel AP (1980) Nitrogenase activity of sugar cane propa-
gated from stem cuttings in sterile vermiculite. Soil Bioi Biochem 12:413-417
Patriquin DG, Boyle CD, Livingstone DC, McClung CR (1981) Physiology of the diazo-
trophic rhizocoenosis in salt march cord grass, Spar tina alterniflora Loisel. In: Vose
PB, Ruschel AP (eds) Associative N 2 -fIxation, vol II. CRC Press, Boca Raton
Pella JJ, Dobereiner J (1974) Efecto del nitrato y amonio en la actividad de la nitrogenase
de bacterias tropicales fIjadoras de nitr6geno atmosferico. Rev Lat Am Microbiol
16:33-44
Pereira PAA, Dobereiner J (1981) A note on nitrogenase and nitrate reductase activities,
and denitrifIcation in fIve Brachiaria spp. genotypes. In: Vose PB, Ruschel AP (eds)
Associative N 2 -fIxation, vol II. CRC Press, Boca Raton
Pramer D (1955) Absorption of antibiotics by plant cells. Science 121: 507-508
Purchase BS (1978) A potential source of nitrogen for Rhodesian agriculture. Rhod
Agric 75:99-104
Quintero MJ, Garza T (1978) Isolation of Spirillum lipoferum Beijerinck from tropical
grasses in Mexico. In: Dobereiner J, Burris RH, Hollaender A (eds) Limitations
and potentials for biological nitrogen fIxation in the tropics. Basic life sciences, vol X.
Plenum, New York London
Rajaramamohan-Rao V, Charyulu PBBN, Nayak DN, Ramakrishna C (1978) Nitrogen
fIxation by free-living organisms in tropical rice soils. In: Dobereiner J, Burris RH,
Hollaender A (eds) Limitations and potentials for biological nitrogen fIxation in the
tropics. Basic life sciences, vol X. Plenum, New York London
Rennie RJ, Larson RI (1981) Dinitrogen fIxation associated with disomic chromosome
substitution lines of spring wheat in the phytotron and in the fIeld. In: Vose PB,
Ruschel AP (eds) Associative N 2 -fIxation, vol I. CRC Press, Boca Raton
Reynders L, Vlassak K (1976) Nitrogen-fIxing Spirillum species in Belgian soils. Agricul-
ture 24: 329-336
Rovira AD (1965) Interactions between plant roots and soil microorganisms. Annu Rev
Microbiol19:241-266
Rubenchik LI (1963) Azotobacter and its use in agriculture. Acad Sci Ukraiman SSR.
Artman A (Translater) Natl Sci Found, Washington
Ruinen J (1956) Occurrence of Beijerinckia species in the "phyllosphere". Nature
177: 22(}-221
II.3 Dinitrogen Fixation in Rhizosphere and Phyllosphere Associations 349

Ruinen J (1961) The phyllosphere. I. An ecologically neglected milieu. Plant Soil

15:81-109
Ruinen J (1965) The phyllosphere. III. Nitrogen fixation in the "phyllosphere". Plant
Soil 22:375-394
Ruinen J (1970) The phyllosphere. V. The grass sheath, a habitat for nitrpgen-fixing
microorganisms. Plant Soil 33: 661-671
Ruinen J (1974) Nitrogen fixation in the phyllosphere. In: Quispel A (ed) The biology of
nitrogen fixation. Elsevier, North-Holland. Biomedical Press, Amsterdam New York
Ruinen J (1975) Nitrogen fixation in the phyllosphere. In: Stewart WDP (ed) Nitrogen
fixation in free-living microorganisms. Cambridge Dniv Press, London
Ruiz-Argiieso T, Emerich DW, Evans HJ (1979) Hydrogenase system in legume nodules:
A mechanism of providing nitrogenase with energy and protection from oxygen dam-
age. J Bacteriol 137: 824-829
Ruschel AP, Silva da WJ (1981) N 2-fixing populations in seeds and plant parts of Teo-
soide and Mazoide plants from Zea mays x Zea mexicana (Teosinte) crosses. In:
Vose PB, Ruschel AP (eds) Associative N 2-fixation, vol I. CRC Press, Boca Raton
Ruschel AP, Victoria RL, Salati E, Henis Y (1978) Nitrogen fixation in sugarcane (Sac-
charum officinarum L.). Ecol Bull (Stockholm) 26:297-303
Ruschel R, Ruschel AP (1981) Inheritance of Nz-fixing ability in sugar cane. In: Vose
PB, Ruschel AP (eds) Associative Nz-fixation, vol II. CRC Press, Boca Raton
Ruscoe AW, Newcomb EH, Burris RH (1978) Pleomorphic forms in strains of Spirillum
lipoferum by light and electron microscopy. Pg. 30th Proc Steenbock-Kettering Int
Symp Nitrogen Fixation. Dniv Madison, Wisc
Sampaio MJAM, Vasconcelos de L, D6bereiner J (1978) Characterization of three groups
within Spirillum /ipoferum Beijerinck. Ecol Bull (Stockholm) 26: 364-365
Sampaio MJAM, Silva da EMR, D6bereiner J, Yates MG, Pedrosa de FO (1981) Au-
totrophy and methylotrophy in Derxia gummosa, Azospirillum brasilense and A. /ipo-
ferum. In: Gibson AH, Newton WE (eds) Current perspectives in nitrogen-fixation.
Proc 4th Int Symp Nitrogen Fixation. Canberra Aust Acad Sci, Canberra, Aust
Scott DB, Scott CA, D6bereiner J (1979) Nitrogenase activity and nitrate respiration
in Azospirillum spp. Arch Microbiol121: 141-145
Shimshick EJ, Herbert RR (1978) Adsorption of rhizobia to cereal roots. Bioch Biophys
Res Commun 84:736-742
Silva da MFS, D6bereiner J (1978) Occurrence of Azospirillum spp. in soils and roots.
In: D6bereiner J, Burris RH, Hollaender A (eds) Limitations and potentials for bio-
logical nitrogen-fixation in the tropics. Basic life sciences, vol X. Plenum, New York
London
Sloger C, Owens LD (1976) N2 fixation by a temperate corn-Spirillum association. 2nd
Int Symp N 2-fixation, Salamanca (Poster)
Smith RM, Schank SC, Quesenberry KH, Milam JM, Hubbell DH (1977) Research
in nitrogen fixation by grasses after inoculation with Spirillum. Dniv Fla Dep Agron
AnnuRep
Smith RM, Thompson DO, Collier JW, Hervey RJ (1954) Soil organic matter, crop
yields, and land use in the Texas Blackland. Soil Sci 77: 377-388
Stephan MP, Pedrosa de 0 F, D6bereiner J (1981) Physiological studies with Azospirillum
spp. In: Vose PB, Ruschel AP (eds) Associative N 2-fixation, vol I. CRC Press, Boca
Raton
Subba Rao NS (1981) Response of crops to Azospirillum inoculation in India. In: Vose
PB, Ruschel A (eds) Associative N 2-fixation, vol I. CRC Press, Boca Raton
Sylvester-Bradley R (1976) Isolation of acetylene-reducing spirilla from the roots bf Pota-
mogetonfiliformis from Loch Leven (Kinross). J Gen Microbiol97: 129-132
Tarrand 11, Krieg NR, D6bereiner J (1978) A taxonomic study of the Spirillum /ipoferum
group, with descriptions of a new genus, Azospirillum gen. nov. and two species,
Azospirillum /ipoferum (Beijerinck) comb. nov. and Azospirillum brasilense sp. nov.
Can J MicrobioI24:967-980
Trolldenier G (1977) Influence of some environmental factors on nitrogen fixation in
the rhizosphere ofrice. Plant Soil 47:203-217
350 J. DOBEREINER: 11.3 Dinitrogen Fixation

Dmali-Garcia M, Hubbell DH, Gaskins MH (1978) Process of infection of Panicum

maximum and Spirillum /ipoferum. Ecol Bull (Stockholm) 26:373-379
Dmali-Garcia M, Hubbell DH, Gaskins MH, Dazzo FB (1980) Association of Azospiril-
lum with grass roots. Appl Environ MicrobioI39:219-226
Vargas MAT (1977) Effect of oxygen on N2 fixation by Spirillum /ipoferum. MS thesis,
Dniv Madison, Wisconsin
Vasantarian VN, Bhat JV (1968) Interrelations of microorganisms and mulberry. II.
Phyllosphere microflora and nitrogen fixation in leaf and root surfaces. Plant Soil
28:258-267
Vasil IK, Vasil V, Hubbell DB (1977) Engineering plant cell or fungal association with
bacteria that fix nitrogen. In: Hollaender A (ed) Genectic engineering for nitrogen-
fixation. Plenum, New York London
Villas Boas FCS; D6bereiner J (1981) Nitrogenase and nitrate reductase activities in
rice plants inoculated with various Azospirillum strains. In: Vose PB, Ruschel AP
(eds) Associative N 2-fixation, vol II. CRC Press, Boca Raton
Vlassak K, Reynders L (1978a) Associative dinitrogen fixation in temperate regions.
In: Proc Isotopes BioI Dinitrogen fixation. IAEA Vienna
Vlassak K, Reynders L (1978b) Conversion of tryptophan to auxins by Azospirillum
sp. Proc Steenbock-Kettering Int Symp N 2-Fixation, Abstr 44. Madison
Volpon AGT, De-Polli H, D6bereiner J (1980) Growth physiology of Azospirillum. Acad
Bras Cie 52:651
Volpon AGT, De-Polli H, D6bereiner J (1981) Changes in efficiency of nitrogen fixation
in various growth stages of batch cultures of Azospirillum /ipoferum. In: Vose PB,
Ruschel AP (eds) Associative Nrfixation, voi I. CRC Press, Boca Raton
Watanabe I, Barraquio WL (1979) Low levels of fixed nitrogen required for isolation
of free-living N 2-fixing organisms from rice roots. Nature 277: 565-566
Watanabe I, Cholitkul W (1979) Field studies on nitrogen fixation in paddy soils. In:
Nitrogen and rice. Int Rice Res Inst, Los Bafios, Manila, Philippines
Watanabe I, Barraquio WL, Guzman de MR, Cabrera DA (1979) Nitrogen-fixing (acety-
lene reduction) activity and population of aerobic heterotrophic nitrogen-fixing asso-
ciated with wetland rice. Appl Environ MicrobioI37:813-819
White JW, Holben FJ, Richer AC (1945) Maintenance level of nitrogen and organic
matter in grassland and cultivated soils over periods of 54 and 72 years. J Am Soc
Agron 37:21-33
Yoshida T, Ancajas RR (1971) Nitrogen fixation in the root zones of rice. Proc Soil
Soc Am 35: 156-157
Zablotowicz RM, Focht DD (1979) Denitrification and anaerobic, nitrate-dependent
acetylene reduction in cowpea Rhizobium. J Gen Microbiol 111 : 445-448
Zablotowicz RM, Eskew DL, Focht DD (1978) Denitrification in Rhizobium. Can J
MicrobioI24:757-760
IT.4 Uptake and Reduction of Nitrate:
Bacteria and Higher Plants
L. BEEVERS and R.H. HAGEMAN

1 Introduction

Although the primary role of plants in converting CO 2 into organic forms

of carbon for use by heterotrophic forms of life is recognized, the equally impor-
tant capacity of plants to convert inorganic nitrogen into organic forms is fre-
quently overlooked. While all animals normally utilize organic nitrogen to satis-
fy their nutritional needs the non-ruminant animals must be supplied with the
pre-fabricated essential amino acids, thus the capability of plants to utilize inor-
ganic nitrate nitrogen or ammonium nitrogen and convert it to organic forms
is of great importance. The currently accepted pathway for the assimilation
of nitrate and ammonia into organic form by green leaves is as follows.

oxoglutarate

NO; ------<-- 2
NO;-----NH3 - glutamine
"- 4
- 2 (glutamate)
~~~ ~
NADH NAD 6Fd red 6Fd ox ATP 2Fdox

1. nitrate reductase 2. nitrite reductase 3. glutamine synthetase 4. glutamate synthetase (GOGAT).

In plant roots the reductant for nitrite reductase and glutamate synthetase
is not known, however some evidence indicates that NADPH may be involved.
In leaves nitrate reductase is localized in the cytoplasm while the other three
enzymes are found associated with the chloroplast.
In bacteria, algae, and higher plants ammonium nitrogen is converted into
organic form primarily by the activities of the enzyme glutamine synthetase
and glutamate synthetase (MIFLIN and LEA 1976, TEMPEST et al. 1973). In fungi,
incorporation of ammonia into organic form appears to occur via glutamic
dehydrogenase (TEMPEST et al. 1973, MIFLIN and LEA 1976, SIMS and FOLKES
1964). Before nitrate nitrogen can be incorporated into organic form it must
first be reduced to ammonium~ The capacity for nitrate reduction is present
in bacteria, fungi, algae and higher plants. Various recent reviews describe the
enzymatic reduction of nitrate in higher plants (KESSLER 1976, BEEVERS and
HAGEMAN 1980, HEWITT and CUTTING 1979, ZUMFT 1976, GUERRERO et al. 1981,
LOSADA et al. 1981); VENNESLAND and GUERRERO (1980) and LOSADA and GUER-
RERO (1979) review the relationship of nitrate reduction and photosynthesis.
PAYNE (1973) and STOUTHAMER (1976) discuss nitrate metabolism in microorgan-
isms.
352 L. BEEVERS and R.H. HAGEMAN:

Because of space limitations, and to minimize overlap with recent reviews,

we have restricted coverage in this article to recent findings on the enzymology
of nitrate metabolism, but have extended the discussion on nitrate nutrition
which, apart from the reviews of HAYNES and GOH (1978), HAGEMAN (1980)
and the recent series of studies of the North Carolina group (RUFTY et al.
1982), has received less extensive treatment.

2 Available Nitrogen Sources

Although higher plants can utilize gaseous ammonium (HUTCHINSON et al. 1972)
and oxides of nitrogen (ROGERS et al. 1979), they obtain the bulk of their nitro-
gen primarily as nitrate and ammonium ions from the soil. Apart from fertilizer
nitrogen and rainfall (lightning and pollution) these ions are derived primarily
by mineralization of soil organic matter. The supply of ammonium and nitrate
ions varies with the environment, soil type, fertilizer practices and previous
cropping history. While plants can absorb soluble organic compounds (e.g.
amino acids) from solution, the availability of such compounds to soil-grown
plants is considered minimal due to rapid mineralization.
Availability of inorganic nitrogen to the plant can be decreased by leaching
of nitrate from the soil or by microbial activity. Inorganic nitrogen (ammonia
is the preferred form, JANSSON 1971) utilized by microorganisms for growth
is rendered unavailable to the plant. Soil organisms possessing dissimilatory
nitrate and nitrite reductive activities release gaseous nitrogen products to the
atmosphere (a process termed denitrification).
In contrast to the mobile nitrate anion the ammonium cation may be bound
to the negatively charged soil (usually clay) particles. These bound ions are
exchangeable and thus available to the plant. The binding of ammonium ions
to the clay particles minimizes leaching loss. The retention advantage of ammo-
nium over nitrate ions in most well aerated soils over 5 °C is, however, short-
lived because nitrification is rapid. Nitrapyrin (pyridine analog) and related
inhibitors of Nitrosomonas have been used to prevent nitrification on a commer-
cial scale (PRAsAD et al. 1971; HUBER et al. 1977). While the development of
a stable highly effective inhibitor of nitrification would appear beneficial with
respect to reduction in loss of nitrogen by leaching, the effect on crop production
by elimination of nitrate from the soil is debatable (HAGEMAN 1980).

2.1 Species Differences in Ammonium and Nitrate Utilization

The differential availability of nitrate and ammonium in the rooting media

raises the question of which form of nitrogen permits maximum growth by
a particular plant species. Work describing comparative studies aimed at estab-
lishing the relative merits of ammonium or nitrate as nitrogen source for plant
growth has been reviewed (HAYNES and GOH 1978, HAGEMAN 1980, STREET
11.4 Uptake and Reduction of Nitrate: Bacteria and Higher Plants 353

and SHEAT 1958). Since soils can provide both sources of inorganic nitrogen,
reliable assessments of the relative efficiencies of nitrate or ammonium can
only be made by using sterile solution or sand culture. However, results from
such nutrient culture experiments are frequently difficult to interpret as the
removal of nitrate from the medium by the plants creates an alkaline medium,
whereas in ammonium medium the external pH declines. The situation is further
complicated by the fact that the pH of the media may exert a differential effect
on the uptake of ammonium or nitrate and other ions; (VAN DEN DRmSCHE
1978, KIRKBY and HUGHES 1979).
When appropriate precautions are taken to prevent pH fluctuation, it is
found that plants vary in their ability to grow on ammonium or nitrate. In
general, calcifuge (acid-loving) plants growing naturally in acid soils when little
nitrification takes place utilize ammonium in preference to nitrate. In contrast,
ca1cicoles or plants with a wide pH tolerance utilize nitrate preferably (HAYNES
and GOR 1978). The differential capability of utilizing nitrate or ammonium
in some instances can be related to variations in the potential to produce the
enzyme nitrate reductase (STEWART et al. 1974) with plants growing in acidic
soil showing limited potential for nitrate reduction.
Although there are demonstrated differences between species in their ability
to grow on ammonium or nitrate sources, comparable growth can be frequently
obtained on either nitrate or ammonium. However, the growing of plants on
ammonium is frequently dependent upon close monitoring of culture conditions.
ARNON (1937) demonstrated that ammonium-grown barley plants required high-
er light intensity, longer photoperiod and additional Cu2 + or Mn2+ to achieve
productivity equivalent to their nitrate-grown counterparts. In view of the less
demanding culture conditions for nitrate-grown plants it is generally considered
that nitrate is a "safer" nitrogen source for crop growth and production.
With respect to the "safer" aspect, nitrate can be absorbed and stored
in various tissues without detriment to the plant (HAGEMAN 1980). Plants also
have regulatory mechanisms that minimize nitrate assimilation when they are
under stress or are depleted of carbohydrate reserves. Plants grown on nitrate
characteristically have high levels of organic acids, predominantly malate, or
malate and aconitate. In contrast, plants cannot accumulate ammonium ions
as the ammonium ion or its dissociation product ammonia is toxic. Ammonium
ions are "detoxified" by union with amino or keto acids to form amides or
amino acids. Plants supplied with adequate ammonium salts have low levels
of organic acids and reserve carbohydrates. The effects of such differential
organic acid contents on plant metabolism are not well studied, however some
workers have reported a positive correlation between organic acid level and
productivity.

2.2 Influence of Ammonium or Nitrate on Cation Uptake

The uptake of ammonium or nitrate is accompanied by changes in accumulation

of other ions and this ionic imbalance has been used to account for the differen-
tial preference of some plants for ammonium or nitrate. Under conditions of
354 L. BEEVERS and R.H. HAGEMAN:

ammonium nutrition there is a decreased intake of cations Ca 2+ and Mg2+

and sometimes K + (Cox and REISENAUER 1973), and plants growing on ammoni-
um frequently develop symptoms of calcium and or magnesium deficiency. Sig-
nificantly, those species normally adapted for ammonium uptake do not show
calcium or magnesium deficiency symptoms when cultivated on ammonium.
Cox and REISENAUER (1973) attributed the decreased calcium and magnesium
uptake during ammonium nutrition to ionic competition at the uptake site.
EpSTEIN (1972) suggested that ammonium and potassium may utilize the same
transport mechanism. However, the uptake system had only one tenth the affini-
ty for ammonia as for alkali cations.
Nitrate uptake is generally associated with an increased uptake and transport
of cations, particularly potassium (JACKSON and WILLIAMS 1968, MINOTTI et al.
1968, BLEVINS et al. 1974, BEN ZIONI et al. 1971). In a nutrient solution where
nitrate is the only source of nitrogen, anion (largely NO~) exceeds cation uptake.
To maintain internal electrical neutrality DIJKSHOORN (1962), suggested that
H+ is absorbed or OH- excreted to balance the excess of NO~ absorbed.
This transport of H+ or OH- would create an alkaline external medium and
neutralize the charge of the internal nitrate.
KIRKBY and KNIGHT (1977), RAVEN and SMITH (1976), and others indicate
that nitrate assimilation produces internal alkalinity according to the following
equation.
N0 3 +8H+ +8e- ~NH3+2H20+0H-

However, DIJKSHOORN (1962) showed that this was valid only for assimilated
nitrate that was absorbed as the companion ion of one of the metallic cations
(e.g. K +, Ca 2+ or Mg 2+). To regulate internal pH, the plant synthesizes organic
acids to neutralize the resultant alkalinity (DIJKSHOORN 1962, BEN ZIONI et al.
1971, BLEVINS et al. 1974, ISRAEL and JACKSON 1982). The organic acid synthe-
sized may be predominantly malate, the synthesis of which is regulated by
the pH of the cell (DAVIES 1973). In some species, aconitate in addition to
malate may be synthesized (NEYRA and HAGEMAN 1976).

2.3 Nitrate Uptake

The kinetics of nitrate uptake have not been as extensively studied as those
of many other ions. This deficiency probably reflects the lack of a convenient
available radioisotope of nitrogen and the fact that the nitrate taken up is
metabolized. Measurements of nitrate uptake which have been made have usual-
ly involved determinations of nitrate depletion from the surrounding medium.
VAN DEN HONERT and HOOYMANS (1955) established that in maize the plot
of uptake versus solution nitrate concentration was hyperbolic with a plateau
between 0.12 and 0.24 mM; half maximal rate, K m, was at 0.023 mM. Similar
hyperbolic relationships between absorption and solution concentration were
reported by LYCKLAMA (1963) with a Km of 0.033 mM. Investigations by RAO
and RAINS (1976a) with barley (Hordeum vulgare) and by DODDEMA and
TELKAMP (1979) with Arabidopsis thaliana suggest that nitrate uptake may follow
11.4 Uptake and Reduction of Nitrate: Bacteria and Higher Plants 355

a dual phase relationship with concentration. This is a feature demonstrated

for many anions and cations (EpSTEIN 1972). The Km for nitrate uptake for
Arabidopsis in Phase I was 0.040 mM and V max at 4.0 J.lmol g-1 h -1. For badey
at the phase I concentrations below 0.5 mM the ~m was 0.11 mM and Vmax
2.35 J.lmol g-1 h -1. In barley, although nitrate uptake increased at concentra-
tions above 0.5 mM, absorption appeared to be linearly and not curvilinearly
related to nitrate concentration. A similar relationship appears in a mutant
of Arabidopsis. However in the wild-type Arabidopsis the Phase II uptake (up
to 48 mM NO;-) had a Km for nitrate of 25 mM and V max of 700 J.lmol g-l
h - 1. Although there are reports of an interaction between nitrate uptake and
absorption of other anions (EpSTEIN 1953) and a similarity of uptake kinetics,
it appears that nitrate uptake is not mediated by the same carrier as other
anions. This conclusion is based on the observations that (1) other anions do
not always interfere with llitrate uptake (RAo and RAINS 1976a), (2) in nitrogen-
deficient plants which show typical chloride uptake nitrate absorption is limited
(BLEVINS et al. 1974). A capacity for nitrate uptake develops in the deficient
plant during a 3-4 h exposure to nitrate (RAo and RAINS 1976b, BLEVINS et al.
1974, JACKSON et al. 1973, MINOTTI et al. 1968, NEYRA and HAGEMAN 1975).
These time-course kinetics for the onset of capacity for nitrate uptake are similar
to those for the development of nitrate reductase. The onset of nitrate absorption
capacity and buildup of nitrate reductase can be prevented by inhibitors of
RNA and protein synthesis (JACKSON et al. 1973). Nitrate uptake is restricted
in molybdenum-deficient plants (LYCKLAMA 1963). RAo and RAINS (1976b) also
propose the concept of an absorption system positively correlated with nitrate
reductase. Collectively these observations have led to the suggestion that nitrate
uptake and reduction may be mediated by a common membrane-located protein
(BUTZ and JACKSON 1977).
However, the observations that tungstate, which prevents the development
of an active nitrate reductase, does not interfere with nitrate uptake (HEIMER
and FILNER 1970, 1971, RAo and RAINs 1976b) and the findings that nitrate
reductase mutants of Arabidopsis can accumulate nitrate (DODDEMA et al. 1978)
are inconsistent with the concept of one protein with both carrier and reductase
activity.

2.4 Influence of Ammonium on Nitrate Uptake and Utilization

Although nitrate uptake is frequently stimulated by and enhances the uptake

of readily absorbed cations such as calcium and potassium, it is observed that
ammonium may impede the absorption of nitrate in some instances (WEISSMAN
1964, PATE 1973, LYCKLAMA 1963, MINOTTI et al. 1969a, SHEN 1969, DODDEMA
et al. 1978, FRITH 1972, FERGUSON and BOLLARD 1969) but not in others (SMITH
and THOMPSON 1971, SCHRADER et al. 1972, OREBAMJO and STEWART 1975a).
It was originally suggested that the decreased absorption of nitrate in the pres-
ence of ammonia was due to an inhibition of nitrate reduction (LYCKLAMA
1963). A lowering of the level of extractable nitrate reductase by ammonium
treatment has been reported by SHEN (1969), OREBAMJO and STEWART (1975b),
356 L. BEEVERS and R.H. HAGEMAN:

RADIN (1975) and JOY (1969). A reduced assimilation of nitrate into organic
N in the presence of ammonium was reported by FERGUSON (1969) and
SCHRADER et al. (1972). Others report that ammonium or products of ammoni-
um assimilation do not interfere with nitrate reductase level (FERGUSON 1969,
BEEVERS et al. 1965, MINOTTI et al. 1969a, OAKS et al. 1977). In view of the
reduced utilization of nitrate in the presence of ammonium without an alteration
in nitrate reductase level it has been suggested (MINOTTI et al. 1969a) that ammo-
nium or to some extent the high acidity adjacent to the cellular boundary mem-
brane resulting from ammonium uptake in excess of nitrate uptake causes an
alteration in membrane permeability, thereby restricting the capacity for nitrate
absorption.
Despite these conflicting reports on the influence of ammonium on nitrate
uptake it is frequently observed that plants grow better on a mixture of inorganic
nitrogen sources (SCHRADER et al. 1972, Cox and REISENAUER 1973, WEISSMAN
1964, WARNCKE and BARBER 1973). Low levels of ammonium are required for
the optimal induction of nitrate reductase and growth of cell tissue cultures
of Paul's Scarlet rose (MoHANTY and FLETCHER 1976) and soybeans (BAYLEY
et al. 1972).

3 Nitrate Reduction

3.1 Bacteria

Certain bacteria can utilize nitrate as the sole nitrogen source for the synthesis
of all nitrogen-containing components of the cell. The assimilation may occur
under both aerobic and anaerobic conditions. In other instances nitrate serves
as a terminal electron acceptor under anaerobic conditions and this process
has been called nitrate respiration (dissimilation) by PAYNE (1973) and STOUTH-
AMER (1976). In both cases the end product of nitrate reduction is nitrite. Nitrate
reductases from bacteria have been divided into two types - nitrate reductase
A, which is membrane-bound and which can use chlorate in addition to nitrate
as a substrate, and nitrate reductase B, which is cytoplasmic and is inhibited
by chlorate (PICIllNOTY and PIECHAUD 1968). Nitrate reductase A plays a role
in anaerobic respiration while nitrate reductase B is usually considered assimila-
tory. However, nitrate reductase B may also serve a respiratory function in
facultative anaerobes and it is debatable whether it is always soluble (PICIllNOTY
1970).

3.2 Dissimilatory Nitrate Reductase

Most of the detailed investigations of bacterial nitrate reductase have been

on the A type. The enzyme can be solubilized from the membranes by detergent,
organic solvents or by incubating the membranes at alkaline pH at 60°C
IIA Uptake and Reduction of Nitrate: Bacteria and Higher Plants 357

(STOUTHAMER 1976). The solubilized enzymes are characterized as molybdopro-

teins with iron-sulfur groups. In general the membrane nitrate reductases appear
to be composed of two subunits of molecular weight 150,000 and 60,000
(STOUTHAMER 1976, VINCENT and BRAY 1978, VANT RrnT et al. 1979). In addition
to a membrane-bound nitrate reductase A other bacteria, depending on culture
conditions may contain varying amounts of a soluble nitrate reductase (GUER-
RERO et al. 1973, VILA et al. 1977, DANIEL and GRAY 1976, BURKE and LASCELLES
1979). The soluble enzyme in Staphylococcus aureus had a molecular weight
of approximately 225,000 and a large subunit of molecular weight 140,000 which
has been shown to possess catalytic activity (BURKE and LASCELLES 1979).
Dissimilatory nitrate reductase (Pichinoty Type A) in membrane fractions
from bacteria utilizes a variety of respiratory intermediates or reduced pyridine
nuc1eotides for nitrate reduction (STOUTHAMER 1976). In some instances, e.g.
E. coli and Klebsiella aerogenes, nitrate reduction to nitrite is associated with
the generation of ATP. In other types, e.g. Paracoccus denitrificans and Bacillus
licheniformis, the reduction of nitrate to nitrite is not accompanied by phosphor-
ylation. In these latter organisms it is considered that phosphorylation is asso-
ciated with the reduction of nitrite during dissimilatory nitrite reduction
(STOUTHAMER 1977, WOOD 1978, VANT RrnT et al. 1979).
When respiratory substrates or pyridine nuc1eotides serve as reductants for
dissimilatory nitrate reductase, reduction is inhibited by azide, cyanide and p-
chloromercuribenzoate (PCMB). Nitrate reduction mediated by respiratory sub-
strates in some instances is inhibited by n-heptyl hydroxyquinoline-N-oxide
(HONO) or dicoumarol. Dissimilatory nitrate reduction by particulate prepara-
tions can utilize reduced viologen dyes as electron donors; with the reductants
nitrate reduction is insensitive to inhibition by PCMB and (HONO) but is
prevented by cyanide and azide. Nitrate reductase assays on solubilized enzymes
must be performed with reduced viologen dyes.
When particulate preparations from bacteria grown anaerobically in the
presence of nitrate are incubated with respiratory substrates or reduced pyridine
nuc1eotides in the absence of nitrate, a reduced cytochrome b spectrum becomes
evident. Cytochrome b is reoxidized when nitrate is added. Thus it appears
that the electrons are transferred from these reductants along the respiratory
chain components to cytochrome b which in tum transfers electrons to the
Mo protein (bacterial nitrate reductase) as shown.

Sulfhydryl CN
Inhibitors HONO

*
reagents Azide

t
NO;

Enzyme
components
HS -
e-
Co-Q
-
Respiratory chain
Nitrate reductase,
(Mo protein)

f e-
Electron A =pyridine nucleotides NO;
donors or Reduced viologens
respiratory substrate
358 L. BEEVERS and R.H. HAGEMAN:

By using reduced viologen dyes as electron donor it has been established

that the Km for nitrate in K. aerogenes and Bacillus licheniformis is 0.11 mM
(\f ANT RIET et al. 1979).

3.3 Assimilatory Nitrate Reduction in Bacteria

Nitrate reductase has been demonstrated by various workers (see BEEVERS and
HAGEMAN 1980) in preparations from photosynthetic bacteria and from the
phototrophic bacterium. In the Cyanobacteria, nitrate reductase has been dem-
onstrated in Anabaena cylindrica, Nostoc muscorum and Anacystis nidulans.
Apart from the report of a pyridine nucleotide-dependent nitrate reductase in
extracts from Rhodospirillum rubrum, most other prokaryotes utilize reduced
flavins or reduced ferredoxin. Although a soluble enzyme has been isolated
from Rhodopseudomonas capsulata with a molecular weight of 180,000, nitrate
reductase in these photosynthetic bacteria and cyanobacteria is frequently asso-
ciated with particles. The enzymes from Anabaena and Anacystis have been
solubilized. The enzyme from Anacystis nidulans was ferredoxin-dependent and
composed of one polypeptide with molecular weight of 75,000 (\fENNESLAND
and GUERRERO 1980).

3.4 Characterization of Nitrate Reductase from Higher Plants

Although there has been an extensive amount of research on the enzyme since
nitrate reductase was first characterized and purified from higher plant extracts
by EVANS and NASON (1953), there is very little information on its physical
properties. This deficiency is partially accounted for by the difficulty experienced
in purifying the enzyme (HAGEMAN and HUCKLESBY 1971). More recent purifica-
tion techniques have involved affinity chromatography on blue dextran Sephar-
ose or blue Sepharose (CAMPBELL and SMARELLI 1978, NOTTON et al. 1977,
NOTTON and HEWITT 1979, Kuo et al. 1980). These highly purified preparations
have a considerably lower specific activity than the enzyme purified from Chlor-
ella (SOLOMONSON et al. 1975). The extensively purified enzyme from higher
plants had a molecular weight of about 200,000, a sedimentation coefficient
of about 8.0, a Stokes radius of about 6.0 nm and a frictional coefficient of
1.55, indicating an assymetric molecule which behaves anomalously on gel filtra-
tion (NOTTON and HEWITT 1979, Kuo et al. 1980). High molecular weights
reported on the basis of molecular sieving should be regarded ,as inaccurate.
Sedimentation coefficients of 8.15 have been reported for enzymes prepared
from tobacco cell tissue cultures (WRAY and FILNER 1970) and maize roots
(ASLAM and OAKS 1976, WALLACE and JOHNSON 1978). A major subunit of
molecular weight of about 100,000 was produced during SDS gel electrophoresis
of SDS and mercaptoethanol-dissociated nitrate reductase from barley (Kuo
et al. 1980). However, a much greater range of polypeptides was produced from
spinach (NOTTON and HEWITT 1979).
II.4 Uptake and Reduction of Nitrate: Bacteria and Higher Plants 359

Nitrate reduction is mediated by NADH in the majority of higher plant

systems (HAGEMAN and HUCKLESBY 1971). The purified enzyme from spinach
shows a specific preference for the A type stereisomer of NADH (GUERRERO
et al. 1977). Some species are able to utilize NADPH as well as NADH for
the reduction of nitrate (EVANS and NASON 1953, HAGEMAN and HUCKLESBY
1971, SHEN et al. 1976, REDINBAUGH and CAMPBELL 1981). This lack of cofactor
specificity is due to the presence of a second nitrate reductase which is functional
with either NADH or NADPH (CAMPBELL 1976, 1978, JOLLY et al. 1976, SHEN
et al. 1976). With soybeans, the NADH nitrate reductase had a molecular weight
of 330,000, while the second nitrate reductase which can use either NADH
or NADPH had a molecular weight of 220,000 (CAMPBELL 1976, JOLLY et al.
1976). STEWART and OREBAMJO (1979) have reported that the tropical legume
Erythrina senegalensis has only the non-specific NAD(P)H nitrate reductase.
Purified nitrate reductase shows diaphorase activity in which NADH can serve
to reduce cytochrome c, ferrocyanide, or other oxidants. Nitrate reduction can
be measured independently of the diaphorase by using reduced viologen dyes
or reduced flavins. Enzymic nitrate reduction mediated by either reduced pyri-
dine nucleotides or viologen dyes or flavins is inhibited by cyanide or azide,
however diaphorase activity is insensitive to these inhibitors (WALLACE and
JOHNSON 1978, WRAY and FILNER 1970). Nitrate or cytochrome c reduction
mediated by reduced pyridine nucleotides is prevented by p-chloromercuriben-
zoate and is relatively heat-sensitive; in contrast nitrate reduction mediated
by reduced viologens or FMNH2 is insensitive to sulfhydryl inhibitors and
is more heat-tolerant (SCHRADER et al. 1968).
The enzyme from higher plants contains flavin (EVANS and NASON 1953).
There appears to be variability in the tightness of the association of the flavin
with the enzyme. Preparations from some species show strong stimulation of
enzymic activity from addition of flavins, whereas preparations from other
species demonstrate limited response (HAGEMAN and HUCKLESBY 1971). Kinetic
properties of the NADH-dependent enzyme indicate a Km for nitrate and
NADH of 200/!M and 10/!M respectively with optimal activity at pH 7.4.
The NAD(P)H-dependent enzyme from soybean (PH 6.5 optimum) and Eryth-
rina have characteristically low affinity for nitrate with Km's for nitrate of
4.5 mM and 8-10 mM having been reported, respectively (CAMPBELL 1976,
JOLLY et al. 1976 and STEWART and OREBAMJO 1980). Km's for NADPH are
15 /!M and for NADH 1.4/!M and 3.9 /!M.
Purified plant nitrate reductase (CAMPBELL and SMARRELLI 1978 and EAGLES-
HAM and HEWITT 1975) shows biphasic responses to nitrate, implying negative
cooperativity towards nitrate. NADH-dependent nitrate reduction occurs by
an ordered ping-pong (by-by) mechanism. This is consistent with reduced pyri-
dine nucleotide being attached and its oxidized form released from a site sepa-
rated physically from a nitrate-binding site.
Purified nitrate reductase from spinach contains molybdenum (NoTTON and
HEWITT 1979) and heme (NOTTON et al. 1977). The heme apparently functions
in the enzyme component which shows diaphorase activity (NOTTON et al. 1979).
In purified preparations nitrate reduction is stimulated by phosphate (HAGEMAN
and HUCKLESBY 1971). The characteristics of nitrate reductase from plants can
be represented schematically as follows.
 360 L. BEEVERS and R.H. HAGEMAN:

Sulfuydryl I CN
Inhibitors
reagents I Azide
Heat (mild)

f ,- !- +-
r
I_e e
Holoenzyme of two parts, Flavin - - - Cyt b ~o-protein ------
I

NIDj/
a diaphorase (left) and a
molybdenum containing I
complex (right) I
I NO;
I
I
Electron donors NAD (P) H oxid I FMNH2
~ Oxidized
Reduced
or acceptors Cytc I Reduced
FMN I Viologen Dyes
DCIP or~o
I
I
I
The functional components of the enzyme have not been physically separated
in higher plants. However, the observations that tungstate, an analog of
molybdenum, inhibits nitrate reductase without impairing diaphorase activity
(WRAY and FILNER 1970, RUCKLIDGE et al. 1976) and that an inactive apoprotein
with diaphorase activity from molybdenum-deficient plants can be reactivated
by mixing with an acid-dissociated product of spinach nitrate reductase are
consistent with a two component model. Additionally, the demonstration of
tobacco mutants deficient in partial activities of nitrate reductase such as cyto-
chrome c reductase which can be activated by simultaneous extraction with
appropriate mutants suggests the in vitro reconstitution of the enzyme from
component subunits with partial activities (MENDEL and MULLER 1978).

4 Molybdenum in Nitrate Reduction

Although the presence of molybdenum in nitrate reductase from plant extracts

has been well documented (HEWITT 1974) and the essentiality, of the metal
demonstrated from deficiency experiments and tungstate antagonism studies,
the mode of operation of molybdenum in reduction is not understood (HEWITT
1974, BEEVERS and HAGEMAN 1980).
Similar uncertainty exists about the role of action of molybdenum in nitrate
reduction in bacterial systems (STOUTHAMER 1976).
11.4 Uptake and Reduction of Nitrate: Bacteria and Higher Plants 361

5 Nitrite Reduction

5.1 Assimilatory Bacteria

Three different assimilatory nitrite-reducing enzymes have been detected in bac-

terial extracts (see STOUlHAMER 1976). The bulk of nitrite assimilation appears
to be mediated by an NADH-dependent nitrite reductase which has been extensi-
vely characterized from Escherichia coli. The purified enzyme has a molecular
weight of 190,000 and a sedimentation coefficient of 8.5-9.5. On SDS polyacryl-
amide electrophoresis a subunit of molecular weight of 88,000 is detected. The
purified enzyme contains flavin but no detectable heme. Catalylic activity is
inhibited by p-chloromercuribenzoate and cyanide. Three molecules of NADH
are oxidized for each NO;- ion reduced and the product of the reaction is
assumed to be ammonium. The enzyme also possesses hydroxylamine reductase
activity (COLEMAN et al. 1978).

5.2 Dissimilatory Bacteria

When denitrifying bacteria are grown anaerobically on nitrite, nitrogen gas

is usually released. However, if cell-free extracts are incubated with nitrite under
anaerobic conditions, nitrous oxide and nitric oxide are produced in addition
to nitrogen (PAYNE 1973, THAUER et al. 1977). The reduction of nitrite to nitro-
gen requires both soluble and particulate components and NADH or respiratory
intermediates such as succinate or lactate (Cox and PAYNE 1973, PAYNE 1973,
THAUER et al. 1977). Dissimilation occurs in two steps; nitrite is first reduced
to nitric oxide by an apparently soluble dissimilatory nitrite reductase. A second
enzmye, nitric oxide reductase, converts nitric oxide to nitrous oxide and a
membrane-associated nitrous oxide reductase converts nitrous oxide to nitrogen.
The enzymes have not been extensively characterized. Nitrite reductase and
nitric oxide reductase usually contain cytochromes which appear to form heme
complexes with the nitrogen oxides (PAYNE et al. 1971, Cox and PAYNE 1973).
The enzymatic conversion of nitrous oxide to nitrogen is inhibited by CO,
CN and azide, indicating the involvement of metals; however these have not
been characterized (THAUER et al. 1977).

5.3 Nitrite Reductase in Plants

Nitrite reductase converts nitrite to ammonium, the six-electron reduction of

nitrite occuring without the release of free intermediates. The nitrite reductase
from leaves has been extensively purified and shown to be dependent upon
reduced ferredoxin as reductant (BEEVERS and HAGEMAN 1980, HEWITT et al.
1976, VENNESLAND and GUERRERO 1980). An enzyme with similar properties
has been isolated from non-green tissue (DALLING et aI. 1973, HUCKLESBY et al.
362 L. BEEVERS and R.H. HAGEMAN:

1972) although the enzyme from non-green tissue can utilize reduced dyes or
ferredoxin as electron donor, the in vivo reductant is not known.
The enzyme, purified from green leaves by a variety of techniques, but more
recently by affinity chromatography on ferredoxin-Sepharose (IDA 1977), has
a molecular weight of 62,000 and is not dissociated by SDS (IDA 1977, VEGA
and KAMIN 1977, HUCKLESBY et al. 1976). Three atoms of iron are present
per mol of enzyme, one iron atom is associated with a heme component identi-
fied as siroheme (VEGA and KAMIN 1977, HUCKLESBY et al. 1976, MURPHY et al.
1974) and the other two iron atoms are associated with two labile sulfides
(i.e. binuclear iron-sulfur center Fe2 S2 ) (HUCKLESBY et al. 1979, VEGA and KA-
MIN 1977, APARICIO et al. 1975). In contrast, the work of LANCASTER et al.
(1979) indicates that the spinach enzyme has 6 mol of iron and 4 mol of acid-
labile sulfur mol- 1 of siroheme (i.e. the enzyme has a tetranuclear iron-sulfur
center).
Electron paramagnetic resonance studies and spectral data indicate the oxi-
dation reduction of iron-sulfur centers and the formation of a NO-heme com-
plex, however other intermediates in the reduction sequence have not been
demonstrated (APARICIO et al. 1975, HUCKLESBY et al. 1979, VEGA and KAMIN
1977). The suggested sequence for reduc.tion of nitrite is an initial reduction
of siroheme by one electron from ferredoxin which permits the formation of
a nitrite-heme complex. Formation of this complex facilitates reduction of the
iron-sulfur center. The further transformation ofthe NO-siroheme to an enzyme
NH3 complex involves the addition of five electrons. The following sequence
shows that these electrons may be transferred directly from ferredoxin or via
the iron center or pairs of electrons may be added simultaneously, one from
each source.

ox
Fe x Sx
Iron center

Primary e
donor

Siroheme-NO ox
comPl"Je

NH3
red

Proposed reaction scheme for nitrite

reduction by chlorophyllus tissue.

The sequence of additions and the intermediates formed have not been deter-
mined (HUCKLESBY et al. 1979, LANCASTER et al. 1979).
11.4 Uptake and Reduction of Nitrate: Bacteria and Higher Plants 363

6 Location of Enzymes of Nitrate Assimilation

in Higher Plants

The observation that living plant cells and organs can be cultured and grown
on nitrate as the primary nitrogen source constitutes convincing evidence for
the ubiquitous occurrence of nitrate and nitrite reductases in plant tissues. How-
ever, in certain organs and tissues the genetic potential for the synthesis of
these enzymes may not be expressed as, for example, in the absence of nitrate
reductase in roots of Xanthium pennsylvanicum (PATE 1973) or the location
of nitrate and nitrite reductase in the mesophyll cells but not in the bundle-sheath
cells of plants showing C 4 photosynthesis (HAREL et al. 1977).
The intracellular localization of nitrate and nitrite reductase has not yet
been clearly resolved. The bulk of the studies indicate that nitrite reductase
in green leaves is associated with chloroplasts (BEEVERS and HAGEMAN 1980,
VENNESLAND and GUERRERO 1980). This chloroplast location is confirmed by
the observations that chloroplasts can reduce nitrite in the light (NEYRA and
HAGEMAN 1974, MAGALHAES et al. 1974, MIFLIN 1974). A particulate association
of nitrite reductase in roots was originally implied from the observation by
MIFLIN (1967) and more recently nitrite reductase has been localized from pro-
plastids from roots (DALLING et al. 1972b, EMES and FOWLER 1979).
Nitrate reductase appears to be soluble (HEWITT et al. 1976). Suggestions
that the nitrate reductase may be associated with chloroplasts or microbodies
in leaf tissue have more recently been interpreted as representing indiscriminate
binding of the enzyme to organelle surfaces (DALLING et al. 1972a). However,
CAMPBELL and SMARELLI (1979) point out the similarities of certain properties
of nitrate reductase and membrane-associated enzymes and suggest that the
enzyme is complexed with a membrane from which it is easily solubilized during
isolation.

7 Provision of Reductant for Nitrate Assimilation

in Higher Plants

The reduction of nitrate in photosynthetic tissue has been closely linked to

light since the early observations of W ARBURG and NEGELEIN (1920) that light
stimulated nitrate utilization in Chlorella. However, the pyridine nucleotide spe-
cificity and the intracellular location of nitrate reductase appears inconsistent
with a direct involvement of photosynthetic products (BEEVERS and HAGEMAN
1980, HEWITT et al. 1976).
Current views of the source of NADH for nitrate reduction in leaves have
evolved from the use of the in vivo assay for nitrate reductase. Because the
assay is conducted under non-physiological conditions (tissue submerged under
water, dark, anaerobic) and based upon the accumulation of nitrite, extrapola-
tion of these data to in situ nitrate reduction must be 'treated with reservations.
364 L. BEEVERS and R.H. HAGEMAN:

The failure of the leaves to assimilate nitrite under dark anaerobic conditions
is attributed to the unavailability of reduced ferredoxin. Under in vivo assay
conditions in certain species, 3-P-glyceraldehyde and glucose 6-P are more effec-
tive than 6-P gluconate or Krebs cycle acids in enhancing nitrite accumulation
(KLEPPER et al. 1971, MANN et al. 1978). These results suggest that the oxidation
of 3-P-glyceraldehyde rather than the pentose pathway or the mitochondria
is the source of NADH for nitrate reduction. ASLAM et al. (1979) demonstrated
that barley seedlings can utilize recently fixed photosynthate or stored reserves
to reduce nitrate; however they did not indicate the enzyme or enzyme system
that generates the reductant. In contrast, MULDER et al. (1959), SAWHNEY et al.
(1978), HAGEMAN et al. (1980) and Woo et al. (1980) have shown that addition
of various Krebs cycle acids to the in vivo assay enhanced nitrite accumulation,
thus implicating mitochondrial involvement in reductant generation. Use of
the in vivo assay with roots also results in the accumulation of nitrite (LEE
1979). Because roots lack ferredoxin, such results indicate that the anaerobic
conditions impair the generation of reductant for nitrite assimilation.
The reductant for nitrite assimilation in green leaves is ferredoxin. In the
light the reduced form is generated via the photosynthetic apparatus. In leaves
in the dark, reduced ferredoxin could be produced via the ferredoxin-NADP-
oxidoreductase given an ample supplyofNADPH. The in situ reductant for
root nitrite reductase is not known. BUTT and BEEVERS (1961) found that root
supplied with nitrite demonstrated enhanced pentose pathway activity, however
in vitro, the root enzyme resembles the leaf enzyme in that it can use reduced
ferredoxin or viologen dyes but not NAD(P)H as reductant. Enzymes of the
pentose pathway and nitrite reductase are associated with the same particulate
fraction from root preparations (EMES and FOWLER 1979) and exposure of roots
to nitrate induces a particulate 6-P-gluconate dehydrogenase (EMES et al. 1979).
A current hypothesis is that roots contain an unknown compound that serves
as an electron carrier between NADPH and nitrite.

8 Regulation of Nitrate Reductase in Higher Plants

8.1 Substrate

Nitrate reductase is present only at very low levels in most plants (soybean
is one exception) not receiving nitrate. The supplying of nitrate results in a
great increase in enzyme level. This nitrate-dependent increase in extractable
nitrate reductase is prevented by inhibitors of protein and nucleic acid synthesis,
indicating that the influence of nitrate reductase is one of induction rather
than activation of some inert precursor (HEWITT et al. 1976). This concept was
confirmed by ZIELKE and FILNER (1971), who have demonstrated de novo syn-
thesis of the enzyme in response to nitrate.
The requirement for sustained RNA synthesis for induction implies that
nitrate may be acting as a coinducer of nitrate reductase in the classical sense
IIA Uptake and Reduction of Nitrate: Bacteria and Higher Plants 365

proposed by JACOB and MONOD (1961). However, confirmation of this idea

has not been provided, and arguments have been proposed that in fungal and
bacterial systems the enzyme may be constitutively synthesized at low levels
and is synthesized at high levels in the presence of nitrate. The constitutively
formed enzyme may be partly responsible for its own repression (STOUTHAMER
1976, COVE and PATEMAN 1969, DANTZIG et al. 1978).
In some instances nitrite may activate a constitutive nitrate reductase compo-
nent (INGLE et al. 1966, KAPLAN et al. 1978) and organic nitro compounds may
stimulate the production of nitrate reductase (SHEN 1972). During rapid vegeta-
tive development of cereal crops grown under normal conditions, availability
of nitrate in the cells of the leaf blade is a major factor affecting in situ nitrate
reduction rate and the diurnal fluctuations of enzyme level (HAGEMAN 1979).
Nitrate reductase was induced in cucumber cotyledons incubated in various
organic acids (especially citrate and to a lesser extent by ascorbate) or
(NH 4hS04 at pH 3.0 (FERGUSON and KNYPL 1974 and KNYPL and FERGUSON
1975). These compounds were ineffective inducers at pH 6.0. Although the phy-
siological significance of the induction at pH 3.0 is not known (FERGUSON and
KNYPL 1974), these results, in conjunction with the observation that ascorbate
induces nitrate reductase in Spirodela oligorrhiza, indicate that the control of
induction of nitrate reductase does not specifically require nitrate under all
conditions.

8.2 Hormonal

Various growth regulators influence the tissue level of nitrate reductase. In

Agrostemma githago embryo and cucumber cotyledons the enzyme is induced
by cytokinins (HIRSCHBERG et al. 1972, KENDE and SHEN 1972, KNYPL 1973).
The growth retardant succinic acid 2-2 dimethyl hydrazide stimulates enzyme
production in cucumber cotyledons. Inhibitor studies and density labeling stu-
dies demonstrate that enzyme production in response to cytokinins is due to
de novo synthesis (HIRSCHBERG et al. 1972, KENDE and SHEN 1972).

8.3 Molybdenum

The production of an active nitrate reductase in response to nitrate depends

upon a supply of molybdate. Molybdenum-deficient fungi (DANTZIG et al. 1978),
higher plants (NOTTON et al. 1974), algae (VEGA et al. 1971) and bacteria (ENOCH
and LESTER 1972) or organisms gtown in conditions where tungstate or vana-
dium replaces molybdenum produce aberrant nitrate reductases which possess
only partial enzyme activity. The enzyme from molybdenum-deficient plants,
which possesses NADH-cytochrome c reductase activity, cross-reacts with anti-
serum produced against active nitrate reductase. The molybdenum-free apoen-
zyme can be activated by addition of a molybdenum-containing complex ob-
tained by acid washing the native enzyme (NOTION and HEWITT 1979).
366 L. BEEVERS and R.H. HAGEMAN:

8.4 Ammonium

Ammonium inhibits the nitrate-stimulated production of nitrate reductase in

barley roots (SMIlH and THOMPSON 1971), Spirodela oligorrhiza (FERGUSON
1969), Lemna minor (JOY 1969, OREBAMJO and STEWART 1974, 1975a) and in
roots but not shoots of cotton plants (RADIN 1975) or in shoot tissues of other
higher plants (OAKS et al. 1977). The repression of nitrate reductase in Lemna
minor is thought to be mediated by some product of ammonium assimilation
(e.g. by glutamine rather than by ammonium per se) (STEWART and RHODES
1978). Work with rose tissue culture (MOHANTY and FLETCHER 1980) and mung
bean hypocotyls (MOHANTY 1979) also indicated that the ammonium effect
on nitrate reductase was mediated via glutamine. When Spirodela was supplied
with equimolar levels of nitrate and glutamine, nitrate and asparagine or nitrate,
the levels of extractable nitrate reductase were suppressed by the presence of
glutamine and enhanced by the presence of asparagine relative to the nitrate
control (FERGUSON 1970). Utilization of nitrate was largely prevented by the
presence of ammonium (FERGUSON 1969) and to a lesser extent by the amino
compounds (FERGUSON 1970).

8.5 Light

Leaves of plants grown in shaded conditions accumulate nitrate and have low
levels of extractable nitrate reductase. When plants cultured in nitrate are trans-
ferred to darkness, extractable nitrate reductase declines and during reillumina-
tion enzyme level increases. The mechanism by which light influences the nitrate-
mediated production of nitrate reductase is not understood, but two proposals
have been advanced. The light stimulation may be associated with increased
capacity for protein synthesis in illuminated leaves (TRAVIS and KEY 1971).
ASLAM et al. (1973) suggest that the main effect of light may be to supply
photosynthate to support respiration, which in tum drives the induction process.
Alternatively or coincidentally light may regulate the availability of nitrate at
the induction site (BEEVERS et al. 1965, JONES and SHEARD 1975). Maximal light
stimulation is dependent on the presence of functional chloroplasts (SLUITERS-
SCHOLTEN 1975, SAWHNEY and NAIK 1972) and appears to be phytochrome-
mediated (JONES and SHEARD 1975, JOHNSON 1976).

8.6 Genetic

Nitrate reductase mutants have been identified and characterized in several

bacteria (STOUlHAMER 1976). The physiological properties of the mutants have
been described and the location, on the circular bacterial"chromosome" of
the genes regulating nitrate reductase have been determined. Mutants deficient
in various components of the complex nitrate reductase molecule have been
identified and in some instances active forms of the enzyme can be formed
by mixing components extracted from appropriate mutants (STOUTHAMER 1976).
11.4 Uptake and Reduction of Nitrate: Bacteria and Higher Plants 367

A similar complex series of mutations affecting nitrate reductase has been

described in the fungi Neurospora and Aspergillus (CODDINGTON 1976, COVE
1970). Mutants deficient in the diaphorase (NADPH cytochrome c reductase)
component or lacking the molybdenum cofactor have been characterized and
appropriate complementation experiments with Neurospora have resulted in the
formation of active enzymes (LEE et al. 1974).
Mutants of the alga Chlamydomonas reinhardii and the higher plant Arabi-
dopsis deficient in their capacity to utilize nitrate have been developed by the
use of the mutagen N-methyl-N'-nitro-N-nitrosoguanidine. Biochemical and
physiological studies indicate that the mutations variously influence nitrate up-
take (SOSA et al. 1978, OosTINDmR-BRAAKSMA and FEENSTRA 1973), as well as
activities of various components of the enzyme (DODDEMA et al. 1978, SOSA
et al. 1978). Mutants deficient in partial activities of nitrate reductase have
been developed in barley (TOKAREV and SHUMNYI 1977, WARNER et al. 1977)
and tobacco (MENDEL and MULLER 1978). The barley mutants characterized
by WARNER et al. (1977), although possessing only traces of nitrate reductase
in the leaf, still possessed nitrite reductase activity and were able to grow on
nitrate as sole nitrogen source. By homogenizing appropriate mutants together,
active enzymes could be detected in extracts of tobacco, indicating in vitro
reconstitution of the enzyme from components with only partial activities (MEN-
DEL and MULLER 1978).
Varietal variations in the level of extractable nitrate reductase have been
demonstrated in many agronomic species. In com, the variability has been
explained on a two-gene model with the genes regulating rate of synthesis and
rate of decay (WARNER et al. 1969).

8.7 In Vivo Controls

The level of extractable nitrate reductase activity fluctuates in response to

drought, and leaf water potential, temperature, light intensity and time of day.
These changes may occur without any detectable alterations in endogenous
nitrate level. SHANER and BOYER (1976) suggest that it is the flux of nitrate,
from the apoplast medium into the cytoplasm, rather than the total nitrate
which controls in vivo levels of activity. It is generally considered that variations
in extractable activity are due to changes in the level of enzyme protein. The
observed turnover of nitrate reductase in tobacco cells (ZmLKE and FILNER
1971) and the detection of proteases that preferentially attack nitrate reductase
(WALLACE 1974,1975) are consistent with enzyme degradation rather than inac-
tivation, accounting for declines in enzyme activity. In some but not all instances
a decline in nitrate reductase can be prevented by cycloheximide, implying a
requirement for the appropriate protease in order to degrade the reductase.
The reports of reduced enzyme level in response to environmental conditions
are based on measurements of activity and not on actual enzyme protein. How-
ever, reduction of in vitro activity could equally well be brought about by
enzyme inactivation rather than destruction of the enzyme. Although convincing
evidence for inactivation of nitrate reductase in algae by NADH and CN-
368 L. BEEVERS and R.H. HAGEMAN:

has been provided, evidence for modulation of activity of higher plant nitrate
reductase by changes in oxido-reduction status is not so apparent. In most
higher plant systems the presence of reductants appears to stabilize the enzyme
(HAGEMAN and HUCKLESBY 1971). Nitrate reductase activity may be modulated
by endogenous inhibitors which have been identified in extracts of soybeans
(JOLLY and TOLBERT 1978).
Although variations in oxido-reduction status and presence of proteases
and inhibitors may be associated with fluctuations in measured levels of extract-
able nitrate reductase, it is not established to what extent these controls operate
in situ. Attempts to estimate the in situ rate of nitrate reduction have been
based on coniparisons of the amount of reduced nitrogen made available as
estimated by the anaerobic in vivo assay of nitrate reductase of the entire seed-
ling with the actual increase in reduced N recorded in the seedling (BRUNETTI
and HAGEMAN 1976). CHANTAROTWONG et al. (1976) have devised an aerobic
in vivo assay for nitrate reductase based on the difference between total nitrate
uptake by the intact plant and the residual nitrate found in the plant over
timed intervals. No comparisons were made between the estimated input of
reduced N and the actual changes in reduced N. OAKS et al. (1979) indicate
that the in vivo assay based on nitrite production under anaerobiosis does
not coincide with nitrate reduction based on incorporation from 15N03 and
it is this latter measurement which most reliably measures nitrate metabolism.
With leaves of some species at certain stages of development, products other
than nitrite may be produced in the in vivo assay.

9 Concluding Thoughts

Current work indicates that nitrate uptake is an active process that may involve
a proteinaceous carrier that is rather specific for nitrate. The understanding
of the precise mechanism of uptake and the distribution and partitioning of
nitrate ·among the various organs and within the cell between cytoplasm and
vacuole is woefully inadequate.
The improvements in enzyme purification and characterization of nitrate
reductase and the recent findings that siroheme and non-heme iron centers
are the active components of nitrite reductase, and that glutamine rather than
glutamate is the first product of ammonia assimilation, have enhanced our
understanding of mechanisms and pathways of nitrate assimilation in higher
plants. The immediate electron donor to nitrite reductase in roots and the pri-
mary sources of energy for regeneration of the reductants for nitrate and nitrite
reductase require additional clarification.
Although the outline of the pathway and mode of action of enzymes involved
in nitrate assimilation are reasonably well established, it is anticipated that
future studies will result in an understanding of the mechanism of the regulation
of enzyme level and methods by which manipulation of nitrate reduction may
regulate crop productivity.
11.4 Uptake and Rllduction of Nitrate: Bacteria and Higher Plants 369

References
Aparicio PJ, Knaff DB, Malkin R (1975) Role of an iron-sulfur center and siroheme
in spinach nitrite reductase. Arch Biochem Biophys 169:102-107
Arnon DI (1937) Ammonium and nitrate nutrition of barley at different seasons in
relation to hydrogen-ion concentration, manganese, copper and oxygen supply. Soil
Sci 44:91-120
Aslam M, Oaks A (1976) Comparative studies on the induction and inactivation of
nitrate reductase in corn roots and leaves. Plant Physiol 57: 572-576
Aslam M, Huffaker RC, Travis LC (1973) The interaction of respiration and photosynthe-
sis in induction of nitrate reductase activity. Plant Physiol 52: 137-141
Aslam M, Huffaker RC, Rains DW, Rao KP (1979) Influence of light and ambient
carbon dioxide concentration on nitrate assimilation by intact barley seedlings. Plant
Physiol 63: 1205-1209
Bayley JM, King J, Gamborg OL (1972) The effect of the source of inorganic nitrogen
on growth and enzymes of nitrogen assimilation in soybean and wheat cells in suspen-
sion cultures. Planta 105: 15-24
Beevers L, Hageman RH (1965) The role of light and nitrate in the induction of nitrate
reductase in radish cotyledons and maize seedlings. Plant PhysioI40:691-698
Beevers L, Hageman RH (1980) Nitrate and nitrite reduction in the biochemistry of
plants. In: Miflin BJ (ed) A comprehensive treatise, vol V. Academic Press, London
New York
Beevers L, Schrader LE, Flesher D, Hageman RH (1965) The role of light and nitrate
in induction of nitrate reductase in radish cotyledons and maize seedlings. Plant
Physiol 40: 691-698
Ben Zioni A, Vaadia Y, Lips SH (1971) Nitrate uptake by roots as regulated by nitrate
reduction products of the shoot. Physiol Plant 24:288-290
Blevins DG, Hiatt AJ, Lowe RH (1974) The influence of nitrate and chloride uptake
on expressed sap pH, organic acid synthesis, and potassium accumulation in higher
plants. Plant Physiol 54: 82-87
Bollard EG (1966) A comparative study of the ability of organic nitrogen compounds
to serve as sole sources of nitrogen for the growth of plants. Plant Soil 25: 153-166
Brunetti N, Hageman RH (1976) Comparison of in vivo and in vitro assays of nitrate
reductase in wheat (Triticum aestivum L.) seedlings. Plant Physiol 58: 583-587
Burke KA, Lascelles J (1979) Partial purification and some properties of the Staphylococ-
cus aureus cytoplasmic nitrate reductase. J Bacteriol139: 120-125
Butt VS, Beevers H (1961) The regulation of pathways of glucose catabolism in maize
roots. Biochem J 80: 21-27
Butz RG, Jackson WA (1977) A mechanism for nitrate transport and reduction. Phyto-
chemistry 16:409-417
Campbell WH (1976) Separation of soybean leaf nitrate reductases by affinity chromatog-
raphy. Plant Sci Lett 7:239-244
Campbell WH (1978) Isolation of NAD(P)H: nitrate reductase from the scutellum of
maize. Z Pflanzenphysiol 88: 357-361
Campbell WH, Smarelli J (1978) Purification and kinetics of higher plant NADH: nitrate
reductase. Plant Physiol61: 611-616
Campbell WH, Smarelli J (1979) Observations on the physical state of NADH: nitrate
reductase. Plant Physiol 63 (Supp!) 249
Chantarotwong W, Huffaker RC, Miller BL, Granstedt RC (1976) In vivo nitrate r~duc­
tion in relation to nitrate uptake, nitrate content and in vitro nitrate reductase activity
in intact barley seedlings. Plant Physiol 57: 519-522
Coddington A (1976) Biochemical studies on the Nit mutants of Neurospora crassa.
Mol Gen Genet 145:195-206
Coleman KJ, Cornish-Bowden A, Cole JA (1978) Purification and properties of nitrite
reductase from Escherichia coli K12. Biochem J 175:483-493
Cove DJ (1970) Control of gene action in Aspergillus nidulans. Proc R Soc London
Ser B 176:267-275
370 L. BEEVERS and R.H. HAGEMAN:

Cove DJ, Pateman JA (1969) Autoregulation of the synthesis of nitrate reductase in

Aspergillus nidulans. J Bacteriol97: 1374-1378
Cox GB, Payne WJ (1973) Separation of soluble denitrifying enzymes and cytochromes
from Pseudomonas perfectomarinus. Can J MicrobioI19:861-872
Cox WJ, Reisenauer HM (1973) Growth and ion uptake by wheat supplied nitrogen
as nitrate, or ammonium or both. Plant Soil 38: 363-380
DaIling MJ, Tolbert NE, Hageman RH (1972 a) Intracellular location of nitrate reductase
and nitrite reductase I. Spinach and tobacco leaves. Biochim Biophys Acta
283:505-512
Dalling MJ, Tolbert NE, Hageman RH (1972b) Intracellular location of nitrate reductase
and nitrite reductase II. Wheat roots. Biochim Biophys Acta 283:513-519
DaIling MJ, Hucklesby DP, Hageman RH (1973) A comparison of nitrite reductase
enzymes from green leaves, scutella and roots of com (Zea mays L.). Plant Physiol
51 :481-484
Daniel RM, Gray J (1976) Nitrate reductase from anaerobically grown Rhizobiumjaponi-
cum. J Gen Microbiol 96:247-251
Dantzig AH, Zurowsk WK, Ball TM, Nason A (1978) Induction and repression of
nitrate reductase in Neurospora crassa. J Bacteriol 133: 671-679
Davies DD (1973) Control of and by pH. In: D.D. Davies: Rate control of biological
processes. SEB Symp 27:513-530
Dijkshoom W (1962) Metabolic regulation of the alkaline effect of nitrate utilization
of plants. Nature 194: 165-166
Doddema H, Telkamp GP (1979) Uptake of nitrate by mutants of Arabidopsis thaliana
disturbed in uptake or reduction of nitrate II. Kinetics. Physiol Plant 45: 332-338
Doddema H, Hoffstra JJ, Feenstra WJ (1978) Uptake of nitrate by mutants of Arabidopsis
thaliana disturbed in uptake and reduction of nitrate I. Effect of nitrogen source
during growth on uptake of nitrate and chlorate. Physiol Plant 43: 343-350
Driesche van den R (1978) Response of Douglas fir seedlings to nitrate and ammonium
nitrogen sources at different levels of pH and iron supply. Plant Soil 49:607-623
Eaglesham AR, Hewitt EJ (1975) Inhibition of nitrate reductase from spinach (Spinacea
oleracea) leaf by adenosine nucleotides. Plant Cell PhysioI16:1137-1149
Emes MJ, Fowler MW (1979) The intracellular location of the enzymes of nitrate assimi-
lation in the apices of seedling pea roots. Planta 144: 249-253
Emes MJ, Ashihara H, Fowler MW (1979) The influence ofN0 3 on particulate 6-phos-
phogluconate dehydrogenase activity in pea roots. FEBS Lett 105:370-373
Enoch HG, Lester RL (1972) Effects of molybdate, tungstate, and selenium compounds
on formate dehydrogenase and other enzyme systems in Escherichia coli. J Bacteriol
110: 1032-1040
Epstein E (1953) Absorption of nitrate affected by bromide. Nature 171: 83
Epstein E (1972) In: Mineral nutrition of plants, principles, and perspectives. Wiley
and Sons, New York
Evans HJ, Nason A (1953) Pyridine nucleotide nitrate reductase from extracts of higher
plants. Plant Physiol 28: 233-254
Ferguson AR (1969) Nitrogen metabolism of Spirodela oligorrhiza. II. Control of the
enzymes of nitrate assimilation. Planta 88: 353-363
Ferguson AR (1970) Nitrogen metabolism of Spirodela oligorrhiza. III. Amino acids
and the utilization of nitrate. Planta 90: 365-369
Ferguson AR, Bollard EG (1969) Nitrogen metabolism of Spirodela oligorrhiza I. Utiliza-
tion of ammonium nitrate and nitrite. Planta 88: 344-352
Ferguson AR, Knypl JS (1974) Specificity of induction of nitrate reductase in plants.
Proc 7th Int Colloq, Hannover
Frith GJT (1972) Effect of ammonium nutrition on the activity of nitrate reductase
in roots of apple seedlings. Plant Cell Physiol 13: 1085-1090
Guerrero MG, Vega JM, Leadbetter E, Losada M (1973) Preparation and characteriza-
tion of a soluble nitrate reductase from Azotobacter chroococcum. Arch Mikrobiol
91 :287-304
Guerrero M G, J etschman K, Volker W (1977) The stereospecificity of nitrate reductase for
hydrogen removal from reduced pyridine nucleotides. Biochim Biophys Acta 482: 19-26
11.4 Uptake and Reduction of Nitrate: Bacteria and Higher Plants 371

Guerrero MG, Vega JM, Losada M (1981) The assimilatory nitrate-reducing system
and its regulation. Annu Rev Plant Physiol32: 169-204
Hageman RH (1979) Integration of nitrogen assimilation in relation to yield. In: Hewitt EJ,
Cutting CV (eds) Nitrate assimilation of plants. Academic Press, London New York
Hageman RH (1980) Effect of form of nitrogen on plant growth. In: Meisinger JJ,
Randall GW, Vitosh ML (eds) Nitrification inhibitors - potentials and limitations,
Spec Publ Am Soc Agron Madison 38:47-62
Hageman RH, Hucklesby DP (1971) In: San Pietro A (ed) Methods of enzymology,
vol 23. Academic Press, London New York
Hageman RH, Reed AJ, Femmer RA, Sherrard JE, Dalling MJ (1980) Some new aspects
of the in vivo assay for nitrate reductase in wheat (Triticum aestivum L.) leaves.
I. Reevaluation of nitrate pool size. Plant Physiol 65: 27-32
Harel C, Lea PJ, Millin BJ (1977) The localization of enzymes of nitrogen assimilation
in maize leaves and their activities during greening. Planta 134: 195-200
Haynes RJ, Goh KM (1978) Ammonium and nitrate nutrition of plants. BioI Rev
53:465-510
Heimer YM, Filner P (1970) Regulation of the nitrate assimilation pathway of cultured
tobacco cells II. Properties of a variant cell line. Biochim Biophys Acta 215: 152-165
Heimer YM, Filner P (1971) Regulation of the nitrate assimilation pathway in cultured
tobacco cells III. The nitrate uptake system. Biochim Biophys Acta 230:362-372
Hewitt EJ (1974) MTP Int Rev Sci Biochem Serl11: 199-245
Hewitt EJ (1975) Assimilatory nitrate-nitrite reduction. Annu Rev Plant Physiol
26:73-100
Hewitt EJ, Cutting CV (1979) In: Nitrogen assimilation of plants. Academic Press,
London New York
Hewitt EJ, Hucklesby DP, Notton BA (1976) In: Bonner J, Varner JE (eds) Plant bio-
chemistry. Academic Press, London New York
Hirschberg K, Hubner G, Borriss H (1972) Cytokinin-induzierte de novo Synthese der
Nitratreductase in Embryonen von Agrostemma githago. Planta 108: 333-337
Honert van den TH, Hooymans JJ (1955) On the absorption of nitrate by maize in
water culture. Acta Bot Neerl4:376-384
Huber VM, Warren HC, Nelson DW, Tsai CY (1977) Nitrification inhibitors - New
tools for food production. BioScience 27: 523-529
Hucklesby DP, Dalling MJ, Hageman RH (1972) Some properties of two forms of
nitrite reductase from corn (Zea mays L.) scutellum. Planta 104:220-233
Hucklesby DP, James DM, Banwell MJ, Hewitt EJ (1976) Properties of nitrite reductase
from Cucurbita pepo. Phytochemistry 15: 599-603
Hucklesby DP, Cammack R, Hewitt EJ (1979) Properties and mechanisms of nitrite
reductase. In: Hewitt EJ, Cutting CV (eds) Nitrogen assimilation in plants. Academic
Press, London New York
Hutchinson GL, Millington RJ, Peters DB (1972) Atmospheric ammonia: Absorption
by plants. Science 175: 771-772
Ida S (1977) Purification to homogeneity of spinach nitrite reductase by ferredoxin-
Sepharose affinity chromatography. J Biochem (Tokyo) 82:915-918
Ingle J, Joy KW, Hageman RH (1966) The regulation of activity of the enzymes involved
in the assimilation of nitrate by higher plants. Biochem J 100:577-588
Israel DW, Jackson WA (1982) Ion balance, uptake and transport processes in N 2 -fixing
and nitrate and urea-dependent soybean plants. Plant Physiol 69: 171-178
Jackson WA, Williams DC (1968) Nitrate-stimulated uptake and transport of strontium
and other cations. Soil Sci Soc Am Proc 32: 689-704
Jackson WA, Flesher D, Hageman RH (1973) Nitrate uptake by dark-grown corn seed-
lings: Some characteristics of apparent induction. Plant Physiol51: 120-127
Jacob F, Monod J (1961) Genetic regulatory mechanisms in the synthesis of proteins.
J Mol BioI 3:318-356
Jansson SL (1971) In: McClaren AP, Skujinis J (eds) Soil biochemistry, vol II. Dekker,
New York
Johnson CB (1976) Rapid activation by phytochrome of nitrate reductase in the cotyledon
of Sinapis alba. Planta 128:127-131
372 L. BEEVERS and R.H. HAGEMAN:

Jolly SO, Tolbert NE (1978) NADH-Nitrate reductase inhibitor from soybean leaves.
Plant Physiol 62: 197-203
Jolly SO, Campbell WH, Tolbert NE (1976) NADPH and NADH nitrate reductases
from soybean leaves. Arch Biochem Biophys 174:431-439
Jones RW, Sheard RW (1975) Phytochrome, nitrate movement and induction of nitrate
reductase in etiolated pea terminal buds. Plant Physiol 55: 954-959
Joy KW (1969) Nitrogen metabolism of Lemna minor II. Enzymes of nitrate assimilation
and some aspects of their regulation. Plant PhysioI44:849-853
Kaplan D, Mayer AJ, Lips SH (1978) Nitrite activation of nitrate reductase in higher
plants. Planta 138:205-209
Kende H, Shen TC (1972) Nitrate reductase in Agrostemma githago comparison of the
inductive effects of nitrate and cytokinin. Biochim Biophys Acta 286: 118-125
Kessler E (1976) Metabolism of inorganic nitrogen compounds. Proc Bot 38:108-117
Kirkby EA, Hughes AD (1979) In: Kirkby EA (ed) Nitrogen nutrition of the plant.
Waverley Press, Leeds
Kirkby EA, Knight AH (1977) Influence of the level of nitrate nutrition on ion uptake
and assimilation, organic acid accumulation and cation-anion balance in whole
tomato plants. Plant PhysioI60:349-353
Klepper LA, Flesher D, Hageman RH (1971) Generation of reduced nicotinamide
adenine dinucleotide for nitrate reduction in green leaves. Plant PhysioI48:58Q-590
Knypl JS (1973) Synergistic induction of nitrate reductase activity by nitrate and benzyl-
amino purine in detached cucumber cotyledons. Z Pflanzenphysiol 70: 1-11
Knypl JS, Ferguson AR (1975) pH-Dependent induction of nitrate reductase in cucumber
cotyledons by citric acid and other compounds. Z Pflanzenphysiol 74:434-439
Kuo T, Kleinhoff A, Warner RL (1980) Purification and partial characterization of
nitrate reductase from barley leaves. Plant Sci Lett 17: 371-381
Lancaster JR, Vega JM, Kamin H, Orme-Johnson NR, Orme-Johnson WH, Krueger
RJ, Siegel LM (1979) Identification of the iron-sulfur center of spinach ferredoxin
nitrite reductase as a tetra nuclear center and preliminary EPR studies of mechanisms.
J BioI Chem 254:1268-1272
Lee KY, Pan SS, Erikson R, Nason A (1974) Involvement of molybdenum and iron
in the in vitro assembly of assimilatory nitrate reductase utilizing Neurospora mutant
nit-I. J BioI Chem 249:3941-3952
Lee RB (1979) The release of nitrite from barley roots in response to metabolic inhibitors
uncoupling agents and anoxia. J Exp Bot 30: 119-133 .
Losada M, GuerreroMG (1979) The photosynthetic reduction of nitrate and its regula-
tion. In: Barker J (ed) Photosynthesis in relation to model systems. ElsevierfNorth-
Holland Biomedical Press, Amsterdam
Losada M, Guerrero MG, Vega JM (1981) The assimilatory reduction of nitrate. In:
Bothe H, Trebst A (eds) Biology of inorganic nitrogen and sulfur. Springer, Berlin
Heidelberg New York
Lycklama JC (1963) The absorption of ammonium and nitrate by perennial ryegrass.
Acta Bot Neerl12:361-423
Magalhaes AC, Neyra CA, Hageman RH (1974) Nitrite assimilation and amino nitrogen
synthesis in isolated spinach chloroplasts. Plant PhysioI58:411-415
Mann AF, Hucklesby DP, Hewitt EJ (1978) Sources of reducing power for nitrate reduc-
tion in spinach leaves. Planta 140:261-263
Mendel RR, Muller AJ (1978) Reconstitution of NADH-nitrate reductase in vitro from
nitrate reductase deficient Nicotiana tabacum mutants. Mol Gen Genet 161 :77-80
Miflin BJ (1967) Distribution of nitrate and nitrite reductase in barley. Nature
214: 1133-1134
Miflin BJ (1974) Nitrite reduction in leaves; studies on isolated chloroplasts. Planta
116: 187-192
Miflin BJ, Lea PJ (1976) The pathway of nitrogen assimilation in plants. Phytochemistry
15:873-885
Minotti PL, Williams DC, Jackson WA (1968) Nitrate uptake and reduction as affected
by calcium and potassium. Soil Sci Am Proc 32:692-698
11.4 Uptake and Reduction of Nitrate: Bacteria and Higher Plants 373

Minotti PL, Williams DC, Jackson WA (1969a) The influence of ammonium on nitrate
reduction in wheat seedlings. Planta 86: 267-271
Minotti PL, Williams DC, Jackson W A (1969 b) Nitrate uptake by wheat as influenced
by ammonium and other cations. Crop Sci 9: 9-14
Mohanty B (1979) Effects of ammonium on the activities of nitrate reductase and gluta-
mine synthetase in mung bean hypocotyls. Plant Physiol Suppl 63: 253
Mohanty B, Fletcher JS (1976) Ammonium influence on the growth and nitrate reductase
activity of Paul's scarlet rose suspension cultures. Plant Physiol 58: 152-155
Mohanty B, Fletcher JS (1980) Ammonium influence on nitrogen-assimilating enzymes
and protein accumulation in suspension cultures of Paul's scarlet rose. Physiol. Plant
48:453-459
Mulder EO, Boxma R, van Veen WL (1959) The effect of molybdenum and nitrogen
deficiencies on nitrate reduction in plant tissues. Plant Soil 10: 335-355
Murphy MJ, Siegel LM, Tove SR, Kamin H (1974) Siroheme: A new prosthetic group
participating in six electron reduction reactions catalyzed by both sulfite and nitrite
reductases. Proc Natl Acad Sci USA 71: 612-616
Neyra CA, Hageman RH (1974) Dependence of nitrite reduction on electron transport
in chloroplasts. Plant Physiol 54:480-483
Neyra CA, Hageman RH (1975) Nitrate uptake and induction of nitrate reductase in
excised corn roots. Plant Physiol 56: 692-695
Neyra CA, Hageman RH (1976) Relationship between carbon dioxide, malate and nitrate
accumulation and reduction in corn (Zea mays L.) seedlings. Plant Physiol58: 726-730
Notton BA, Hewitt EJ (1979) Structure and properties of higher plant nitrate reductase
especially Spinacia oleracea L. In: Hewitt EJ, Cutting CV (eds) Nitrogen assimilation
of plants. Academic Press, London New York
Notton BA, OrafL, Hewitt EJ, Pover RC (1974) The role of molybdenum in the synthesis
of nitrate reductase in cauliflower (Brassica oleracea L. var Botrytis) and spinach
(Spinacea oleracea L.). Biochim Biophys Acta 364:45-58
Notton BA, Fido RJ, Hewitt EJ (1977) The presence of functional haem in a higher
plant nitrate reductase. Plant Sci Lett 8: 165-170
Notton BA, Fido RJ, Watson EF, Hewitt EJ (1979) Presence of haem in the tungsten
analogue of nitrate reductase and its relationship to dehydrogenase function. Plant
Sci Lett 14:85-90
Oaks A, Aslam M, Boesel I (1977) Ammonium and amino acids as regulators of nitrate
reductase in corn roots. Plant Physiol 59: 391-394
Oaks A, Stulen I, Boesel IL (1979) Influence of amino acids and ammonium on nitrate
reduction in corn seedlings. Can J Bot 57: 1824-1828
Oostindier-Braaksma FJ, Feenstra WJ (1973) Isolation and characterization of chlorate
resistant mutants of Arabidopsis thaliana. Mutat Res 19: 175-185
Orebamjo TO, Stewart OR (1974) Some characteristics of nitrate reductase induction
in Lemna minor L. Planta 117: 1-10
Orebamjo TO, Stewart OR (1975a) Ammonium repression of nitrate reductase formation
in Lemna minor L. Planta 122:27-36
Orebamjo TO, Stewart OR (1975b) Ammonium inactivation of nitrate reductase in
Lemna minor L. Planta 122:37-44
Pate JS (1973) Uptake, assimilation and transport of nitrogen compounds in plants.
Soil BioI Biochem 5: 109-119
Payne WJ (1973) Reduction of nitrogenous oxides by microorganisms. Bacteriol Rev
37:409-452
Payne WJ, Riley PS, Cox CD (1971) Separate nitrite, nitric oxide, and nitrou~ oxide-
reducing fractions from Pseudomonas perfectomarinus. J Bacteriol106: 356-361
Pichinoty F (1970) Les nitrate-reductases bacteriennes IV. Regulation de la biosynthese
et de l'activite de l'enzyme B. Arch Mikrobiol71 :116-122
Pichinoty F, Piechaud M (1968) Detection and characterization of bacterial nitrate reduc-
tases A and B. Ann Inst Pasteur 114:77-98
Prasad R, Rajale OB, Lakkhdive BA (1971) Nitrification retarders. Adv Agron
23:337-387
374 L. BEEVERS and R.H. HAGEMAN:

Radin JW (1975) Differential regulation of nitrate reductase induction in roots and shoots
of cotton plants. Plant Physiol 55: 178-182
Rao KP, Rains DW (1976a) Nitrate absorption by barley. I. Kinetics and energetics.
Plant Physiol 57: 55-58
Rao KP, Rains DW (1976b) Nitrate absorption by barley II. Influence of nitrate reductase
activity. Plant Physiol 57: 59-62
Raven JA, Smith FA (1976) Nitrogen assimilation and transport in vascular plants in
relation to intracellular pH regulation. New Phytol 76:415-431
Redinbaugh MG, Campbell WH (1981) Purification and characterization of NAD(P)H:
nitrate reductase and NADH: nitrate reductase from corn roots. Plant Physiol
68:115-120
Riet Vant J, Wientjes FB, van Doorn J, Planta RJ (1979) Purification and characterization
of the respiratory nitrate reductase of Bacillus licheniformis. Biochim Biophys Acta
576:347-360
Rogers HH, Campbell JC, Yolk RJ (1979) 15N02 uptake and incorporation by Phaseolus
vulgaris L. Science 206: 333--335
Rucklidge G, Notton BA, Hewitt EJ (1976) Reconstitution in vitro of nitrate reduc-
tase from apoprotein of molybdenum-deficient spinach. Biochem Soc Trans 4: 77-
80
Rufty TW, Yolk RJ, McClure PR, Israel DW, Raper CD (1982) Relative content of
NO; and reduced N in xylem exudate as an indicator of root reduction of concur"
rently absorbed 15NO - 1. Plant Physiol 69: 166-170
Sawhney SK, Naik MS (1972) Role of light in the synthesis of nitrate reductase and
nitrite reductase in rice seedlings. Biochem J 130:475-485
Sawhney SK, Naik MS, Nicholas DJD (1978) Regulation of NADH supply for nitrate
reduction in green plants via photosynthesis and mitochondrial respiration. Biochem
Biophys Res Commun 81:1209-1217
Schrader LE, Ritenour GL, Eilrich GL, Hageman RH (1968) Some characteristics of
nitrate reductase in higher plants. Plant Physiol 43: 930-940
Schrader LE, Domska D, Jung PE, Peterson LA (1972) Uptake and assimilation of
ammonium-N and nitrate-N and their influence on the growth of corn (Zea mays
L.). Agron J 64:690-695
Shaner DL, Boyer JS (1976) Nitrate reductase activity in maize (Zea mays L.) leaves.
I. Regulation by nitrate flux. Plant Physiol 58 :499-504
Shen TC (1969) The induction of nitrate reductase and the preferential assimilation
of ammonium in germinating rice seedlings. Plant Physiol44: 1650-1655
Shen TC (1972) Nitrate reductase of rice seedlings and its induction by organic-nitro
compounds. Plant Physiol 49: 546-549
Shen TC, Funkhouser EA, Guerrero MG (1976) NADH- and NAD(P)H-nitrate reduc-
tases in rice seedlings. Plant Physiol 58: 292-294
Sims AP, Folkes AP (1964) A kinetic study of the assimilation of 15N ammonia and
the synthesis of amino acids in an exponentially growing culture of Candida utilis.
Proc R Soc London Ser B 159:479-502
Sluiters-Scholten CMT (1975) Photosynthesis and the induction of nitrate reductase and
nitrite reductase in bean leaves. Planta 123: 175-184
Smith FW, Thompson JF (1971) Regulation of nitrate reductase in excised barley roots.
Plant PhysioI48:219-223
Solomonson LP, Lorimer GH, Hall RL, Borchers R, Bailey JL (1975) Reduced nicotin-
amide adenine dinuc1eotide~nitrate reductase of Chlorella vulgaris. Purification, pros-
thetic groups and molecular properties. J BioI Chem 250:4120-4127 ,
Sosa FM, Ortega T, Bara JL (1978) Mutants from Chlamydomonas reinhardii affected
in their nitrate assimilation capability. Plant Sci Lett 11: 51-58
Stewart GR, Orebamjo TO (1979) Some unusual characteristics of nitrate reduction
in Erythrina senegalensis D.C. New Phytol 83:311-320
Stewart GR, Rhodes D (1978) Control of enzyme levels in the regulation of nitrogen
assimilation. In: Smith H (ed) Regulation of enzyme synthesis and activity. Academic
Press, London New York
Stewart GR, Lee JA, Orebamjo TO, Havill DC (1974) Ecological aspects of nitrogen
11.4 Uptake and Reduction of Nitrate: Bacteria and Higher Plants 375

metabolism. In: Bieleski RL, Ferguson AR, Cresswell MM (eds) Mechanisms of

regulation of plant growth. R Soc N Z Bull 12:41-47
Street HE, Sheat DEG (1958) The absorption and availability of nitrate and ammonia.
In: Ruhland W (ed) Encyclopedia of plant physiology, vol VIII. Springer, Berlin
Gottingen Heidelberg
Stouthamer AJ (1976) Biochemistry and genetics of nitrate reductase in bacteria. Adv
Microbiol PhysioI14:315-375
Stouthamer AH (1977) In: Haddock BA, Hamilton EA (eds) Microbial energetics. Cam-
bridge Univ Press, Cambridge
Tempest DW, Meers JL, Brown CM (1973) In: Prusiner S, Stadtman ER (eds) The
enzymes of glutamine metabolism. Academic Press, London New York
Thauer RK, Jungerman K, Decker K (1977) Energy conservation in chemotrophic anaer-
obic bacteria. Bacteriol Rev 41 : 100--180
Tokarev BI, Shumnyi VK (1977) Clarification of barley mutants with lowered nitrate
reductase activity after treatment of the grain with ethylmethanesulfonate. Genetika
(Moscow) 13:2097-2103
Travis RL, Key JL (1971) Correlation between polyribosome level and the ability to
induce nitrate reductase in dark grown corn seedlings. Plant Physiol 48: 617-620
Vega JM, Kamin H (1977) Spinach nitrite reductase. Purification and properties of a
siroheme-containing iron-sulfur enzyme. J BioI Chern 252: 896-909
Vega JM, Herrera J, Aparicio PJ, Paneque A, Losada M (1971) Role of molybdenum
in nitrate reduction in Chlorella. Plant Physiol 48: 294-299
Vennesland B, Guerrero MG (1980) Reduction of nitrate and nitrite. In: Gibbs M,
Latzko E (eds) Photosynthesis II. Encyclopedia of plant physiology. new ser vol 6.
Springer, Berlin Heidelberg New York .
Vila R, Barcena JA, Llobell A, Paneque A (1977) Characterization of a membrane-bound
nitrate reductase from Azotobacter chroococcum. Biochem Biophys Res Commun
75:682-688
Vincent SP, Bray RC (1978) Electron paramagnetic resonance studies on nitrate reductase
from Escherichia coli K12. Biochem J 171 :639-647
Wallace W (1974) Purification and properties of a nitrate reductase-inactivating enzyme.
Biochim Biophys Acta 341 :267-276
Wallace W (1975) Effects of a nitrate reductase inactivating enzyme and NAD(P)H
on the nitrate reductase from higher plants and Neurospora. Biochim Biophys Acta
377:239-250
Wallace W, Johnson CB (1978) Nitrate reductase and soluble cytochrome c reductase(s)
in higher plants. Plant Physiol 61: 748-752
Warburg 0, Negelein E (1920) The reduction of nitrate in green cells. Biochem Z
110:66-115
Warncke DE, Barber SA (1973) Ammonium and nitrate uptake by corn (Zea mays
L.) as influenced by nitrogen concentration and NHt INO:; ratio. Agron J 65: 950--953
Warner RL, Hageman RH, Dudley JW, Lambert RJ (1969) Inheritance of nitrate reduc-
tase activity in Zea mays. Proc Natl Acad Sci USA 62:785-792
Warner RL, Lin CJ, Kleinhofs A (1977) Nitrate reductase-deficient mutants in barley.
Nature 269: 406-407
Weissman GS (1964) Effect of ammonium and nitrate nutrition on protein level and
exudate composition. Plant Physiol 39: 947-952
Woo KC, Jukinen M, Canvin DT (1980) Reduction of nitrate via a dicarboxylate shuttle
in a reconstituted system of supernantant and mitochondria from spinach leaves.
Plant PhysioI56:433-436
Wood PM (1978) Periplasmic location of the terminal reductase in nitrite respiration.
FEBS Lett 92:214-218
Wray JL, Filner P (1970) Structural and functional relationships of enzyme activities
induced by nitrate in barley. Biochem J 119:715-725
Zielke HR, Filner P (1971) Synthesis and turnover of nitrate reductase induced by nitrate
in cultured tobacco cells. J BioI Chern 246: 1772-1779
Zumft WG (1976) Anorganische Biochemie des Stickstoffs; die Mechanismen der Stick-
stoffassimilation. Naturwissenschaften 63: 457-464
11.5 Uptake and Reduction of Nitrate:
Algae and Fungi
W.R. ULLRICH

1 .Introduction

Algae have been used for nitrate and nitrite assimilation studies since the begin-
ning of this century (WARBURG and NEGELEIN 1920). In the meantime a large
body of knowledge has been accumulated and requires separate treatment in
this chapter of the Volume, although many processes in nitrate assimilation
are common in algae and higher plants. Some differences are due to the different
evolutionary backgrounds of the groups or to different structural properties
of the enzymes, but some are merely due to the relatively large surface of
contact that algal cells have with the external medium. For reasons of evolution,
the structural characteristics and the enzymes of blue-green algae (cyanobac-
teria) are very different from those of eucaryotes and require special paragraphs.
However, the differences between e.g. green algae or diatoms and higher plants

(N H:) - - 1 - - - - - - , 0 " -

n-1H+-__P -v-1
(n-2J

~~---------I--N03
~~--------I--K+
Tissue

Fig. 1. Scheme of nitrate assimilation and related alkalinization under steady state condi-
tions. Upper part represents cells of microalgae, the whole scheme vacuolated cells. R'
nitrate reductase; R" nitrite reductase; in brackets transport related to ammonium accu-
mulation and release; A counter-transport (antiport); ceo-transport; P H+ extrusion
pump (ATPase). Protein synthesis in cytoplasm omitted. Stoichiometry of alkalinization
in micro algae in steady state: H+ net uptake (or net OH- re1ease)=NO; uptake -
N02" re1ease+NHt release
11.5 Uptake and Reduction of Nitrate: Algae and Fungi 377

seem to be much less clearly defined with respect to nitrate assimilation, and
many of them can be explained with regard to morphological organization.
As a main nutrient, nitrate belongs to the metabolizable anions, but it can
also be stored to a large extent in its original form. The process of nitrate
assimilation comprises several steps: uptake, storage in vacuoles, transport to
other cells or organs, reduction, and assimilation of the reduced nitrogen. In
higher plants this system is normally complete and involves the participation
of the whole plant. In microalgae (e.g. blue-green algae or green algae with
small single cells) little or no storage of nitrate is possible, because the cells
do not contain storage vacuoles, nor is there any transport to other cells. As
a result, nitrate uptake and nitrate reduction are tightly coupled in these cells
and can easily control each other (Fig. 1). An intermediate situation is encoun-
tered in the vacuolate unicellular algae, for example the diatoms, the green
algae Hydrodictyon or Platymonas, and the Characeae, in which translocation
from one giant cell to another still plays only a minor role, but where storage
in the large vacuoles may account for a large proportion of the nitrate uptake
rates under appropriate conditions. In some cases the regulation of nitrate
uptake can be separated from that of nitrate reduction, leading to a rather
complicated pattern of induction/repression and activation/inactivation pro-
cesses. In many cases the direct linkage between uptake and reduction and
the complicated response to external factors has made investigation very diffi-
cult, and has not allowed us to apply the biophysical equations as used for
uptake studies with chloride or potassium ions (WALKER 1976, RAVEN 1976).

2 Nitrate and Nitrite Reduction in Algae

As in higher plants and fungi, nitrate reduction in algae occurs in two steps,
first the reduction of nitrate to nitrite, and then the reduction of nitrite to
ammonia. A more detailed description of the processes and the enzymes involved
was given by VENNESLAND and GUERRERO (1979) and recent literature was re-
viewed by GUERRERO et al. (1981), so the following part of this chapter is re-
stricted to a short survey and to selected literature only.

2.1 Nitrate Reductase of Eucaryotic Algae

The enzyme shows many properties common with that of higher plants (see
Chap. II.4). Whereas NADH is the predominant electron donor in nitrate reduc-
tion by higher plants, for several species of algae NADPH has been reported
to be as good or better for algal nitrate reductases, for example those from
Ankistrodesmus braunii (AHMED and SPILLER 1976, DIEZ et al. 1977), Dunaliella
(LECLAIRE and GRANT 1972), and Chlamydomonas (SOSA and CARDENAS 1977).
According to this specificity, algal nitrate reductase has been coded as EC 1.6.6.1
when NADH-specific, EC 1.6.6.2 when specific for both, and EC 1.6.6.3 when
378 W.R. ULLRICH:

(Electron donors) Fig. 2. Schematic illustration of electron

FMNH transport in nitrate reduction, regardless of
NAD(P)H+H+ subunit structure of nitrate reductase. Posi-
tions of Fe (cytochrome b-557) and Mo
in the chain, as assumed recently (cf. GUER-
RERO et al. 1981)

NAD(W Cyt C H20 NO:Z

Ferricyanide
(Electron acceptors)

Diaphorase Terminal reductase

(part a) (part b)

specific for NADPH. Generally the preference of the enzyme for one or the
other electron donor is only relative. Purification of algal nitrate reductase
was achieved more or less completely following classical procedures (for litera-
ture see HEWITT 1975, VENNESLAND and GUERRERO 1979, BEEVERS and HAGEMAN
1980), but in some cases serious difficulties of instability remained and did
not allow high specific activities. The recent technique of affinity chromatogra-
phy using blue-dextran sepharose or agarose columns has led to much shorter
and easier procedures and to the preparation of homogeneous enzyme proteins
in a few steps (SOLOMONSON 1975, DE LA ROSA et al. 1980).
Algal nitrate reductase is characterized as a protein complex of molecular
weight between 356,000 (Chlorella) and 467,000 (Ankistrodesmus) and is com-
posed of a varying number of subunits (between 2 and 8, DE LA ROSA et al.
1981); it contains flavin, molybdenum and heme iron. In nitrate reductase from
Chlorella the flavin is apparently more tightly bound than in reductase from
other plants and algae. The enzyme activities are normally divided into two
parts: (a), diaphorase activity concerning the transfer of electrons from
NAD(P)H to the flavin group (FAD) of the enzyme, and thence to artificial
electron acceptors such as cytochrome c, ferricyanide or others, without reduc-
ing nitrate; (b), terminal nitrate reductase activity concerning the transfer of
electrons from part (a) or from artificial electron donors such as reduced flavins
or methylviologen to nitrate (Fig. 2). The diaphorase moiety of the enzyme
has a different sensitivity to various inhibitors, activating or inactivating condi-
tions. It contains the flavin. The terminal nitrate reductase moiety contains
the heme iron, in most cases identified as cytochrome b-557, and the molybde-
num, and exhibits the typical sensitivity to metal-binding inhibitors. A physical
separation of the two enzyme moieties has not been achieved as yet.
In algae, as in higher plants, nitrate reductase is an enzyme with a short
half-life of only a few hours, and therefore is subject to a rapid turnover. Hence
there are many reports on its induction, mainly by light and nitrate (see reviews
by HEWITT 1975, LOSADA and GUERRERO 1979, GUERRERO et al. 1981, and also
Dmz et al. 1977) as well as on its repression by ammonia or amino acid nutrition
(MORRIS and SYRETT 1963, and the same reviews). In several species of algae
at least part of the nitrate reductase available in a normal cell seems to be
constitutive, i.e. its formation does not require special induction (VEGA et al.
1971, Dmz et al. 1977, HIPKIN and SYRETT 1977) or the formation of an inactive
precursor protein occurs even in the presence of ammonium (FUNKHOUSER and
11.5 Uptake and Reduction of Nitrate: Algae and Fungi 379

RAMADOSS 1980). In addition to this regulation of amounts by protein synthesis

and gene activity, the actual activity of existing nitrate reductase can be drasti-
cally reduced by inhibitors. Chlorate acts as a competitive inhibitor, competing
with nitrate and forming chlorite, which is more toxic for the cells than nitrite
(TROMBALLA and BRODA 1971, SOLOMONSON and VENNESLAND 1972a, VEGA et al.
1972). Azide, cyanate, and thiocyanate are also competitive with nitrate, while
cyanide and sulfide inhibit by complexing molybdenum or iron. In the presence
of the electron donor NADH, cyanide is very effective both in vivo and in
vitro at inhibiting terminal nitrate reductase activity (VEGA et al. 1972, SOLOMON-
SON and VENNESLAND 1972b). A stoichiometric HCN-nitrate reductase complex
is formed and has been studied for Chiarella vulgaris (LORIMER et al. 1974).
This cyanide complex, however, retains its full diaphorase activity and is stable
under reducing conditions. The enzyme can be reactivated easily by oxidants
such as ferricyanide, and less efficiently by O 2 , nitrate or redox dyes, but O 2
can also contribute to inactivation (CHAPARRO et al. 1979). Nitrate can prevent
the formation of this cyanide-nitrate reductase complex, probably by maintain-
ing the enzyme in an oxidized form. A competitive inhibition of nitrate reduction
by nitrite has been reported for the enzyme from Chiarella vulgaris in vitro,
but the affinity for nitrite was lower than for nitrate (SOLOMONSON and VENNES-
LAND 1972b). The diaphorase moiety is sensitive to mercuribenzoate (pCMB),
an SH -group inhibitor.
Rather complex results have been shown in algae for the in vivo effect
of ammonia on nitrate reduction, despite the fact that ammonia does not inhibit
the isolated enzyme in vitro. Repression of the formation of nitrate reductase
has been postulated for many years (MORRIS and SYRETT 1963), reversible inacti-
vation in vivo only more recently. The inactive enzyme can be isolated from
cells and then its reactivation achieved with ferricyanide. Alternatively, reactiva-
tion occurs very quickly with blue light in the presence of FAD (APARICIO
et al. 1976). There is still some discussion whether the inactive form generated
by ammonia inhibition may possibly result from the formation of the afore-
mentioned cyanide complex or not (CHAPARRO et al. 1979). Since inactivation
in vitro also occurs by reduction in the presence ofNADH and ADP, an uncou-
pling effect by ammonia in vivo was thought to be responsible, but in some
algal strains cyanide was always found after inactivation in vivo. Extensive
studies in various species of algae, recently with the blue-green alga Anacystis
nidulans, showed that cyanide is actually formed in considerable amounts in
algal cells, mainly via amino acid oxidases (PISTORIUS et al. 1979).
An irreversibly inactive form of nitrate reductase can be produced with
tungstate instead of molybdate. An exchange of molybdenum and tungsten
in vitro is not possible, but tungsten is incorporated in vivo, e.g. in Chiarella
enzyme when added to the nutrient medium instead of molybdenum (VEGA
et al. 1971). Vanadium is reported to produce a similar inactive enzyme, but
its action is not quite clear yet (HEWITT 1975).
The total activity of nitrate reductase found in algal cells normally considera-
bly exceeds the activity necessary to account for the rates of nitrate assimilation
in vivo. Km values for nitrate shown by nitrate reductase preparations from
green algae or diatoms normally resemble those found for higher plants (100
to 200 IlM).
380 W.R. ULLRICH:

2.2 Nitrate Reductase in Blue-Green Algae

It differs from that in eucaryotic photosynthetic organisms in various respects.

It can be isolated very efficiently by affinity chromatography (MANZANO et al.
1978). Nitrate reductase of Anabaena or Anacystis is particle-bound, but can
be solubilized by some treatments during the isolation procedure. It cannot
accept electrons from reduced pyridine nucleotides, but only from single-electron
donors (HATTORI and MYERS 1967, MANZANO et al. 1976). According to recon-
stitution experiments with chlorophyll-containing particles, ferredoxin in the
photosynthetic electron transport chain is probably the direct source of electrons
for nitrate reduction in these cells (MANZANO et al. 1976). In contrast to the
enzyme from other algae and higher plants, nitrate reductase from blue-green
algae is a single polypeptide chain with a molecular weight of about 75,000.
It is inactivated by ammonia, even in vitro when bound to particles, and by
reduction in the light, and it is reactivated by O 2 (ORTEGA et al. 1977).

2.3 Nitrite Reductase in Algae

Nitrite reductase has been purified to a different extent from various algae
as from Chlorella (ZUMFT 1972), Anacystis (MANZANO et al. 1976), and even
from the red alga Porphyra (Ho et al. 1976). That from Chlorella has a molecular
weight of 63,000, as is the case for several higher plants. Iron is present in
the enzyme molecule in a siroheme group and most probably also (as shown
for higher plant nitrite reductase) in an iron-sulfur center. The reductant for
nitrite reduction is reduced ferredoxin, which is also a very effective stabilizing
agent on sepharose columns during the purification procedure. Reduced ferre-
doxin can be replaced as the electron donor by flavodoxin, which is found
also in vivo (ZUMFT 1972, BOTHE 1977), and by several artificial one-electron
donors such as reduced viologens. Although hydroxylamine does not appear
as a free intermediate in the reduction from nitrite to ammonia, it can be reduced
to ammonia by nitrite reductase at a low rate (CZYGAN 1965, ZUMFT 1972).
Algal nitrite reductase is a soluble enzyme except in blue-green algae (e.g. Ana-
cystis), where a special treatment is required to dissolve it from the particles.
In most algae, nitrite reductase has a higher affinity to its substrate than nitrate
reductase, the reported Km values lying between 20 and 200 (up to 450) J.lM.
Nitrite reductase of algae is sensitive to a series of inhibitors among which
cyanide, carbon monoxide, and sulfhydryl group blockers as pCMB or mersaly-
late are the most important. Nitrite can protect the enzyme against CO inhibi-
tion. The iron-sulfur center is completely reduced only when CO binds to siro-
heme and prevents its re-oxidation. In vivo in several algae, an inhibition of
nitrite and nitrate assimilation by uncouplers as DNP or CCCP was observed
(HOFMANN 1972, ULLRICH 1974, LARSSON and ANDERSSON 1982), but not with
the isolated enzymes. For more biochemical information see the article by VEN-
NESLAND and GUERRERO (1979) and the preceding (Chap. 11.4, this Vol.) by
BEEVERS and HAGEMAN.
11.5 Uptake and Reduction of Nitrate: Algae and Fungi 381

2.4 Location of Nitrate and Nitrite Reduction in Algal Cells

In higher plants, according to DALLING et al. (1972), nitrate reductase seems

to be generally located in the cytoplasmic compartment (cytosol), while nitrite
reductase is in chloroplasts or other plastids. Similar attempts have been made
to localize these enzymes in the algae, but the results were less clear, perhaps
due to the much greater difficulties encountered in isolating intact chloroplasts
from most algal cells. The limited evidence suggests that nitrate reductase occurs
in the cytosol, but the recovery of nitrate reductase in the soluble fraction
after homogenization of the cells is, of course, not a final proof that the enzyme
is also soluble in vivo. A hypothesis proposed by BuTZ and JACKSON (1977)
suggesting that the nitrate uptake carrier located in the plasmalemma also func-
tions as the nitrate reductase is not in accordance with many results from algae,
where independent activity and regulation of nitrate uptake and reduction were
shown. There is some evidence supporting a chloroplast localization for nitrite
reductase in contrast to nitrate reductase, especially the quenching of photo-
system II chlorophyll fluorescence by nitrite, not by nitrate, in Ankistrodesmus
under anaerobic conditions (KESSLER and ZUMFT 1973). This compartmentation
has some implications in intracellular transport. Both processes require reducing
power, but in a different form. Photosynthetic reducing power can be used
directly for nitrite reduction with reduced ferredoxin. Under non-photosynthetic
conditions, respiratory reducing power available as NADH must pass the chlo-
roplast envelope by special transport shuttles (HEBER 1974) and then be trans-
formed into reduced ferredoxin. Nitrite also has to pass through the chloroplast
envelope from the cytosol, and nitrite reduction causes alkalinization in the
chloroplast. In algae the rates of nitrate and especially of nitrite reduction are
normally much lower in the dark than in the light, which is consistent with
the more complicated transport steps needed to provide reducing power at
the site of nitrite reduction and also with the much lower rates of reducing
power production in the dark. According to studies with intact chloroplasts
(HEBER 1974) the shuttle systems for transport of intermediates have very high
affinities to their substrates, so photosynthetic electron transport will not easily
become limiting, even for nitrate reduction in the cytoplasmic compartment.

2.5 Stoichiometry Between Nitrate Reduction and O 2 Exchange

The consumption of reducing power for nitrate (and nitrite) reduction leads
to changes in the 02/C02 ratios in respiration and photosynthesis. In some
cases respiratory O 2 consumption and especially CO 2 evolution are increased,
indicating a change in the metabolic rates, since nitrate reduction will,compete
with O2 for NADH. Also in photosynthesis the results are variable and differ
from species to species. GRANT and TURNER (1969) found a regular stoichiometry
between extra-0 2 evolution and nitrate assimilation only in Chlorella pyrenoidosa,
not in five other species from various groups of algae. Similarly MORRIS and
AmmD (1969) reported the expected ratio of 1.5 with nitrite only for Chlorella
382 W.R. ULLRICH:

and otherwise rather variable ratios, most of them exceeding the expected values
of 02/NO; =2, also in Ankistrodesmus. For this alga KESSLER (1957) reported
a ratio of 1.5 with nitrite, and ULLRICH and EISELE (1977) found relatively
stable ratios with nitrate which were lower (about 1.65) in the absence of CO 2,
but higher (about 2.1) at CO 2 saturation. The varying ratios and the lack of
extra-0 2 evolution in many algal species shows the involvement of various
metabolic processes and limitations when nitrate is reduced.

3 Nitrate Uptake in Algae

3.1 <;eneralltemnarks
Unicellular micro algae and giant algal cells differ to some degree in their proper-
ties of nitrate uptake, mainly due to the fact that the microalgae are devoid
of storage vacuoles, the existing microvacuoles being specialized for the storage
of polyphosphates (volutin granula) or lipids. This makes it extremely difficult
to measure reliable pool sizes of nitrate in such cells. Many algal species contain
vacuoles, but the storage still plays a minor role. In most micro algae, uptake
and reduction of nitrate are closely coupled processes which become limiting
to each other and which can be separated only under special conditions. Never-
theless, nitrate reduction can sometimes be shown to be not limiting, when
the production of reducing power and the activity of nitrate reductase are consid-
erably higher than the observed rates of nitrate assimilation.
On the basis of our present knowledge nitrate uptake has some characteristic
properties that are distinctly different from those of isolated nitrate reductase:
a far lower Km for nitrate (0.5 to 50 J.1M instead of 100 to 200 J.1M); different
sensitivity to various inhibitors such as tungstate, ammonium or methylammon-
ium and other uncouplers; and an immediate response to various conditions,
whereas nitrate reductase inactivation or reactivation may take many minutes
or even several hours.
As to the use of Michaelis-Menten kinetics in uptake studies, a few general
remarks may be required. Nitrate uptake as a function of concentration follows
a saturation curve in almost all organisms. This is regarded as characteristic
for uptake via a "carrier" or "permease". In many instances the rate limitation
for uptake could just as well be due to a limited driving force on H+ -anion
co-transport, presumably through the electrochemical proton gradient (A.uH+,
see Sect. 3.10). At higher nitrate concentrations the saturation curve may remain
unchanged or become linear with nitrate concentration (e.g. SERRA et al. 1978a,
for Skeletonema); but in some cases high nitrate concentration leads to a serious
inhibition of nitrate assimilation, probably due to control by nitrate uptake,
especially in the absence of carbon sources.
Another problem in using Michaelis-Menten equations and constants arises
from the frequent observation that once a rate of uptake has established at
a certain initial concentration, this rate remains constant until almost all nitrate
is consumed, irrespective of the Km calculated from the initial rates obtained
11.5 Uptake and Reduction of Nitrate: Algae and Fungi 383

over a wide range of concentrations (CALERO et al. 1980, see also figures in
TISCHNER and LORENZEN 1979). This requires further investigation and indicates
that the uptake system is complex and may involve a binding protein. The
same phenomena have been studied for various ions in erythrocytes. There
is sometimes interference of uptake with nitrate release from storage vacuoles
as in Platymonas (RICKETTS and EDGE 1977, ULLRICH and GROGER unpub-
lished).

3.2 Substrate AtTmity

The nitrate affinity of the uptake system differs from species to species, but
it varies also with the external conditions and with cell age or pretreatment.
FALKOWSKI (1977) presents a model for a bisubstrate kinetics (nitrate and light)
as an example to show how various external factors may interfere with nitrate
uptake and complicate a kinetic treatment. Thus under optimal conditions Km
may appear much higher than under restricted conditions in the same algal
culture. As mentioned before, in microalgae and not only there, nitrate uptake
may be directly limited by its consumption within the cells. These problems
have led many authors to use the term Ks instead of Km.
There are many reports of Ks values in algae, mainly measured for ecological
purposes, and usually in oceanic species. In general there seems to exist a weak
inverse correlation between cell size and Ks (EpPLEY et al. 1969). As another
general tendency, the freshwater algae (mostly from meso- or eutrophic habitats)
show lower affinity (higher Ks) to nitrate than those algae which occur in the
littoral zones of the sea or in the free ocean. We can speculate that this is
related to the much higher nitrate concentrations found in most freshwaters
than in the open sea, where nitrate concentrations are so low that the plankton
algae mainly feed on other nitrogen sources (EpPLEY et al. 1969). There is,
however, little correlation with taxonomy. According to HATTORI (1962) Ana-
baena cylindrica, a freshwater blue-green alga, shows a rather high apparent
Ks of 70 ~M. A number of green algae species from very different ecological
systems have been studied. The highly saline Dunaliella tertiolecta shows a medi-
um Ks value of 1.4 JlM (EpPLEY et al. 1969), Chlorella sorokiniana when adapted
to high nutrient levels a Ks of 4.3 ~M (TISCHNER and LORENZEN 1979). In
some neritic and oceanic species Ks can be as low as 0.2 JlM, irrespective of
whether they belong to Dinophyceae, centric or pennate diatoms, Coccolithinae
or Chrysophyceae (EpPLEY et al. 1969, FALKOWSKI 1975, SERRA et al. 1978 a).
As mentioned before, some algae show considerable changes in nitrate uptake
affinity under various external conditions; for example, the marine diatom Ske-
letonema costatum (EpPLEY et al. 1969), and the seaweeds Codium fragile
(HANISAK and HARLIN 1978) and Laminaria longicruris (HARLIN and CRAIGm
1978) show a decrease of Ks towards lower temperature, or an increase of
Ks with increasing light intensity (GRANT and TURNER 1969, FALKOWSKI 1977).
In synchronously grown Ankistrodesmus braunii (Chlorophyceae), the Ks for
nitrate seems to be extremely variable. In the absence of CO 2 , the apparent
Ks (the concentration for half-saturation of uptake) is about 500 ~M (ULLRICH
384 W.R. ULLRICH:

1974, EISELE 1976), while at CO 2 saturation in the light, values below 10 IJ.M
were found (EISELE 1976, ULLRICH unpublished data).

3.3 Light Dependence

For many years and for many species of algae, light stimulation of nitrate
uptake has been reported (KESSLER 1964); however, low nitrate assimilation
rates in the dark or at low light intensities do not necessarily reflect a limitation
by the uptake system but may be due to the lower rates of intracellular consump-
tion. Correspondingly, light stimulation is most pronounced in micro algae with-
out storage vacuoles. Thus in Ankistrodesmus or Chlorella light-dark ratios of
20: 1 to SO: 1 can easily be measured, especially after a short nitrate-starvation
period (ULLRICH-EBERIUS 1973, CALERO et al. 1980, Fig. 3). By contrast, phos-
phate uptake is much less stimulated by light in Ankistrodesmus, probably be-
cause phosphate is metabolized without reducing power (ULLRICH-EBERIUS
1973). GRANT and TURNER (1969) found a light stimulation of nitrate uptake
between 2- and 1S-fold in six algal species of different taxonomic groups, but
they were not grown in synchronous cultures. With Anabaena cylindrica
HATTORI (1962) showed a very strong initial light stimulation in nitrate and
nitrite uptake, but the rates soon decreased. In Chlamydomonas reinhardii light
stimulation was nine fold in non-starved cells, but only about three fold in
nitrogen-starved cells which generally showed higher uptake rates (THACKER
and SYRETT 1972).
Like the Ks values, light saturation rates (V max) of nitrate uptake depend
on various factors, for example on adaptation to "sun" or "shade" conditions

NO; uptake

il;Oi
white 400
30 }J mol mg-1 ChI!
dark
I -0-0
consu,,:ed)

0_0_0--_0 / 50

0
A red 240 / Q96 0

20
~~/1'2·'
o·
1
"mol mg- chi h-1
white /
700 0

• '240(100%) 40

/0 0
white 400/

10 blue 186 0

;0 '26.0 ~ 01'2.0 (50%) Fig. 3. Effect of white and monochro-

30

o /0 f1 mol mg-1 chi h-1

matic light on nitrate uptake in Chlor-
ella fusca with immediate response to

I/O
blue.18y'
light changes. Light intensities in
W m - 2 ; absolute and relative rates (%)
20 8 20.8 (87%) 20 indicated in italics. A comparison be-
tween saturating white light, red light
and dark; B comparison between satur-
blue 190 /0
ating blue light, in the absence and pres-
/"
10 ~/./0 n2 (47%)
10
ence of complementary weak red light,
and saturating white light. (CALERO
et al. 1980)
o 30 60 90 120 150 min
11.5 Uptake and Reduction of Nitrate: Algae and Fungi 385

in phytoplankton species (BATES 1976). Light saturation may be reached earlier

in the absence of CO 2 than in its presence. Use of monochromatic light revealed
a complicated light dependence in Chlorella fusca (CALERO et al. 1980, see
Fig. 3). In the presence of CO 2 the rates of nitrate uptake in saturating blue
or red light were only about one half of those in saturating white light, but
an additional low intensity of the complementary light led to rates approaching
those in white light. Nitrate uptake rates responded immediately upon changes
from one light quality to the other or to dark. Nitrite or ammonia uptake
and photosynthetic O 2 evolution were unaffected by monochromatic light at
saturation. Also nitrate reductase activity remained unchanged and in excess
to that necessary for the in vivo rates of nitrate assimilation. At low light
intensities or in the presence of DCMU only blue light stimulated nitrate uptake,
suggesting an effect on respiratory metabolism with its energy and reducing
power supply (VOSKRESENSKAYA 1972). Studies with nitrate reductase in vitro
(ROLDAN et al. 1978) and the uptake experiments in vivo (CALERO et al. 1980),
which suggested that the most effective wavelength was 457 nm, imply that
flavin could be involved in the stimulation of the respiration-supplied rates
by blue light.
Light stimulation of nitrate uptake can also be accompanied by or due
to an increase in nitrate reductase activity, as has been shown for Chlorella
(TISCHNER and LORENZEN 1979), but such changes are not immediate. They
were found after the onset of the light period in synchronous cultures.

3.4 pH-Dependence

Most of the few data available in the literature show or suggest a pH optimum
for nitrate uptake in the neutral or slightly alkaline range. It is found at pH 7.5
to 8.5 for many species in the light, and at pH 6 in the dark in Ankistrodesmus
(ULLRICH-EBERIUS 1973), but RIGANO (1978) for good reasons grows his acido-
philic unicellular red alga (Cyanidium caldarium) at pH 1.9 with nitrate as sole
nitrogen source. In synchronously grown Ankistrodesmus braunii, pH depen-
dence with the optimum between 7.8 and 9.0 was similar to that of many
other species when measured in the absence of CO 2 , but this pH dependence
was completely abolished by CO 2 or glucose as a carbon source leading to
pH independence between pH 5 and 9 (EISELE and ULLRICH 1975, 1977,
Fig. 4 A). It is likely that similar effects can be found also in other species,
but most experiments were carried out at the low CO 2 concentration of the
air. The varying pH effects may reflect the intracellular pH control dependent
on energy supply (SMITH and RAVEN 1979), but the internal pH seems to be
very stable in some algae (LANE and BURRIS 1981, LARSSON and ANDERSSON
1982). '

3.5 Dependence on Carbon Sources

There is some interdependence between the influence of pH and that of carbon

sources in the light. DAVIS (1953) found considerable stimulation by glucose
386 W.R. ULLRICH:

. . . . . .::::::::::~SO,
!.:\~'A
·•__
~ +---
-.;----::::.~tOI
(3°/oC~· ~
(air)

KCI (3"1oC02) -~ / +
NaCI • Sucrose (air)

---0
o
\o

6 7 .8pH900.01
\
o
-0_0 (0.5mM KN03. pH 8)

0.05
KCI (air)
N~a_CI-:-:-_~=-,
0.10 M 0.15

Fig.4A, B. Effect of carbon sources on pH dependence (A) and on anion sensitivity

(B) of nitrate uptake in Ankistrodesmus braunii. Relative uptake rates, in A related to
pH 8, in B to 0.5 mM KN0 3 in dilute medium, both in the light. (Data for A from
EISELE and ULLRICH 1977, for B from ULLRICH and EISELE 1977)

of nitrate uptake in Chlorella at pH 6.5 in the light. MORRIS and AHMED (1969)
confirmed this with Ankistrodesmus, but not with their strain of Chlorella. In
Ankistrodesmus braunii in the light at pH5 to 6, CO 2 increases the rates of
nitrate uptake up to tenfold or more, at pH 8 only 1.5-.to 3-fold (EISELE and
ULLRICH 1977). Glucose can replace CO 2 in Ankistrodesmus in that it abolishes
pH dependence, but uptake is not enhanced to the same degree. In the dark,
nitrate uptake rates are strongly dependent on respiration rates so that under
anaerobic conditions they may approach zero after a short time (GRANT and
TURNER 1969). CO 2 or glucose can also change the sensitivity of nitrate uptake
rates to nitrate and chloride at higher concentrations (ULLRICH and EISELE
1977, Fig. 4B) and considerably lower the Ks in Ankistrodesmus (cf. Sect. 3.2).
On the other hand, carbon sources increase the regulatory inhibition of nitrate
uptake by ammonium (EISELE 1976, SYRETT and LEFTLEY 1976). As shown by
GRANT and TURNER (1969) the stimulatory effect on nitrate uptake of glucose
is restricted to rather few species, probably to those possessing a specific glucose
uptake system. For Chlorellafusca CALERO et al. (1980) report a five- to tenfold
stimulation of nitrate uptake by glucose in the dark. In Chlamydomonas, acetate
instead of glucose enhances nitrate uptake (THACKER and SYRETT 1972). While
the stimulation by CO 2 in the light is obviously due to the manyfold increase
of all synthetic rates by photosynthesis, organic carbon sources, in the light
or in the dark, may contribute to the supply of reducing power, ATP, and
carbon skeletons for amino acid formation. There is no corresponding increase
in phosphate uptake by CO 2 in the light in Ankistrodesmus (ULLRICH 1971,
1972), possibly because phosphate is not reduced after it has been taken up.

3.6 Inhibition by Anions

Chlorate inhibits nitrate uptake in Chlorella fusca in a competitive way (TROM-

BALLA and BRODA 1971), but an indirect effect by inhibition of nitrate reduction
cannot be excluded. Nitrite may, however, act as a competitive inhibitor of
11.5 Uptake and Reduction of Nitrate: Algae and Fungi 387

nitrate uptake rather than of nitrate reduction, for which higher concentrations
of both nitrate and nitrite may be necessary (EpPLEY and ROGERS 1970, HANISAK
and HARLIN 1978). A peculiarity in Ankistrodesmus braunii is the sensitivity
of nitrate assimilation to nitrate at high concentrations in the simultaneous
absence of a carbon source. Then nitrate at 30 to 100 mM strongly inhibits
nitrate assimilation (ULLRICH 1974). Chloride has the same effect, phosphate
less, sulfate is only slightly inhibitory, and mere osmotics are not at all inhibitory
(ULLRICH and EISELE 1977, Fig. 4 B). Nitrate reductase in vitro shows almost
no response to these salts, and there is no difference between enzymes of fresh-
water organisms and those of high salt tolerance (HElMER 1973). Thus the re-
sponse seems to be due to the uptake system itself.

3.7 Inhibition by Ammonia and Amino Compounds

Ammonia is the main alternative nitrogen source for most plants, and it is
taken up and assimilated by many algae at similar or even higher rates than
nitrate. In many algae, ammonia inhibits or completely suppresses nitrate assimi-
lation. The interpretation is given on three different levels: (a), repression of
nitrate reductase on the level of gene action, requiring several hours even in
fast-growing algae; (b), reversible inactivation of nitrate reductase in vivo but
not in vitro (except in particles from blue-green algae), requiring many minutes
up to a few hours; (c), inhibition of nitrate uptake, taking place within a few
minutes or even less (PISTORIUS et al. 1976, TISCHNER and LORENZEN 1979,
ULLRICH unpublished results). The mechanism of this proposed uptake inhibi-
tion is unknown as yet, but a rapid membrane depolarization could be involved,
as shown for Neurospora (SLAYMAN 1977). In Ankistrodesmus the inhibition
occurs only in the additional presence of a carbon source. Thus in the absence
of CO 2 or glucose, this alga can take up nitrate in the light and reduce it
to nitrite and ammonia, both of which then accumulate in the medium; the
process is only slightly inhibited by addition of external ammonium (EISELE
1976, SYRETT and LEFTLEY 1976). The ammonium inhibition of nitrate uptake
in the presence of CO 2 is immediately relieved, once most of the ammonium
in the medium has been consumed (EISELE 1976, TISCHNER and LORENZEN 1979).
In the presence of glucose instead of CO 2 , the ammonium inhibition is much
less pronounced but still visible. Ammonium was less inhibitory in Codium
(HANISAK and HARLIN 1978). A very clear example of inhibition at the uptake
site is shown for the diatom Phaeodactylum which accumulates nitrate in the
absence of ammonium (CRESSWELL and SYRETT 1979). The special ammonium
effect on nitrate uptake must he attributed to some amino acid as the product
of ammonium assimilation rather than to ammonium itself (SYRETT and LEFTLEY
1976), or to strong and continuing membrane depolarization (see Sect. 3,10).

3.8 Effect of Metabolic Inhibitors and Uncouplers

Most inhibitors effective in reducing nitrate uptake are those which affect the
energy metabolism of the cells. They prevent either nitrate reduction by blocking
388 W.R. ULLRICH:

the production of reducing power or else the general energy supply for all
uptake processes. Another site of action could be the ATP-dependent proton
pump at the plasmalemma. Cyanide, azide, sulfide, and pCMB are also known
to be direct inhibitors of nitrate reductase (see Sect. 2 and VENNESLAND and
GUERRERO 1979). The inhibition of nitrate and nitrite assimilation by uncouplers
(HOFMANN 1972, ULLRICH 1974), which cannot be explained by sensitivity of
the enzymes, could reflect a break-down of transport and of the ability of
the cells to maintain pH levels (SMITH and RAVEN 1979) but in Scenedesmus
it was independent of the stable internal pH (LARSSON and ANDERSSON 1982).
Tungstate and vanadate apparently do not block nitrate uptake completely
(SERRA et al. 1978 b), nor do they inhibit or inactivate nitrate reductase in a
direct way (see Sect. 2, and Chap. IIA, this Vol.). More recently, nitrobenzalde-
hyde and derivatives have been found to specifically compete with nitrate at
its uptake site (TISCHNER and LoRENZEN 1981).

3.9 Stoichiometry Between the Uptake of Nitrate and that of Other Ions

In all plants nutrition with nitrate or nitrite produces an alkalinization of the

surrounding soil or water medium, whereas- with ammonium the environment
is acidified. In higher plants and in algae with nitrate storage capacity, stoichio-
metric relations of this alkalinization are not constant due to simultaneous
transport of cations or other anions for storage in the vacuoles or for transport
to other parts of the plant (cf. Fig. 1). According to the chemical equations
of nitrate and nitrite reduction one proton is consumed per nitrate or nitrite
reduced, as long as the ammonia so formed is used for polypeptide synthesis.
If ammonium or free amino acids are the accumulated products and ammonium
is released to the medium, another proton is consumed to form NHt and
hence the alkalinization is increased. Thus the stoichiometry of 1 OH- for 1
N0 3 , when carbon sources are present to support protein synthesis, increases
to almost 2 OH - for 1 N0 3 when ammonium is accumulated in the medium.
An undisturbed stoichiometry of this type was found in the medium of the
microalga Ankistrodesmus braunii (EISELE and ULLRICH 1975, 1977). There was
no significant concomitant uptake of cations as K +, N a +, or Ca2 +. The pH
regulation in long-term and short-term analysis was calculated in a careful study
for the vacuolate alga Hydrodictyon africanum (RAVEN and DE MICHELIS 1979).

3.10 Transport Mechanism

Such stoichiometric ratios are interesting not only with respect to the biochemis-
try of nitrate reduction and to pH regulation in the respective compartments
of the cells, but also with respect to the transport mechanism. The existence
of an alkaline pH optimum, and the excess production of OH- ions in nitrite
reduction, together may suggest that the uptake of nitrate proceeds via a
counter-transport of OH- against N0 3 . So far in unicellular algae no transient
pH changes or membrane potential changes have been measured with nitrate,
11.5 Uptake and Reduction of Nitrate: Algae and Fungi 389

but corresponding membrane potential changes in frond cells of the higher

water plant Lemna gibba show a transient depolarization of the highly negative
interior upon addition of nitrate which is similar to that found upon addition
of glucose, amino acids, or phosphate (NOVACKY et al. 1980, ULLRICH-EBERIUS
et al. 1981). The nitrate signal is small in non-induced material (NOVACKY et al.
1978), but has a saturation value of 65 mV in induced plants (ULLRICH and
NOVACKY 1981). These data obtained with Lemna can be best explained by
a nitrate-proton co-transport, in which nitrate is transported with an excess
of at least one proton over neutralization by a carrier in a passive process
following the electrochemical gradient across the plasmalemma, LlfiH+ (inside
negative). The excess protons would then be extruded through the plasmalemma
by the active, ATP-dependent H+ -extrusion pump (Fig. 1). In the case of
nitrate, its reduction would contribute by consuming one or even two of the
protons transported with the nitrate, so that the H+ -extrusion pump would
be at least partly relieved of its work. In the marine diatom Phaeodactylum,
sodium-dependence of nitrate uptake suggests aNa +INO) co-transport (REES
et al. 1980). On the other hand, FALKOWSKI (1975) reported on the presence
of a nitrate- or chloride-activated membrane ATPase in six different species
of marine phytoplankton (one species gave negative results) and proposed a
mechanism involving directly ATP dependent trarisport for nitrate.

4 Nitrite Uptake in Algae

In contrast to nitrate and ammonium, nitrite is of minor importance for nitrogen

nutrition of algae and higher plants, due to its lower stability and its presence
only at low concentrations under natural conditions. As an intermediary product
of nitrate reduction, it is sometimes accumulated in small amounts within the
cells, more often released to the medium and readily taken up again as soon
as the conditions have improved for its reduction. In many algae studied, nitrite
uptake reached higher rates in the light than nitrate uptake, especially under
anaerobic conditions (KESSLER 1957, HATTORI 1962, GRANT and TURNER 1969,
MORRIS and AHMED 1969, ULLRICH 1974, CALERO et al. 1980), consistent with
the fact that nitrite reduction is more directly linked with photosynthetic electron
transport than nitrate reduction (KESSLER and ZUMFT 1973).
In the dark, or in the absence of carbon sources in the light, nitrite is
often formed in excess from nitrate and released to the medium (KESSLER 1957,
ULLRICH 1974, OHMORI 1978, RIGANO 1978). In most algae studied, the Ks
values for the uptake of nitrite are slightly higher than those for nitrate. This
implies that in most cases nitrite uptake also follows a saturation kinetics curve,
indicating the involvement of a carrier. At high concentrations of nitrite, devia-
tions from this saturation kinetics towards diffusion kinetics may occur
(ULLRICH 1974).
When present in combination with nitrate, nitrite plays a variable role. In
some cases it inhibits nitrate uptake in a competitive type of reaction, in which
390 W.R. ULLRICH:

the combined uptake rates of both ions may greatly exceed that of nitrate
alone (EISELE 1976). For unknown reasons nitrite assimilation in vivo is often
more sensitive than nitrate uptake and reduction to uncouplers such as dinitro-
phenol (KESSLER 1964, HOFMANN 1972), so that dinitrophenol can be used in
the dark for the in vivo assay of nitrate reductase (e.g. FISCHER and SIMONIS
1979). The competitive inhibition of nitrate uptake by nitrite and vice versa
may lead to the conclusion that both anions are taken up via the same transport
system, but in the algae this is unlikely: in Chlorella fusca the effects of mono-
chromatic light were limited to nitrate (CALERO et al. 1980) and in many algae
the effect of ammonia is less pronounced on nitrite, uptake than on nitrate
uptake. There are also often differences in pH dependence, but the interpretation
of these data has to include the possibility of effects on nitrite reduction. There-
fore, the location of nitrite reduction within the chloroplasts may playa role.
In Ankistrodesmus (ULLRICH 1974) and also in Chlorella (CALERO et al. 1980)
a non-linear time course of nitrite uptake is observed, with a rapid initial uptake
and reduction followed by an almost complete stand-still, followed in tum by
a recovery to high rates. This time course can be explained in accordance with
experiments in intact chloroplasts of higher plants. In such chloroplasts nitrite
causes acidification and thus a drastic but transient inhibition of photosynthetic
O 2 evolution (PURCZELD et al. 1978), but in Scenedesmus no internal pH changes
were observed (LARSSON and ANDERSSON 1982).

5 General Remarks on Regulation of Nitrate

and Nitrate Uptake

If the intracellular nitrate or nitrite pool approaches saturation through the

operation of the accumulation process, and at the same time nitrate or nitrite
reduction is low or zero, then the transport mechanism will slow down or
stop net influxes. If we assume the functioning of a proton co-transport system
(see Sect. 3.10), the slow-down of transport may also be due to intracellular
acidification or depolarization at rates that cannot be sufficiently counteracted
by the active and probably ATP-dependent H+ extrusion pump. Even if we
assume the operation of a specific nitrate-transport ATPase as proposed by
FALKOWSKI (1975) the energy supply to the pumping system will still be strongly
regulatory in ion uptake. Nitrate reductase activity, in tum, is likely to be
very dependent on nitrate accumulation by the uptake system, since all known
preparations of nitrate reductase have a relatively high Km. Although direct
effects of light are possible, the main reason for light stimulation of nitrate
or nitrite uptake may be found in the energy supply from photosynthesis or
respiration. Even the lower saturation rates in monochromatic light, observed
only for nitrate uptake, can be explained in this manner (VOSKRESENSKAYA 1972,
CALERO et al. 1980). But in many cases nitrate uptake, especially by algae with-
out storage vacuoles, is certainly regulated by nitrate reduction. This applies
also to light stimulation.
11.5 Uptake and Reduction of Nitrate: Algae and Fungi 391

Although it is not always easy to separate regulation of nitrate uptake from

that of nitrate reduction, nitrate uptake can also be regulated at the level of
protein synthesis. During nitrate starvation in Ankistrodesmus and Platymonas,
nitrate uptake capacity increases several-fold during the first 3 h (EISELE 1976,
ULLRICH et al. 1981) and this stimulation is sensitive to the protein synthesis
inhibitor, cycloheximide (ULLRICH et al. 1981). However, in synchronous Chlor-
ella cultures, nitrate uptake is stimulated soon after the start of the light period,
whereas the main light-induced increase of nitrate reductase activity follows
only after a lag phase (TISCHNER and LORENZEN 1979).
As described before, ammonium when added in the presence of carbon
sources, immediately stops nitrate assimilation without affecting nitrate reduc-
tase activity in the first 30 to 60 min. Although it has been postulated for
a long time that an intermediate of ammonium assimilation rather than ammoni-
um itself must be responsible for this inhibition at the uptake site, no special
inhibitor has yet been identified. In Ankistrodesmus, in contrast to many other
algae, ammonium has no regulatory effect in the absence of carbon sources,
but nitrate uptake and reduction still proceed at considerable rates (see
Sect. 3.5). In the unicellular red alga, Cyanidium caldarium, RIGANO et al. (1979)
could completely rule out regulation of nitrate uptake by ammonium by adding
methionine sulfoximine, an inhibitor of glutamine synthetase. Like Ankistrodes-
mus in the absence of carbon sources, Cyanidium in the presence of this inhibitor
and in the simultaneous presence of CO 2 , shows uptake and reduction of
nitrate without ammonium control. The same was found recently with Anacystis,
and also with azaserine, an inhibitor of glutamate synthase, instead of methio-
nine sulfoxiInine (FLORES et al. 1980). Some more details on regulation have
been dealt with in the preceding parts of this chapter.

6 Uptake and Reduction of Nitrate and Nitrite in Fungi

There has been much research on fungal nitrate and nitrite assimilation, which
can be only briefly summarized in this chapter. Fungi are interesting organisms,
because at least some of them are completely autotrophic in nitrogen nutrition
in spite of their heterotrophy in carbon nutrition. They have special uptake
systems for nitrate and nitrite subject to genetic or metabolic regulation, nitrate
and nitrite reductases with a well-studied regulation system and, for ammonium
assiInilation, glutaInine synthetase, glutamate synthase and also glutamate dehy-
drogenase. Recent reviews were given by HEWITT (1975), by GARETT and AMY
(1978), by BEEVERS and HAGEMAN (1980), and by GUERRERO et al. (1981).
Nitrate reductase has been isolated and purified from various species of
fungi, but the most complete picture has been elaborated for Neurospora crassa
and Aspergillus nidulans (GARRETT and NASON 1967, 1969, for more literature
see GARRETT and AMY 1978, VENNESLAND and GUERRERO 1979). The procedure
is similar to that described for the enzyme of algae or higher plants. More
recently, affinity chromatography on blue dextran sepharose columns has been
392 W.R. ULLRICH:

used (GUERRERO and GUTIERREZ 1977, DOWNEY and STEINER 1979). The proper-
ties of nitrate reductase from Neurospora and Aspergillus were similar to those
from algae and higher plants. The enzyme is a metalloflavoprotein containing
FAD, molybdenum, and a hemoprotein identified as cytochrome b-557. It is
sensitive to SH-group inhibitors as mercuribenzoates (PCMB), and also to met-
al-complexing agents as cyanide and azide. The most conspicuous difference
to the nitrate reductase of algae and higher plants is the substrate specificity
for NADPH (code number EC 1.6.6.3). The enzyme can also catalyze several
partial reactions and consists of the two moieties, diaphorase and terminal
nitrate reductase. The diaphorase is sensitive to heat and SH-group inhibitors,
the terminal reductase to cyanide and azide. The sensitivity to cyanide, sulfide,
or azide can be prevented by preincubation of the enzyme with nitrate and
can be reversed by oxidizing NADPH via the diaphorase moiety using ferricya-
nide or cytochrome c as electron acceptors. AMY et al. (1977) proposed a cyclic
system in Neurospora with four forms of more or less active nitrate reductase
based on oxidation or reconstitution of SH -groups and complexing with FAD
or NADP. Molecular weights vary betwen 196,000 for Aspergillus, where four
subunits of 49,000 could be separated (DOWNEY and STEINER 1979) and more
than 228,000 for Neurospora (GARRETT and NASON 1969), where the enzyme
was split into two subunits of 130,000 and 115,000. Genetical control of nitrate
reductase has been widely studied in both species by various mutants whose
extracts allowed recombination of enzyme activities (see GARRETT and AMY
1978).
Fungal nitrite reductase differs much more from that of photosynthetic or-
ganisms than does nitrate reductase. It is specific for NADH or NADPH and
requires FAD, but does not accept electrons from reduced ferredoxin, probably
because ferredoxin is not usually available in these heterotrophic organisms.
Fungal nitrite reductase has a higher molecular weight (290,000), perhaps due
to the additional diaphorase, similar to that of nitrate reductase, by which
electrons from two-electron donors become available. Over-reduction in the
presence of FAD and NAD(P)H as well as oxidation, e.g. by hydrogen peroxide,
can lead to inactivation of the enzyme, which is generally labile in vitro and
must be stabilized by addition of FAD and dithionite. The terminal part of
the enzyme that transfers the electrons to nitrite, contains siroheme, as shown
by spectral analysis and comparison with bacterial sirohemes (VEGA et al. 1975).
Correspondingly, nitrite reductase is sensitive to cyanide, CO, and other iron-
z
complexing agents. CO was shown to compete with NO as a ligand in siro-
heme. Also the reduction of nitrite, a six-electron process, seems to occur at
z
a siroheme-NO complex.
An independent nitrate uptake system in Neurospora crassa was reported
to cause 50-fold accumulation of nitrate within the mycelium against the medium
with a Ks of 0.25 mM for nitrate (SCHLOEMER and GARRETT 1974). The uptake
system was apparently only formed in the presence of nitrate or nitrite; its
formation was inhibited by cas amino acids, cycloheximide, puromycin, and 6-
methylpurine. Ammonium and nitrite are non-competitive inhibitors of nitrate
uptake only, but uptake of both nitrate and nitrite is inhibited by metabolic
poisons such as cyanide and dinitrophenol, indicating that both processes are
11.5 Uptake and Reduction of Nitrate: Algae and Fungi 393

somehow related to active transport, but do not proceed via the same carrier.
Neurospora mutants devoid of nitrate reductase still show normal uptake kinet-
ics, thus demonstrating that uptake and reduction are completely independent
processes (SCHLOEMER and GARRETT 1974).
Also in fungi regulation of nitrate and nitrite reduction seems' to act at
several levels. A series of genes was found to be responsible for nitrate and
nitrite reductase activities (COVE 1966, GARRETT and AMY 1978). Ammonium
may act as a repressor in Aspergillus (COVE 1966), nitrate and nitrite as inducers.
In some fungal species, ammonium and few amino acids promote the decay
of nitrate reductase in vivo and even more that of the nitrate uptake system.
This effect can be prevented by protein synthesis inhibitors (LEWIS and FINCHAM
1970, GOLDSMITH et al. 1973). This suggests the involvement of special proteins
in the enzyme decay, probably proteolytic inactivators (SORGER et al. 1978).
In Neurospora, the repression of nitrate reductase seems to be not due to ammo-
nia itself but to glutamine (PREMAKUMAR et al. 1979). In addition to the genetic
control, strong and rapid membrane depolarization upon addition of ammoni-
um, as shown also in Neurospora (SLAYMAN 1977), could lead to a break-down
of anion-proton cotransport and thus cause some of the ammonium effects
by direct inhibition of nitrate uptake.
Also in fungi light may contribute to the regulation of nitrate reduction.
Methylviologen-nitrate reductase activity in Neurospora crassa is stimulated by
blue light, whereas the NADPH-nitrate reductase activity is decreased (KLEMM
and NINNEMANN 1979), an observation that can be explained by light sensitivity
of the flavin and siroheme constituents of the enzyme.

References

Ahmed J, Spiller H (1976) Purification and some properties of the nitrate reductase
from Ankistrodesmus braunii. Plant Cell Physiol17: 1-10
Amy NK, Garrett RH, Anderson MB (1977) Reactions of the Neurospora crassa nitrate
reductase with NAD(P) analogs. Biochim Biophys Acta 480: 83-95
Aparicio PJ, Roldl'm JM, Calero FS (1976) Blue light reactivation of nitrate reductase
from green algae and higher plants. Biochem Biophys Res Commun 70:1071-1077
Bates SS (1976) Effects of light and ammonium on nitrate uptake by two species of
estuarine phytoplankton. Limnol Oceanogr 21 :212-218
Beevers L, Hageman RH (1980) Nitrate and nitrite reduction. In: Stumpf PK, Conn
EE (eds) The biochemistry of plants, vol V. Academic Press, London New York
Bothe H (1977) Flavodoxin. In: Trebst A, Avron M (eds) Encyclopedia of plant physiolo-
gy, New Ser, vol V. Springer, Berlin Heidelberg New York ,
Butz RG, Jackson WA (1977) A mechanism for nitrate transport and reduction (review).
Phytochemistry 16:409-417
Calero F, Ullrich WR, Aparicio PJ (1980) Regulation by monochromatic light of nitrate
uptake in Chlorellafusca. In Senger H (ed) The blue light syndrome. Springer, Berlin
Heidelberg New York, pp 411-421
Chaparro A, Rosa de la MA, Vega JM (1979) Involvement of oxygen in Chlorellafusca
nitrate reductase inactivation by reduced nicotinamide adenine dinucleotide. Z Pflan-
zenphysiol 95: 77-85
394 W.R. ULLRICH:

Cove DJ (1966) The induction and repression of nitrate reductase in the fungus Aspergillus
nidulans. Biochim Biophys Acta 113:51-56
Cresswell RC, Syrett PJ (1979) Ammonium inhibition of nitrate uptake by the diatom,
Phaeodactylum tricornutum. Plant Sci Lett 14:321-325
Czygan FC (1965) Zur Frage der Zwischenprodukte der Nitratreduktion bei Griinalgen.
Planta 64:301-311
Dalling MJ, Tolbert NE, Hageman RH (1972) Intracellular location of nitrate reductase
and nitrite reductase. Biochim Biophys Acta 283:505-512
Davis EA (1953) Nitrate reduction by Chlorella. Plant PhysioI28:53~544
Diez J, Chaparro A, Vega JM, Relimpio AM (1977) Studies on the regulation of as simila-
tory nitrate reductase in Ankistrodesmus braunii. Planta 137:231-234
Downey RJ, Steiner FX (1979) Further characterization of the reduced nicotinamide
adenine dinucleotide phosphate: nitrate oxidoreductase in Aspergillus nidulans. J Bac-
teriol137:105-114
Eisele R (1976) Photosynthetische Nitrataufnahme und Nitratreduktion bei Ankistrodes-
mus braunii. Ph D thesis, Tech Hochsch Darmstadt
Eisele R, Ullrich WR (1975) Stoichiometry between photosynthetic nitrate reduction
and alkalinisation by Ankistrodesmus braunii in vivo. Planta 123: 117-123
Eisele R, Ullrich WR (1977) Effect of glucose and CO 2 on nitrate uptake and coupled
OH- flux in Ankistrodesmus braunii. Plant Physiol59: 18-21
Eppley RW, Rogers IN (1970) Inorganic nitrogen assimilation of Ditylum brightwellii,
a marine plankton diatom. J PhycoI6:344-351
Eppley RW, Rogers IN, McCarthy JJ (1969) Half-saturation constants for uptake of
nitrate and ammonium by marine phytoplankton. Limnol Oceanogr 14:912-920
Falkowski PG (1975) Nitrate uptake in marine phytoplankton: Comparison of half-
saturation constants from seven species. Limnol Oceanogr 20:412-417
Falkowski PG (1977) A theoretical description of nitrate uptake kinetics in marine phy-
toplankton based on bisubstrate kinetics. J Theor BioI 64: 375-379
Fischer S, Simonis W (1979) Tagesperiodische Schwankungen und lichtinduzierte
Zunahme der Nitratreduktase-Aktivitat bei Synchronkulturen von Ankistrodesmus
braunii. Z Pflanzenphysiol92: 143-152
Flores E, Guerrero MG, Losada M (1980) Short-term ammonium inhibition of nitrate
utilization by Anacystis nidulans and other cyanobacteria. Arch Microbiol
128:137-144
Funkhouser EA, Ramadoss CS (1980) Synthesis of nitrate reductase in Chlorella. II.
Evidence for synthesis in ammonia-grown cells. Plant Physiol 65: 944-948
Garrett RH, Amy NK (1978) Nitrate assimilation in fungi. Adv Microb Physiol18: 1-65
Garrett RH, Nason A (1967) Involvement of a b-type cytochrome in the assimilatory
nitrate reductase of Neurospora crassa. Proc Natl Acad Sci USA 58: 1603-1610
Garrett RH, Nason A (1969) Further purification and properties of Neurospora nitrate
reductase. J BioI Chern 244:2870-2882
Goldsmith J, Livoni JP, Norberg CL, Segel IH (1973) Regulation of nitrate uptake
in Penicillium chrysogenum by ammonium ion. Plant Physiol 52: 362-367
Grant BR, Turner 1M (1969) Light-stimulated nitrate assimilation in several species of
algae. Comp Biochem PhysioI29:995-1004
Guerrero MG, Gutierrez M (1977) Purification and properties of the NAD(P)H: nitrate
reductase of the yeast Rhodotorula glutinis. Biochim Biophys Acta 482: 272-285
Guerrero MG, Vega JM,Losada M (1981) The assimilatory nitrate-reducing system
and its regulation. Annu Rev Plant Physiol 32: 16~204
Hanisak MD, Harlin MM (1978) Uptake of inorganic nitrogen by Codiumfragile subspec.
tomentosoides (Chlorophyta). J Phyco114:450-453
Harlin MM, Craigie JS (1978) Nitrate uptake by Laminaria longicruris (Phaeophyceae).
J PhycoI14:464-467
Hattori A (1962) Light-induced reduction of nitrate, nitrite and hydroxylamine in a
blue-green alga, Anabaena cylindrica. Plant Cell Physiol 3: 355-369
Hattori A, Myers J (1967) Reduction of nitrate and nitrite by subcellular preparations
of Anabaena cylindrica. II. Reduction of nitrate to nitrite. Plant Cell Physiol8: 327-337
11.5 Uptake and Reduction of Nitrate: Algae and Fungi 395

Heber U (1974) Metabolite exchange between chloroplasts and cytoplasm. Annu Rev
Plant Physiol 25: 393-421
Heimer YM (1973) The effects of sodium chloride, potassium chloride and glycerol on
the activity of nitrate reductase of a salt-tolerant and two non-tolerant plants. Planta
113: 279-281
Hewitt EJ (1975) Assimilatory nitrate-nitrite reduction. Annu Rev Plant Physiol
26:73-100
Hipkin CR, Syrett PJ (1977) Some effects of nitrogen-starvation on nitrogen and carbohy-
drate metabolism in Ankistrodesmus braunii. Planta 133:209-214
Ho CH, Ikawa T, Nisizawa K (1976) Purification and properties of a nitrate reductase
from Porphyra yezoensis Ueda. Plant Cell Physiol17:417-430
Hofmann A (1972) Ober die energetische Kopplung der Nitratassimilation von Griinalgen
an Atmung und Photosynthese. Planta 102:72-84
Kessler E (1957) Untersuchungen zum Problem der photochemischen Nitratreduktion
in Griinalgen. Planta 49: 505-523
Kessler E (1964) Nitrate assimilation by plants. Annu Rev Plant Physiol 15: 57-72
Kessler E, Zumft WG (1973) Effect of nitrite and nitrate on chlorophyll fluorescence
in green algae. Planta 111 :41-46
Klemm E, Ninnemann H (1979) Nitrate reductase - a key enzyme in blue light-promoted
conidiation and absorbance change of Neurospora. Photochem Photo bioi 29: 629-632
Lane AE, Burris JE (1981) Effects of environmental pH on the internal pH of Chlorella
pyrenoidosa, Scenedesmus quadricauda, and Euglena mutabilis. Plant Physiol
68:439-442
Larsson C-M, Andersson M (1982) Uptake and photoreduction of NO; and NO; in
Scenedesmus: Interactions with CO 2 fixation. In: Akoyunoglou G (ed) Photosynthesis
Vol IV, Proc 5th Int Congr Photosynthesis, Halkidiki, Greece
LeClaire JA, Grant BR (1972) Nitrate reductase from Dunaliella tertiolecta, purification
and properties. Plant Cell Physiol 13: 899-907
Lewis CM, Fincham JRS (1970) Regulation of nitrate reductase in the basidiomycete
Ustilago maydis. J Bacteriol 103: 55-61
Lorimer GH, Gewitz HS, Volker W, Solomonson LP, Vennesland B (1974) The presence
of bound cyanide in the naturally inactivated form of nitrate reductase of Chlorella
vulgaris. J Bioi Chern 249: 6074-6079
Losada M, Guerrero MG (1979) The photosynthetic reduction of nitrate and its regula-
tion. In: Barber J (ed) Photosynthesis in relation to model systems. Amsterdam,
Elsevier, pp 365-408
Manzano C, Candau P, Gomez-Moreno C, Relimpio AM, Losada M (1976) Ferredoxin-
dependent photosynthetic reduction of nitrate and nitrite by particles of Anacystis
nidulans. Mol Cell Biochem 10:161-169
Manzano C, Candau P, Guerrero MG (1978) Affinity chromatography of Anacystis
nidulans ferredoxin-nitrate reductase and NADP reductase on reduced ferrodoxin-
sepharose. Anal Biochem 90:408-412
Morris I, Ahmed J (1969) The effect oflight on nitrate and nitrite assimilation by Chlorella
and Ankistrodesmus. Physiol Plant 22:1166-1174
Morris I, Syrett PJ (1963) The development of nitrate reductase in Chlorella and its
repression by ammonium. Arch MikrobioI47:32-41
Novacky A, Fischer E, Ullrich-Eberius CI, Liittge U, Ullrich WR (1978) Membrane
potential changes during transport of glycine as a neutral amino acid and nitrate
in Lemna gibba G1. FEBS Lett 88: 264-267
Ohmori M (1978) Nitrite excretion by a blue-green alga, Oscillatoria rubescens DC. Arch
Hydrobiol 83 :485-493
Ortega T, Rivas J, Cardenas J, Losada M (1977) Metabolic interconversion offerrodoxin-
nitrate reductase and NADP reductase of Nostoc muscorum. Biochem Biophys Res
Commun 78: 185-193
Pistorius EK, Gewitz HS, Voss H, Vennesland B (1976) Reversible inactivation of nitrate
reductase in Chlorella vulgaris in vivo. Planta 128: 73-80
Pistorius EK, Jetschrnann K, Voss H, Vennesland B (1979) The dark respiration of
396 W.R. ULLRICH:

Anacystis nidulans. Production of HCN from histidine and oxidation of basic amino
acids. Biochim Biophys Acta 585: 630-642
Premakumar R, Sorger GJ, Gooden D (1979) Nitrogen metabolite repression of nitrate
reductase in Neurospora crassa. J Bacteriol 137: 1119-1126
Purczeld P, Chon CJ, Portis AR, Heldt HW, Heber U (1978) The mechanism of the
control of carbon fixation by the pH in the chloroplast stroma. Studies with nitrite-
mediated proton transfer across the envelope. Biochim Biophys Acta 501 : 488-498
Raven JA (1976) Transport in algal cells. In: Liittge U, Pitman MG (eds) Transport
in plants II. Encyclopedia of plant physiology new ser, vol 2. Springer, Berlin Heidel-
berg New York
Raven JA, DeMichelis MI (1979) Acid-base regulation during nitrate assimilation in
Hydrodictyon africanum. Plant Cell Environ 2:245-257
Rees TAV, Cresswell RC, Syrett PJ (1980) Sodium-dependent uptake of nitrate and
urea by a marine diatom. Biochim Biophys Acta 596: 141-144
Ricketts TR, Edge PA (1977) The effect of nitrogen refeeding of starved cells of Platy-
monas striata Butcher. Planta 134: 169-176
Rigano C (1978) Studies in vivo on the control by ammonia of nitrate reduction to
nitrite in the unicellular alga Cyanidium caldarium. Plant Sci Lett 13:301-307
Rigano C, DiMartino-Rigano V, Vona V, Fuggi A (1979) Glutamine synthetase activity,
ammonia assimilation and control of nitrate reduction in the unicellular red alga
Cyanidium caldarium. Arch Microbiol121: 117-.120
Roldan JM, Calero F, Aparicio PJ (1978) Photoreactivation of spinach nitrate reductase:
role offlavins. Z PflanzenphysioI90:467-474
Rosa de la MA, Diez J, Vega JM, Losada M (1980) Purification and properties of
assimilatory NAD(P)H-nitrate reductase from Ankistrodesmus braunii. Eur J Biochem
106:249-256
Rosa de la MA, Vega JM, Zumft WG (1981) Composition and structure of assimilatory
nitrate reductase from Ankistrodesmus braunii. J BioI Chern 256: 5814-5819
Schloemer RH, Garrett RH (1974) Nitrate transport system in Neurospora crassa. J
Bacteriol 118: 259-269
Serra JL, Llama MJ, Cadenas E (1978a) Nitrate utilization by the diatom Skeletonema
costatum. I. Kinetics of nitrate uptake. Plant Physiol 62: 987-990
Serra JL, Llama MJ, Cadenas E (1978b) Nitrate utilization by the diatom Skeletonema
costatum. II. Regulation of nitrate uptake. Plant PhysioI62:991-994
Slayman CL (1977) Energetics and control of transport in Neurospora. In: Jungreis AM,
Hodges TK, Kleinzeller A, Schultz SG (eds) Water relations in membrane transport
in plants and animals. Academic Press, London New York
Smith FA, Raven JA (1979) Intracellular pH and its regulation. Annu Rev Plant Physiol
30:289-312
Solomonson LP (1975) Purification ofNADH-nitrate reductase by affinity chromatogra-
phy. Plant Physiol 56: 853-855
Solomonson LP, Vennesland B (1972a) Nitrate reductase and chlorate toxicity in Chlor-
ella vulgaris Beijerinck. Plant Physiol 50:421-424
Solomonson LP, Vennesland B (1972b) Propen:ies of a nitrate reductase of Chlorella.
Biochim Biophys Acta 267:544-557
Sorger GJ, Premakumar R, Gooden D (1978) Demonstration in vitro of two intracellular
inactivators of nitrate reductase from Neurospora. Biochim Biophys Acta 540:33-47
Sosa FM, Cardenas J (1977}NADPH as electron donor for nitrate reduction in Chlamy-
domonas reinhardi. Z Pflanzenphysiol 85: 171-175 .
Syrett PJ, Leftley JW (1976) Nitrate and urea assimilation by algae. 'In:Sunderland
N (ed) Perspectives in experimental biology, vol II. Botany. Pergamon, Oxford New
York
Thacker A, Syrett PJ (1972) The assimilation of nitrate and ammonium by Chlamydomo-
nas reinhardi. New Phytol 71 :423-433
Tischner R, Lorenzen H (1979) Nitrate uptake and nitrate reduction in synchronous
Chlorella. Planta 146:287-292
11.5 Uptake and Reduction of Nitrate: Algae and Fungi 397

Tischner R, Lorenzen H (1981) Nitrate uptake and reduction in Chlorella- Characteriza-

tion of nitrate uptake in nitrate-grown and nitrogen-starved Chlorella sorokiniana.
In: Bothe H, Trebst A (eds) Biology of inorganic nitrogen and sulfur. Springer,
Berlin Heidelberg New York
Tromballa HW, Broda E (1971) Das Verhalten von Chlorellafusca gegeniiber Perchlorat
und Chlorat. Arch MikrobioI78:214-223
Ullrich WR (1971) Nitratabhangige nichtcyclische Photophosphorylierung bei Ankistro-
desmus braunii in Abwesenheit von CO 2 und Oz. Planta 100: 18-30
Ullrich WR (1972) Der EinfluB von COz und pH auf die 3zP-Markierung von Polyphos-
phaten und organischen Phosphaten bei Ankistrodesmus braunii im Licht. Planta
102:37-54
Ullrich WR (1974) Die nitrat- und nitritabhangige photosynthetische 0z-Entwicklung
in N z bei Ankistrodesmus braunii. Planta 116: 143-152
Ullrich WR, Eisele R (1977) Relations between nitrate uptake and nitrate reduction
in Ankistrodesmus braunii. In: Thellier M, Monnier A, Demarty M, Dainty J (eds)
Echanges ioniques transmembranaires chez les vegetaux. Colloq CNRS
Ullrich WR, Novacky A (1981) Nitrate-dependent membrane potential changes and their
induction in Lemna gibba G1. Plant Sci Lett 22: 211-217
Ullrich WR, Schmitt H-D, Arntz E (1981) Regulation of nitrate uptake in green algae
and duckweeds. Effect of starvation and induction. In: Bothe H, Trebst A (eds)
Biology of inorganic nitrogen and sulfur. Springer, Berlin Heidelberg New York
Ullrich-Eberius CI (1973) Beziehungen der Aufnahme von Nitrat, Nitrit und Phosphat
zur photosynthetischen Reduktion von Nitrat und Nitrit und zum ATP-Spiegel bei
Ankistrodesmus braunii. Planta 115: 25-36
Ullrich-Eberius CI, Novacky A, Fischer E, Liittge U (1981) Relationship between energy-
dependent phosphate uptake and the electrical membrane potential in Lemna gibba
G1. Plant Physiol 67: 797-801
Vega JM, Herrera J, Aparicio PJ, Paneque A, Losada M (1971) Role of molybdenum
in nitrate reduction by Chlorella. Plant Physiol 48: 294-299
Vega JM, Herrera J, Relimpio A, Aparicio PJ (1972) NADH-nitrate reductase de Chlor-
ella: nouvelle contribution al'etude de ses proprietes. Physiol Veg 10:637-652
Vega JM, Garrett RH, Siegel LM (1975) Siroheme: a prosthetic group of the Neurospora
crassa assimilatory nitrite reductase. J BioI Chern 250: 7980-7989
Vennesland B, Guerrero MG (1979) Reduction of nitrate and nitrite. In: Gibbs M,
Latzko E (eds) Photosynthesis II Encyclopedia of plant physiology new ser, vol 6.
Springer, Berlin Heidelberg New York
Voskresenskaya NA (1972) Blue light and carbon metabolism. Annu Rev Plant Physiol
23:219-234
Walker NA (1976) Membrane transport: theoretical background. In: Liittge U, Pitman
MG (eds) Transport in plants II. Encyclopedia of plant physiology new ser, vol
2, part A. Springer, Berlin Heidelberg New York
Warburg 0, Negelein E (1920) Uber die Reduktion der Salpetersaure in griinen Zellen.
Biochem Z 110:66-115
Zumft WG (1972) Ferredoxin: nitrite oxidoreductase from Chlorella. Purification and
properties. Biochim Biophys Acta 276:363-375
III. Metabolism of Sulfur and Phosphorus
111.1 Reduction and Other Metabolic Reactions
of Sulfate
J.A. SCHIFF

1 Introduction
Sulfur, like nitrogen, is required to form the ubiquitous proteins and coenzymes
of plants. It is not surprising, then, that this element is fairly uniformly distrib-
uted throughout the tissues and organs of higher plants and in the cells of
plant-like microorganisms. Since sulfur is indispensable for plant growth and
development, sulfur deficiency results in discoloration and abnormal growth
of plant tissues. Sulfur deficiency influences nitrogen metabolism as well (FRIED-
RICH and SCHRADER 1978). The importance of sulfur in the nutrition of plants
and, therefore, in agriculture is discussed in several reviews (McLACHLAN 1975,
THOMPSON et al. 1970, ASLANDER 1958, BAUMEISTER 1958) and in the quarterly
periodical Sulphur Research and Development, published by The Sulphur Insti-
tute, Washington, D.C.
This chapter is intended to summarize our present understanding of sulfate
metabolism in the biological world, particularly the biochemical and enzymatic
reactions in cells which are involved in sulfate utilization and reduction. The
better we understand these cellular reactions the better will be our understanding
of the role of sulfur in inorganic plant nutrition.
A concise chapter of this sort cannot deal with all the details of the subject.
Useful general reviews of sulfate metabolism are those of Roy and TRUDINGER
(1970), SCHIFF and HODSON (1973), SIEGEL (1975) and ANDERSON (1980); nutri-
tional and ecological aspects are discussed by ANDERSON (1978). The sulfated
polysaccharides are described by PERCIVAL and McDOWELL (1967) in HARVEY
and McLACHLAN (1973) and by CRAIGIE and MCCANDLESS (1979); further refer-
ences will be found in SCHIFF and HODSON (1973).
The accumulation of sulfuric acid in the vacuoles of the marine brown alga
Desmerestia has been summarized (SCHIFF 1963) and the production of volatile
sulfur compounds by algae was discussed by CHALLENGER (1959). The metabo-
lism of reduced sulfur compounds formed from sulfate including the sulfur-
containing coenzymes, peptides such as glutathione, and the sulfur-containing
amino acids are reviewed by various authors in GREENBERG (1975); further
information on glutathione and the y-glutamyl cycle is available (GRIFFITH et al.
1979).
Regulation of the metabolic pathways leading to the sulfur-containing amino
acids is comprehensively discussed by UMBARGER (1978) and by GIOVANELLI
et al. (1980).
The remainder of this chapter will be concerned with our current understand-
ing of the reactions of sulfate metabolism with particular emphasis on the enzy-
mology, biochemistry and evolution of the sulfate-reducing pathways.
 402 J.A. SCHIFF:

2 The Place of Sulfate Reduction in the Sulfur Cycle

Unlike most of the other elements required by living systems, carbon, nitrogen
and sulfur undergo extensive metabolic transformations. Sulfur, unlike carbon
and nitrogen, can be utilized in its most highly oxidized naturally occurring
form, sulfate; sulfate reduction is necessary for the formation of sulfur-contain-
ing amino acids and proteins (SCHIFF and HODSON 1973, Roy and TRUDINGER
1970, SIEGEL 1975).
Figure 1 shows the relation of sulfate to the many reactions which sulfur
undergoes in the biosphere. Sulfate is used by living systems to form sulfate
esters of polysaccharides, phenols, steroids and other organic compounds
through a series of activation and transfer reactions. Sulfate is reduced by certain
anaerobic organisms such as the bacterium Desulfovibrio in their energy metabo-
lism where, during anaerobic respiration, sulfate is used in place of oxygen
to oxidize substrates releasing energy and hydrogen sulfide (PECK 1959, PECK
1962a, b, SIEGEL 1975, Roy and TRUDINGER 1970). Many organisms reduce
sulfate to the thiollevel found in several coenzymes, and the amino acids cysteine
and methionine which serve as building blocks for peptides and proteins (GREEN-
BERG 1975); this process is called assimilatory sulfate reduction. Most organisms
(including higher animals and higher plants) oxidize reduced sulfur to sulfate
(SCHIFF and HODSON 1973, SINGER 1975) although the aerobic chemosynthetic
bacteria are the only organisms which have been observed to couple the energy
released (some 180 kcal per mol) to the reduction of carbon dioxide (SIEGEL
1975). The anoxygenic photosynthetic bacteria use reduced sulfur compounds
as photosynthetic electron donors thereby oxidizing them to sulfate (SIEGEL
1975). Recent reviews will be found in BOTHE and TREBST (1981).

Reduction (+.!W)
Assimilatory reduction
(plants, microorganisms)

Dissimilatory reduction
Polysaccharide (Desulfovibrio)
sulfate
Steroidal
sulfates
S02-
4 '-----
S20 23--
___ --So----- H2S~

{
Cysteine

1~ ::::= Protein
Chemosynthetic sulfur bacteria
Methionine

Animals, plants (? ), microorganisms

Oxidation ( -~F)
Fig. 1. Reactions of sulfur in the biosphere
111.1 Reduction and Other Metabolic Reactions of Sulfate 403

3 Phylogenetic Distribution of Reactions Involving

Sulfate Transfer and Reduction

Reactions resulting in the esterification of organic compounds by sulfate are

widely distributed (SCHIFF and HODSON 1973, DEMElo 1975). Sulfate esters of
organic compounds are formed by higher animals in the form of steroid sulfates,
phenol sulfates, and polysaccharide sulfates such as chondroitin sulfate. Luci-
feryl sulfate formed by sulfate transfer from PAPS serves as a reservoir of
luciferin for bioluminescence (CORMIER et al. 1970, 1974). The algae produce
sulfated polysaccharide as wall constituents, frequently in copious amounts
(SCHIFF and HODSON 1973, CALLOW et al. 1978, BINGHAM and SCHIFF 1979,
RAMUS 1974, RAMus and GROVES 1974, QUATRANO 1978, CRAIGIE and MCCAND-
LESS 1979). The better-known sulfated polysaccharides of commercial impor-
tance such as carrageenan and agar are produced by marine members of the
red algae or Rhodophyta. There is at least one report of sulfated polysaccharides

0 0
0 >-
N Multicellular .c
0 0..
a; higher plants 0
:l: <II
:l:

Green algae
B B
.~ ~
0 0
~ Yello w alg ae ~
Red algae

0 0
Q; Blue-green Q;
c c
0 algae 0
:l: :l:
(eyanobac erial

OJ - H' 100 Keal/ los

50 2- HS- 200 Kea l /M los
Oissimil a ory sol- HS- presen _

Fig, 2. Loss of energetically expensive reduction reactions in various contemporary groups

of organisms. Dissimilatory sulfate reduction is restricted to two small genera of anaerobic
bacteria (Desulfovibrio and Desulfomaculum). Clear areas indicate organisms that can
carry out the reduction of sulfate, nitrate and carbon dioxide [a small group of aerobic
chemosynthetic bacteria (not shown) are capable of carrying out carbon dioxide reduction
given oxidizable inorganic substrates as energy sources]. The various patterns indicate
the loss of these reduction reactions in various groups. Tpe loss of all three reduction
pathways has occurred in the evolution of the protozoa and higher animals
404 I.A. SCHIFF:

among the prokaryotes (bacteria and cyanobacteria or blue-green algae) in Halo-

coccus (STEBER and SCHLEIFER 1975), but none in higher plants. These gelatinous
materials may have evolved as adaptations to periodic wetting and drying as
water conservation mechanisms, a particular problem for benthic and terrestrial
algae. In the higher animals, the sulfated polysaccharides are used to provide
a matrix for cellular structures such as cartilage and skin. From our present
knowledge the occurrence of sulfated polysaccharides seems to represent the
adaptation of particular groups of organisms to special needs.
The distribution of reduction reactions, however, seems to follow a more
regular evolutionary pattern (Fig. 2) (SCHIFF and HODSON 1973). Reduction
reactions are expensive, energetically, and require the investment of hundreds
of kcal per mol to allow the reaction to take place (Fig. 2). The reduction
of nitrate, carbon dioxide and sulfate seems to have been lost in the evolution
of the primitive animals, the protozoa, and this deficiency persists in the evolu-
tion of the multicellular animals. These reactions may have been lost when
photosynthetic reactions trapping light energy were lost. Animals require energy
for movement, and having lost photosynthetic energy found themselves with
a tight energy budget. Being predators, they could fulfill their architectural
and energy requirements for reduced sulfur, nitrogen and carbon by eating
organisms which could perform these reductions, such as microorganisms or
plants, originating the food chains we know today. Contemporary animals
obtain their reduced sulfur compounds by eating or harboring sulfate-reducing
plants or microorganisms, or by eating other animals which do so.

4 Sulfate Uptake, Activation and Transfer

Sulfate uptake and its control has been well studied in many systems (SCHIFF
and HODSON 1973, SIEGEL 1975, SMITH 1975, 1976, CACCO et al. 1977, RENASTO
and FERRARI 1975). Uptake seems to be accomplished through active transport
mediated by a carrier enzyme system. Although the sulfate content of Lemna
increased greatly with increased sulfate in the medium, the concentration of
other cellular sulfur compounds showed little change (DATKO et al. 1978).
As far as is known, the cellular metabolism of sulfate begins with a series
of activation reactions with ATP to form the nucleoside phospho sulfates adeno-
sine 5' -phosphosulfate (APS) and adenosine 3' -phosphate 5' -phosphosulfate
(PAPS) (Fig. 3) (SCHIFF and HODSON 1973, DEMElO 1975, PASTERNACK et al.
1965, DEMElO et al. 1955, HILZ and LIPMANN 1955, ROBBINS and LIPMANN
1957, 1958, REUVENY and FILNER 1977, REUVENY 1977, FARLEY et al. 1978).
(In crabgrass, Digitaria sanguinalis, a C 4 plant, 90% of the enzyme activity
forming APS was found in extracts from bundle-sheath cells (GERWICK and
BLACK 1979)). APS is formed against an unfavorable equilibrium since the
free energy of hydrolysis of the phospho sulfate bond is higher than that of
the pyrophosphate linkage (RoY and TRUDINGER 1970). To offset this unfavor-
able equilibrium, pyrophosphate removal through inorganic pyrophosphatase
activity and/or reaction of APS with another molecule of ATP to form PAPS
111.1 Reduction and Other Metabolic Reactions of Sulfate 405

o
II ATP sulfurylase
-O-S-O-
II + W
o

ATP Sulfate

o
0
II II
+ HO-P-O-P-OH
I I
0_ 0_

APS Pyrophosphate

+ ATP
APS kinase
..

APS

ADP

Pyrophosphate

PAPS o
o 0 Inorganic II
2 _O-P-O_ + 2H'
II II
HO-P-O-P-OH + H2 0 Pyrophosphatase I
I I OH
0_ 0_
Orthophosphate
Fig. 3. Enzymatic reactions involved in sulfate activation. The enzymes shown are ATP
sulfurylase (ATP: sulfate adenylyl transferase 2.7.7.4), APS kinase (ATP: adenylyl sulfate
3' phosphotransferase 2.7.1.25), and inorganic pyrophosphatase (pyrophosphate phos-
phohydrolase 3.6.1.1)

are used to allow an accumulation of nucleoside phospho sulfates. A widely

distributed 3' (2'), 5'-diphosphonucleoside 3' (2')-phosphohydrolase (DPNPase)
converts PAPS to APS and, together with inorganic pyrophosphatase, may
facilitate APS formation despite the unfavorable equilibrium of the first reaction
forming APS (Fig. 4) (TSANG and SClllFF 1976c). This enzyme may also provide
406 J.A. SCHIFF:

control of APS and PAPS levels in organisms where both are required for
different purposes.
All sulfate transfer reactions resulting in sulfate esterification appear to use
PAPS as the sulfate donor as far as is known (DEMElO 1975, SCHIFF and HODSON
1973). APS is a substrate for hydrolases that release sulfate (TSANG and SCHIFF
1976c) and for a novel enzyme reaction (adenylyl sulfate: NH3 adenylyl transfer-
ase (APSAT) which converts APS and ammonia to adenosine 5'-phosphorami-
date (FANKHAUSER et al. 1981 a) (this activity was previously thought to be
a cyclase-forming cAMP; cAMP shares many properties with adenosine 5'-
phosphoramidate; TSANG and SCHIFF 1976c). A molecule with the properties
of adenosine 5'-phosphoramidate has been isolated from Chlorella (FANKHAUSER
et al. 1981 b). Sulfonic acids which contain sulfur at the redox level of sulfite,
occur in many living systems, the plant sulfolipid of chloroplast thylakoids
being an important example (SCHIFF and HODSON 1973, HARWOOD and NI-
CHOLLS 1979, HARWOOD 1980, BENSON 1963, HAINES 1973, HOPPE and SCHWENN
1981). Although the formation of the sulfonic acid group of compounds such
as taurine or isethionate (HOSKIN and KORDIK 1977) can arise through oxidation
of reduced sulfur in animals (SmGEL 1975) or, perhaps, from sulfate via PAPS
and serine (MARTIN et al. 1974), the formation of the sulfonic acid group of
the plant sulfolipid appears to come more-readily from sulfate via as yet un-
known reactions. The formation of sulfonic acids is an area requiring more
study. Sulfur dioxide and sulfite are readily metabolized by plants and their
chloroplasts (ZmGLER 1977, ZmGLER and HAMPp 1977, PLESNICAR 1977, SILVIUS
et al. 1976) but S02 can also cause changes in metabolism (PAUL and BASSHAM
1978) and produce injury (BRESSAN et al. 1978).
Natural products containing both oxidized and reduced sulfur are formed
by plants including glucosinolates and polyacetylenes (see HODSON and SCHIFF
1973).

5 Sulfate Reduction

Dissimilatory sulfate reduction occurs in certain anaerobic bacteria such as

Desulfovibrio which use sulfate as an oxidant in place of molecular oxygen.
In this process APS is the nucleotide sulfate donor for reduction, resulting
in the accumulation of hydrogen sulfide and A TP synthesis through oxidative
phosphorylation (RoY and TRUDINGER 1970, SmGEL 1975, PECK 1970, ISHIMOTO
and FUJIMOTO 1959, PECK et al. 1965). A light-dependent emission of hydrogen
sulfide from higher plants has been described (WILSON et al. 1978).
Assimilatory sulfate reduction, where sulfate is reduced to form the thiol
groups of sulfur-containing amino acids, coenzymes and other organic com-
pounds, now appears to exhibit two distinct patterns (TSANG and SCHIFF 1975,
SCHIFF and HODSON 1973):
a) In most oxygen-evolving photosynthesizers including all eukaryotic algae
and some prokaryotic blue-green algae (GOLDSCHMIDT et al. 1975, TSANG and
SCHIFF 1975, SCHMIDT 1977), all higher plants tested (SCHMIDT 1975a) and
 Sulfate Transferases o ~ ....
....
esters
NH, -~-O-S-O_ ....
....
NXI
I.
N~1,.fH
N
0 CH,-O 6_ ~
H H
Id
~ H (1)
N 0
Cysteine
OH II ~
o-p-o_ H H o·
Adenosine 3' phosphate I I I ~
I OH DPNPase Chiarella -s-c-c-coo--..
5' phosphosulfate (PAPS) I I , §
H NH, '\ p..
1 0 0 \
I , \ NH, II II OAS sulfhydrase----t
o
0_
II II 0II - AT\.... N~~"
I. I -I H 0 CH,-o-r-o-rr-O-\f:\
0 0 o H NH, 11 [
-O-S-O- J - -O-~-O ATP "N N H H - +NH [G-S-] II I I /
3 H3 C- C- 0 -C-C-coo- ~
& 1 0 ',ll\,,,,l~. l) ApS'l APS AA I
APS ~
I P-Pi OH OH _ ammonia sulfohydrolase O-Acetyl serine I 2-
Sulfate I ~ul:ate I. Adenosine 5' phasphosulfate ad~~YI ) sulfotransfera~ 0 [G S-S-] . I ri'
outside I Inside 2Pi (APS) transferase
1
s at ,.3.- J-.~Q>h. t.----...... FerredoxIn I Id
(1)
""Q'q, '<lc

SOi- /' SO'-

4
--A- ',Cto> '4..0
s" "'' 4.;%
oxidized I
I ~
a.
5' AMP <it"
[G-S-sO,J .',& 1 o
+ + I
FerredoxIn ~ I ~
Adenosine 5' 5' AMP reduced I I o...,
phosphoramidate I CIl
+ R-S- I S
o I ;;J'
II Sulfite I (t
When R-S- = Monothiol, product is sulfite -o-s-o- - - - - - - - - - - - - S ' -
When R-S- = Ring-forming dithiol, product is sulfite Sulfite reductase
Sulfide
When R-S- = Vicinal dithiol, product is thiosulfate inside
1
----+---
I
o
It
-O-S-O-
Sulfite
outside
Fig. 4. APS pathway of sulfate reduction in Chiarella. Reactions thought to be on the main path of sulfate reduction in vivo are shown
as solid lines. Reactions blocked in various mutants that cannot grow on sulfate (Sat- mutants) are shown by dashes through the reactions
deleted. Side reactions are shown by alternate dashes and dots. Sulfite (or thiosulfate) is only produced through chemical reactions requiring
o
"'"
thiols and probably only occurs in vivo when sulfite (or thiosulfate) is available to the cells from outside. G-S - designates reduced glutathione -.J
408 J.A. SCHIFF:

spinach chloroplasts (SCHMIDT and SCHWENN 1971, SCHMIDT 1975b, FANK-

HAUSER and BRUNOLD 1978), the nucleotide sulfate donor for reduction is APS
via a highly specific APS sulfotransferase (adenylyl sulfate: thiol sulfotransfer-
ase) which will not use PAPS (Fig. 4) (TSANG and SCHIFF 1976a). Where this
form of reduction has been studied in detail, e.g. Chlorella, Euglena and spinach
chloroplasts (SCHMIDT and SCHWENN 1971, BRUNOLD and SCHIFF 1976, SCHMIDT
et al. 1974, ABRAMS and SCHIFF 1973) a low molecular weight thiol carrier (which
may be glutathione in Chlorella; TSANG and SCHIFF 1976a) is also involved,
but the APS sulfotransferase will transfer promiscuously to other added thiols
in the presence or absence of the carrier (TSANG and SCHIFF 1976a). Ferredoxin
(in the Chlorella and higher plant system) or NADPH (in Euglena extracts)
must be added to achieve reduction of the sulfo group of APS to the thiol
level of amino acids.
b) In organisms which lack oxygen-evolving photosynthesis [besides animals,
which do not reduce sulfate to the thiol level (SCHIFF and HODSON 1973)],
such as yeast and E. coli and other bacteria (TSANG and SCHIFF 1975, TORRI
and BANDURSKI 1967, WILSON et al. 1961, ASAHI et al 1961, LEINWEBER and
MONTY 1963, DREYFUSS and MONTY 1963a, b, FUllMOTO and ISHIMOTO 1961,
TSANG and SCHIFF 1976b, 1978b, WILSON and BIERER 1976) and a few oxygen-
evolving blue-green algae (SCHMIDT 1977), PAPS is the nucleoside phosphosul-
fate donor for reduction via a specific PAPS sulfotransferase (3' phosphoadeny-
lyl sulfate: thiol sulfotransferase) which will not use APS (TSANG and SCHIFF
1976b, PECK 1961) (Fig. 5). The transfer of the sulfo group of PAPS and further
reduction requires thioredoxin in yeast and E. coli (PORQuE et al. 1970, TSANG
and SCHIFF 1976b, 1978b). Only reduced pyridine nucleotide needs to be added
to obtain reduction, but the PAPS sulfotransferase will only function in the
presence of thioredoxin (TSANG and SCHIFF 1976b) or glutaredoxin (TSANG
and SCHIFF 1976b, TSANG 1981); other added thiols in the absence ofthioredoxin
or glutaredoxin are inactive. Interestingly, APS sulfotransferases requiring thio-
redoxin have been found recently in certain cyanobacteria (SCHMIDT 1982). Thio-
redoxin has also been implicated in the regulation of chloroplast enzymes by
light (BUCHANAN 1980).

5.1 Detailed Reactions of the Two Assimilatory Pathways

5.1.1 The APS Pathway

Our most detailed information comes from work on the enzymes from Chlorella,
Euglena, spinach chloroplasts (SCHMIDT and SCHWENN 1971, SCHMIDT 1975b,
BRUNOLD and SCHIFF 1976, SCHMIDT et al. 1974, ABRAMS and S.CHIFF 1973,
GOLDSCHMIDT et al. 1975, TSANG and SCHIFF 1976a, 1978a, FANKHAUSER and
BRUNOLD 1978) and duckweed (Lemna) (BRUNOLD and SCHMIDT 1978) (Fig. 4).
It has taken some time to sort out the reactions of APS sulfotransferase because
the enzyme transfers the sulfo group of APS not only to the normal carrier
[which may be glutathione (TSANG and SCHIFF 1978a) in Chlorella or a some-
what larger molecule in spinach chloroplasts (SCHMIDT and SCHWENN 1971)]
lILt Reduction and Other Metabolic Reactions of Sulfate 409

but to practically any thiol added (TSANG and SCHIFF 1976a). Thiols forming
rings on oxidation (such as dithiothreitol) form sulfite in this reaction, mono-
thiols (such as glutathione and thioethanolamine) form the Bunte salt (R-S-
SO;) of the thiol and, in the presence of excess thiol, sulfite. Vicinal dithiols
such as 2,3-dithiopropanol (BAL) form thiosulfate CS-SO;) wherf} the oxi-
dized sulfur comes from APS and the reduced sulfur from the thiol. Of all
the thiols, glutathione stands out because it has a much higher activity with
the enzyme than any other monothiol and shows kinetics suggestive of a regula-
tory interaction. Glutathione seems to be the only small molecule in Chlorella
extracts which is active in the APS sulfotransferase reaction (TSANG and SCHIFF
1978a). If thiols are omitted from the reaction, it is possible to demonstrate
that the enzyme transfers the sulfo group of APS to an enzyme-bound acceptor
in the crude extracts (ABRAMS and SCHIFF 1973). This labeled material shows
the properties of a Bunte salt or organic thiosulfate (R - S - SO;) and is thought
to be the sulfo form of the thiol carrier. Reconstruction experiments using
glutathione - S - SO; show that sulfite and thiosulfate can be formed chemically
by a non-enzymatic reaction with the appropriate thiol (TSANG and SCHIFF
1976a). While the transfer of the sulfo group from APS to the enzyme-bound
carrier is heat-labile (and, therefore, enzymatic), the subsequent reaction of
the enzyme-bound Bunte salt with added thiols is non-enzymatic and the same
products are formed with each thiol as in the model experiments (TSANG and
SCHIFF 1976a, ABRAMS and SCHIFF 1973).
Thus we view the reaction catalyzed by APS sulfotransferase as the transfer
of the sulfo group of APS to the thiol of a carrier such as glutathione (G-S-)
to form the Bunte salt (G-S-SO;). As will be discussed shortly, further
physiological reduction appears to involve this bound sulfo group. The reductive
release of sulfite or thiosulfate by addition of thiols results from chemical side
reactions of this carrier-bound sulfo group. Regulation of APS sulfotransferase
by HzS and cysteine has been described in Lemna (BRUNOLD and SCHMIDT
1976, 1978).
There are two candidates for enzymes catalyzing reduction to the thiollevel.
One is sulfite reductase (hydrogen-sulfide:NADP+ oxidoreductase 1.8.1.2)
which is present in these systems and will bring about the reduction of free
sulfite to free sulfide with reduced pyridine nucleotides (SIEGEL 1975, SIEGEL
et al. 1973, 1974, SIEGEL and DAVIS 1974, SCHMIDT et al. 1974, TSANG and SCHIFF
1976b, ASADA et al. 1969). The other enzyme is an organic thiosulfate reductase
(ferredoxin: sulfoglutathione oxidoreductase) (formerly called thiosulfonate re-
ductase) which reduces glutathione-S-SO; to bound sulfide (probably gluta-
thione-S-S-) or, dithionite to sulfide, with reduced ferredoxin (SCHMIDT 1973,
SCHMIDT et al. 1974). Studies of a mutant Chlorella (Sat~) which cannot grow
on or reduce sulfate have shown that organic thiosulfate reductase activity
is extremely low or absent while normal levels of sulfite reductase activity are
present (SCHMIDT et al. 1974). Thus the presence of sulfite reductase alone does
not allow sulfate reduction to proceed in vivo and shows that organic thiosulfate
reductase is the preferred enzyme for sulfate reduction in this system. We view
the physiological reaction as the reduction of G - S - SO; to G - S - S - which
seems to be the primary reduction reaction in vivo in the APS system. Sulfite
410 J.A. SCHIFF:

reductase would probably act on free sulfite only if it arose through side reac-
tions in vivo or if it was taken up from outside by the cell.
These systems contain a very active O-acetyl serine sulfhydrase [O-acetyl-L-
serine acetate-lyase (adding H 2 S)4.2.99.8] which forms cysteine from free sulfide
or from carrier-bound sulfide (BRUNOLD and SCHIFF 1976, SCHMIDT et al. 1974).
Two electrons would be required to convert G - S - S - to the thiol group of
cysteine in this reaction.
An important advance in studies of this system was finding conditions for
the reduction of APS to form cysteine in cell-free extracts without the addition
ofthiols (SCHMIDT et al. 1974) which form free sulfite and sulfide through chemi-
cal side reactions (TSANG and SCHIFF 1976a). Studies of this cell-free system
from wild-type Chlorella and mutants blocked for organic thiosulfate reductase
and APS sulfotransferase have proven to be consistent with the interpretations
offered above and have led to a scheme in which what are thought to be the
normal reactions in vivo are shown in solid lines and the side reactions in
broken lines (Fig. 4).
Studies of the cellular location and synthesis of the various enzymes of
this pathway have been undertaken. The activating enzymes for sulfate, APS
sulfotransferase, organic thiosulfate reductase (reductant, ferredoxin) and 0-
acetyl serine sulfhydrase have been found in spinach chloroplasts (SCHMIDT
and SCHWENN 1971, SCHWENN and TREBST 1976, FANKHAUSER and BRUNOLD
1979). It is likely that this is also the case in Chlorella, judging from the close
similarity of green algal and higher plant systems. In Euglena, however, which
is an animal-like cell resembling the protozoa but containing chloroplasts, APS
sulfotransferase is found in the mitochondria and microbodies and the organic
thiosulfate reductase (reductant, NADPH) and OAS sulfhydrase are found in
the mitochondria (BRUNOLD and SCHIFF 1976); thus the mitochondria of Euglena
contain all of the presently known enzymes of the bound pathway of APS
reduction. It is interesting that green algal and higher plant cells cannot be
induced to lose their chloroplasts, perhaps because certain enzyme systems (such
as sulfate reduction) which are essential for viability are plastid-localized. In
Euglena, however, chloroplasts and plastid DNA can be lost (SCHIFF 1972).
Euglena does not seem to localize essential reactions (other than those essential
for photosynthesis) in the plastids.
In the course of studies on the APS sulfotransferase another activity was
found which acts on APS (TSANG and SCHIFF 1976c), an APS sulfohydrolase
(adenylyl sulfate sulfohydrolase 3.6.2.1) which forms sulfate and AMP from
APS. Other sulfohydrolases are known (DODGSON and ROSE 1975). PAPS is
much more stable than APS in extracts; perhaps this is why the DPNPase
reaction forming APS from PAPS is advantageous as an APS source.

5.1.2 The PAPS Pathway

This system has been best studied in yeast and Escherichia coli (TSANG and
SCHIFF 1975, TORII and BANDURSKI 1967, WILSON et al. 1961, FUJIMOTO and
ISlllMoTO 1961, LEINWEBER and MONTY 1963, DREYFUSS and MONTY 1963a,
b, TSANG and SCHIFF 1976b, WILSON and BIERER 1976) (Fig. 5). PAPS sulfo-
 .....
.....
~
.....
I
RED.NADP I CysG,I,J
I o
I
I O-Acetyl serine
o H NH2
=
PAP\~HSO; II I I [
Ht_H3C_C_0-~-~_COO- o
PAPS
[
- OAS SuJfhydrase ~
Sulfotransferase
--CysH HH ~

-S-<:-C-ooO-
I I
g.
NH2 0 0 H NH2 ~
N....... ......." H 0 CH2-0-P-0-S-O_ Cysteine ~
"" S-
~ N
III~
~ H H 0_
I I I0 I a.
ATP I N H NADP"~Tr~_S_S-G NADPH +H'" 8
+ 0 V 'S- \~
Thioredoxin Glutathione '"o
....,
ase
APS kin~. OH O-~-O_ reductase S reductase
Adenosine 3' phosphate 6H NADPH/ "-i I 2G-S-~ "-NADP ~
I 5' phosphosulfate (PAPS) lit'
+ H'" 's ft
I
NH2 0 0
o I 0
II I II ATP II II
\. N ...... H 0
-O-S-O-
II
--+--
I
-O-S-O-
II
JC: I ~ CH2-0-l-0-~-O-
o 0
ATP "'" ~ N--t;-~ 0_ 0
I sulfuryIase P-Pi WI
Sulfate I Sulfate i OH OH
outside I inside 2Pi Adenosine 5' phosphosulfate
(APS)
Fig. 5. PAPS pathway of sulfate reduction in E. coli. Thioredoxin is required in the PAPS sulfotransferase reaction and formation of
sulfite, but the reactions it undergoes have not been elucidated fully. Mutants (Cys) blocked in various enzyme reactions are shown by ...,.
dashes through the reactions deleted. G-S - designates reduced glutathione. T<~ =designates reduced thioredoxin .....
412 J.A. SCHIFF:

transferase (also called PAPS reductase before thioredoxin was identified as

a component) is specific for PAPS as the sulfate donor and is also completely
specific for thioredoxin; other thiols will not replace thioredoxin (TSANG and
SCHIFF 1976b) or glutaredoxin (TSANG and SCHIFF 1978b, TSANG 1981). As
in Chlorella, an enzyme-bound intermediate with the properties of a Bunte
salt (R-S-SO~) can be detected as a product of the PAPS sulfotransferase
reaction (TSANG and SCHIFF 1976b, TORII and BANDURSKI 1967). The formation
of free sulfite requires thioredoxin reductase (NADPH: oxidized-thioredoxin
oxidoreductase 1.6.4.5) and reduced pyridine nucleotide (TORII and BANDURSKI
1967, WILSON et al. 1961, ASAIn et al. 1961, PORQuE et al. 1970) but these can
be replaced by appropriate thiols (TSANG and SCHIFF 1976b) to yield sulfite
or thiosulfate as in the APS system. A glutaredoxin can accept electrons from
the glutathione-glutathione reductase system [NAD(P)H: oxidized-glutathione
oxidoreductase 1.6.4.2)] (TSANG and SCHIFF 1978b, HOLMGREN et al. 1978,
TSANG 1981). The product of this PAPS sulfotransferase-thioredoxin system
is sulfite when NADPH is used as the reductant. Free sulfite is reduced to
sulfide with reduced pyridine nucleotides via sulfite reductase (SmGEL 1975).
E. coli and Salmonella mutants blocked for sulfate reduction lack sulfite
reductase activity, suggesting that this enzyme system is the physiological reac-
tion in these organisms (DREYFUS and MONTY 1963a, b, JONEs-MORTIMER 1968a,
b, LEINWEBER and MONTY 1963, HODSON and SCHIFF 1971, TSANG and SCHIFF
1978b). O-acetyl serine sulfuydrase is also present, (KREDICH and TOMPKINS
1966) which can form cysteine from O-acetyl serine and sulfide. Taking together
the information presented here, and suggestions made previously by various
workers, it is possible to write a scheme consistent with current knowledge
(Fig. 5).
The role of thioredoxin in this system is still under study. Reduced thiore-
doxin (Tr< =) could act as the thiol carrier acceptor for the sulfo group of
PAPS in the PAPS sulfotransferase reaction forming T~=-S03 (sulfothiore-

doxin). Elimination of sulfite would yield oxidized thioredoxin (T<~) which

could then be reduced as shown in Fig. 5. In view of the ready elimination
of sulfite from Bunte salts of non-vicinal dithiols (TSANG and SCHIFF 1976a),
some mechanism would be necessary to prevent the immediate release of sulfite
from sulfothioredoxin since the bound intermediate formed from PAPS in E.
coli is rather stable (TSANG and SCHIFF 1976b). Perhaps linkage of one of the
thiol groups of thioredoxin to a thiol group of the transferase via an S - S bond
accomplishes this. Another possibility is suggested by recent work which shows
that a large proportion of the thioredoxin has a thiol group substituted with
phosphate (PIGmT and CONLEY 1978). Perhaps this substitution of one thiol
group by phosphate prevents spontaneous elimination of sulfite from sulfothio-
redoxin until timely enzymatic removal ofthe phosphate. It is possible, of course,
that thioredoxin is not the carrier and that the bound intermediate contains
the sulfo group bound to a different acceptor. In this case, thioredoxin would
act as the immediate reductant.
111.1 Reduction and Other Metabolic Reactions of Sulfate 413

It is worth noting that although the bound intermediates in both the APS
and PAPS systems show many reactions of Bunte salts, frequently not all of
the activity is exchangeable with sulfite (TSANG and SCHIFF 1976b, TORII and
BANDURSKI 1967) and further work is needed to prove that these are definitely
organic thiosulfates and that they account for all of the activity in both types
of systems. Also unexplained is why thiosulfate can be formed from APS in
Chlorella extracts in the absence of added thiols (SCHIFF and LEVINTHAL 1968,
LEVINTHAL and SCHIFF 1968) when NADPH is used as the reductant; a donor
of reduced sulfur must be present in the extracts themselves.

5.2 Location of Sulfate Reduction in Tissues and Organs

of Multicellular Plants
We have already mentioned that sulfate reduction is localized in the chloroplasts
of higher plants. However, cell, tissue and organ cultures from a wide variety
of plant parts including roots can be grown on media where sulfate is the
sole source of sulfur (WHITE 1963, WHITE and GROVE 1965, THOMAS and DAVEY
1975, BUTENKO 1968). Non-green tissues of plants such as beet discs (ELLIS
1963) and roots (PATE 1965) reduce sulfate. Enzymes of sulfate reduction such
as APS sulfotransferase (SCHMIDT 1976, FANKHAUSER and BRUNOLD 1978) and
O-acetylserine sulfhydrylase (FANKHAUSER and BRUNOLD 1979, TAMURA et al.
1976) are found in roots and hypocotyls. Enzymes which may be involved
in sulfate reduction but which may participate in other functions, such as ATP
sulfurylase (ELLIS 1969, ONAJOBI et al. 1973, CACCO et al. 1977) and sulfite reduc-
tase (MAYER 1967) are also found in roots and hypocotyls. It appears that
sulfate reduction may be widely distributed throughout the plant. There may
be large quantitative differences, however, in the rates, since the activity of
APS sulfotransferase in spinach roots, for example, is far lower than in fully
green leaves (FANKHAUSER and BRUNOLD 1978). In green leaves and chloroplasts,
sulfate reduction is a light-dependent process, but in non-green tissues energy
and reducing power for sulfate reduction must come from non-photosynthetic
sources. It remains for future work to determine whether the sulfate reduction
system in non-green tissues is localized in plastids or in other organelles, but
O-acetyl serine sulfhydrylase is found in plastids in spinach roots (FANKHAUSER
and BRUNOLD 1979). In this connection it is worth remembering that sulfate
reduction in higher plant chloroplasts requires reduced ferredoxin but when
sulfate reduction is localized in mitochondria as in Euglena, NADPH is the
reductant.

6 Speculations on the Origin and Evolution of Pathways

of Sulfate Reduction

It is likely that the most primitive form of sulfate reduction is the dissimilatory
pathway of the anaerobes (RoY and TRUDINGER 1970, SIEGEL 1975, PECK 1970)
(Fig. 6) which may have survived with little modification from the anaerobic
414 J.A. SCHIFF:

Higher plants
Animals
(PAPS-->sulfate esters) lAPS 1_'1 Cy-st-ein-e-'I

" ,,
----
Loss of
PAPS reduction
FPox FPred
~

(PAPS-->sulfate esters)
...----.....
Cytcox Cyt C,ed

ADP + Pi ~

I PAPS I ~ lAPS 1---1 Cysteine I

Aerobic r~eSPiration~EarIY assimilatory sulfate reductIOn

I APS I - I Cysteine I .
. Aeroblc
----------- H
--
2H+
- - - - - - - - - - -Anaerobic
--
ATP ~
Anaerobic '\ ~
respiration ) Cyt c30x Cyt c3red
+ ADP+Pi ~
Dissimilatory ~ Desulfovibrio
sulfate Ferredoxin red. ·Ferredoxin ox.
reduction
ATP ~6e-
SO~- --\1APs"1Y: so0---[!J
PPi
+
2Pi \
Primitive anaerobic procaryotes

Fig. 6. Suggested evolutionary relationships of sulfate-reducing pathways in various

groups. Dissimilatory sulfate reduction, as part of anaerobic respiration, probably arose
in the anaerobic phase of the origin of life and has persisted in the modern procaryotic
anaerobes such as Desulfovibrio which uses sulfate as an oxidant in respiration forming
ATP and sulfide. This pathway begins with APS .. With the release of oxygen into the
atmosphere, oxygen became the electron acceptor in respiration and a separation of
respiration and sulfate reduction occurred leading to the establishment of assimilatory
sulfate reduction, still using APS, as the activated sulfate for reduction. For as yet un-
known reasons, the PAPS pathway for reduction evolved from the APS pathway and
modern prokaryotes (bacteria and blue-green algae or cyanobacteria) have one or the
other. The APS pathway has persisted in the evolution of the eukaryotic photosynthetic
oxygen evolvers including the eukaryotic algae and higher plants. The PAPS pathway
leads to yeast, among the fungi. The animals have lost the ability to reduce PAPS to
the thiol level of the amino acids but retain PAPS as a donor of the sulfo group in
esterification reactions via specific sulfotransferases. The eukaryotic algae also use PAPS
as the donor of the sulfo group in esterification reactions although they use APS as
the substrate for reduction

phase of the origin of life and evolution. In the absence of molecular oxygen,
other oxidants were pressed into service to oxidize substrates. The result was
anaerobic respiration employing sulfate or nitrate as oxidants with the conse-
quent formation of their reduced forms, sulfide and nitrogen gas. In organisms
such as Desulfovibrio the formation of A TP from the oxidation of substrates
111.1 Reduction and Other Metabolic Reactions of Sulfate 415

is coupled to dissimilatory sulfate reduction which serves as a sink for the

electrons removed in the oxidation of the substrate (hydrogen in the present
example). This dissimilatory pathway uses APS as the activated form of sulfate
for reduction and the pathway proceeds through the formation of free sulfite
and sulfide. In an anaerobic environment, there is no danger from chemical
oxidation of these intermediates by molecular oxygen. Anaerobic conditions
also allow the undisturbed participation of autoxidizable low potential reduc-
tants such as ferredoxin in the sulfate reduction pathway. Multielectron reduc-
tions utilizing low potential reductants such as those involved here, in sulfite
reduction, in nitrite reduction and in nitrogen fixation very likely evolved under
the early anaerobic conditions in which life originated.
When oxygen came into the atmosphere through the activities of the first
oxygenic photo synthesizers (very likely early prokaryotes related to the cyano-
bacteria or blue-green algae) (see SClllFF 1972, SClllFF and LYMAN 1982) aerobic
respiration using oxygen as the electron acceptor evolved and the organisms
carrying out anaerobic respiration were forced into anaerobic niches (such as
deep mud) to survive. With the availability of oxygen, I suppose, came the
separation of respiration and sulfate reduction in the aerobic organisms. Respi-
ration now employed the reduction of oxygen to water (Fig. 6) in the formation
of ATP. Sulfate reduction, although no longer required for respiration, was
still necessary for the formation of reduced sulfur compounds including cysteine
and methionine of protein. This separation of sulfate reduction from respiration
led to the establishment of assimilatory sulfate reduction as a separate process
to provide these reduced sulfur compounds.
This assimilatory sulfate reduction process was, at first I suppose, similar
to the process of dissimilatory sulfate reduction in the anaerobes in that it
employed APS as the activated intermediate and reduced ferredoxin as the
reductant. Free sulfite and sulfide react with oxygen, however, and this may
have been the point at which carrier-bound intermediates such as organic thio-
sulfates (G-S-SO;) and persulfides (G-S-S-) which are not autoxidizable,
were introduced into the APS pathway. This process would have been present
in the early aerobic prokaryotes and is found among contemporary prokaryotes
such as the cyanobacteria or blue-green algae (TSANG and SClllFF 1976b,
SCHMIDT 1977). Chloroplasts and mitochondria of modern eukaryotes may have
originated from the invasion of non-photosynthetic non-respiratory eukaryotes
by appropriate procaryotic endosymbionts such as primitive blue-green algae
and bacteria (SClllFF 1972, SClllFF and LYMAN 1982). In view of this, it is not
surprising that the APS pathway of assimilatory sulfate reduction with ferre-
doxin as reductant is found in eukaryotic algae such as Chlorella (TSANG and
SCHIFF 1975, SCHMIDT et al. 1974) and higher plants (SCHMIDT 1975a, BRUNOLD
and SCHMIDT 1978) and in spinach has been shown to be localized in the·chloro-
plasts (SCHMIDT and SCHWENN 1971, SCHWENN and TREBST 1976). This isprob-
ably one of the reasons why chloroplast function can never be eliminated com-
pletely from green algae and higher plants even when reduced carbon is avail-
able; essential functions such as sulfate reduction (and nitrite reduction) are
localized in the plastids. Organisms more closely related to the animals such
as Euglena do not put essential functions other thim carbon dioxide fixation
in their plastids and when provided with a reduced carbon source plastid func-
416 J.A. SCHIFF:

tion can be eliminated completely. In Euglena, sulfate reduction is localized

in the mitochondria and microbodies (BRUNOLD and SCHIFF 1976). In this organ-
ism, the APS pathway characteristic of oxygenic eukaryotes is present but the
reductant is NADPH rather than reduced ferredoxin which is available in eu-
karyotes only in the chloroplasts as far as we know.
Although the APS pathway is pervasive among eukaryotes having oxygenic
photosynthesis, another pathway appears to have evolved at the same time.
While some prokaryotes use the APS pathway of assimilatory sulfate reduction,
others use a PAPS pathway (TSANG and SCHIFF 1975, SCHMIDT 1977), suggesting
that the PAPS pathway evolved from the APS pathway at or after the establish-
ment of assimilatory sulfate reduction (Fig. 6). This pathway is found in bacteria
such as E. coli (TSANG and SCHIFF 197~, 1976b, 1978b, DREYFUSS and MONTY
1963a, b, FUJIMOTO and ISHIMOTO 1961) and in fungi such as yeast (ToRII and
BANDURSKI 1967, WILSON and BIERER 1976) where the reductant is NADPH.
Since PAPS is formed from APS it presumably arose later in evolution than
APS but why PAPS was selected for assimilatory sulfate reduction in these
organisms is not readily explained. As we have already noted (Fig. 2) sulfate
reduction was lost in the evolution of the protozoa and higher animals; the
pathway from sulfate to PAPS formation has been conserved, however, and
PAPS serves as the source of sulfate for esterification reactions via PAPS sulfo-
transferases of various specificities (DEMElO 1975). PAPS also serves as the
donor for esterification reactions in organisms that use APS for sulfate reduc-
tion, suggesting that PAPS formation was fairly common among primitive pro-
karyotes, but was selected for use in assimilatory sulfate reduction only in certain
bacteria and in non-photosynthetic eukaryotes such as the fungi.

Acknowledgments. The support of a series of grants from the National Science Foundation
(pCM 76-21486) is gratefully acknowledged. The technical assistance of Mrs. LINDA
CORRADO is appreciated as is the secretarial help of Mrs. MARGARET KING. NANCY
O'DONOGHUE prepared the illustrations. HEINZ FANKHAUSER provided several references
and helpful suggestions.

References
Abrams WR, Schiff JA (1973) Studies of sulfate utilization by algae. 11. An enzyme-
bound intermediate in the reduction of adenosine 5' phospho sulfate (APS) by cell-free
extracts of wild-type Chiarella and mutants blocked for sulfate reduction. Arch Mikro-
bioI 94:1-10
Anderson JW (1978) Sulphur in biology. Arnold, London
Anderson JW (1980) Assimilation of inorganic sulfate into cysteine. In: Stumpf PK,
Conn EE (eds) Plant biochemistry, vol V. Academic Press, London New York
Asahi T, Bandurski RS, Wilson LG (1961) Yeast sulfate-reducing system. II. Enzymatic
reduction of protein disulfide. J BioI Chern 236:1830-1835
Asada K, Tamura G, Bandurski RS (1969) Methyl viologen-linked sulfite reductase
from spinach leaves. J BioI Chern 244:4904-4915
Aslander A (1958) Nutritional requirements of crop plants. Encycl Plant Physiol IV: 977-
1019
Baumeister W (1958) Hauptniihrstoffe. In: Ruhland W (ed) Encyclopedia of plant physi-
ology, vol IV. Springer, Berlin Gottingen Heidelberg
111.1 Reduction and Other Metabolic Reactions of Sulfate 417

Benson AA (1963) The plant sulfolipid. Adv Lipid Res 1 :387-394

Bingham S, Schiff JA (1979) Conditions for attachment and development of single cells
released from mechanically disrupted thalli of Prasiola stipitata Suhr. BioI Bull
156:257-271
Bothe H, Trebst A (eds) (1981) Biology of inorganic nitrogen and sulfur. Springer, Berlin
Heidelberg New York
Bressan RA, Wilson LC, Pilner P (1978) Mechanisms of resistance to sulfur dioxide
in the Cucurbitaceae. Plant Physiol 61: 761-767
Brunold C, Schiff JA (1976) Studies of sulfate utilization by algae. 15. Enzymes of
assimilatory sulfate reduction in Euglena and their cellular localization. Plant Phys~ol
57:430-436
Brunold C, Schmidt A (1976) Regulation of adenosine 5' phosphosulfate sulfotransferase
activity by H 2 S in Lemna minor. Planta 133: 85-88
Brunold C, Schmidt A (1978) Regulation of sulfate assimilation in plants. 17. Cysteine
inactivation of adenosine 5' phosphosulfate sulfotransferase in Lemna minor L. Plant
Physiol 61 : 342-347
Buchanan BB (1980) Role of light in the regulation of chloroplast enzymes. Annu Rev
Plant Physiol31 :341-374
Butenko RG (1968) Plant tissue culture and plant morphogenesis. Isr Program Sci Transl
US Dep Commerce, Springfield, VA
Cacco G, Saccomani M, Ferrari G (1977) Development of sulfate uptake capacity and
ATP-sulfurylase activity during root elongation in maize. Plant Physiol 60: 582-584
Callow ME, Coughlan SJ, Evans LV (1978) The role of Golgi bodies in polysaccharide
sulphation in Fucus zygotes. J Cell Sci 32: 337-356
Challenger F (1959) Aspects of the organic chemistry of sulfur. Academic Press, London
New York
Cormier MJ, Hori K, Karkhanis YD (1970) Studies on the bioluminescence of Renilla
reniformis. VII. Conversion of luciferin into luciferyl sulfate by luciferin sulfokinase.
Biochemistry 9: 1184-1190
Cormier MJ, Hori K, Anderson JM (1974) Bioluminescence in coelenterates. Biochim
Biophys Acta 346: 137-164
Craigie JS, McCandless E (1979) Sulfated polysaccharides in red and brown algae. Annu
Rev Plant Physiol 30:41-53
Datko AH, Mudd SH, Giovanelli J, Macnicol PK (1978) Sulfur-containing compounds
in Lemna perpusilla 6746 grown at a range of sulfate concentrations. Plant Physiol
62:629-635
DeMeio RH (1975) Sulfate activation and transfer. In: Greenberg DM (ed) Metabolism
of sulfur compounds. Academic Press, London New York
DeMeio RH, Wizerkaniuk M, Schreibman I (1955) Enzymatic system-synthesizing sul-
furic acid esters of phenols. J BioI Chern 213 : 439-443
Dodgson KS, Rose FA (1975) Sulfohydrolases. In: Greenberg DM (ed) Metabolism
of sulfur compounds. Academic Press, London New York
Dreyfuss J, Monty KJ (1963a) The biochemical characterization of cysteine-requiring
mutants of Salmonella typhimurium. J BioI Chern 238: 1019-1024
Dreyfuss J, Monty KJ (1963 b) Coincident repression of the reduction of 3' phospho aden-
osine 5' phosphosulfate, sulfite, and thiosulfate in the cysteine pathway of Salmonella
typhimurium. J BioI Chern 238:3781-3783
Ellis RJ (1963) Cysteine biosynthesis in beet discs. Phytochemistry 2:129-136
Ellis RJ (1969) Sulphate activation in higher plants. Planta 88: 34-42
Fankhauser H, Brunold C (1978) Localization of adenosine 5' phospho sulfate sulfotrans-
ferase in spinach leaves. Planta 143: 285-289
Fankhauser H, Brunold C (1979) Localization ofO-acetyl-L-serine sulfhydrylase in Spina-
cia oleracea L. Plant Sci Lett 14: 185-192
Fankhauser H, Schiff JA, Garber L (1981a) Purification and properties of adenylyl
sulfate: ammonia adenylyl transferase from Chlorella catalyzing the formation of
adenosine 5' phosphoramidate from adenosine 5' phosphosulphate and ammonia.
Biochem J 195: 545-560 '
418 J.A. SCHIFF:

Fankhauser H, Berkowitz GA, Schiff JA (1981 b) A nucleotide with the properties of

adenosine 5' phosphoramidate from Chlorella cells. Biochem Biophys Res Commun
101:524-532
Farley JR, Nakayama G, Cryns D, Segel IH (1978) ATP sulfurylase from Penicillium
chrysogenum. Equilibrium binding, substrate hydrolysis, and isotope exchange studies.
Arch Biochem Biophys 185:376-390
Friedrich JW, Schrader LE (1978) Sulfur deprivation and nitrogen metabolism in maize
seedlings. Plant Physiol 61 : 900-903
Fujimoto D, Ishimoto M (1961) Sulfate reduction in E. coli. J Biochem 50:533-537
Gerwick BC, Black CC (1979) Sulfur assimilation in C 4 plants. Intercellular compartmen-
tation of ATP sulfurylase in crabgrass leaves. Plant Physiol64:590-593
Giovanelli J, Mudd SH, Datko AH (1980) Sulfur amino acids in plants. In: Stumpf
PK, Conn EE (eds) Plant biochemistry, vol V. Academic Press, London New York
Goldschmidt EE, Tsang ML-S, Schiff JA (1975) Studies of sulfate utilization by algae.
13. Adenosine 5' phosphosulfate (APS) as an intermediate in the conversion of adeno-
sine 3' phosphate 5' phosphosulfate (PAPS) to acid volatile radioactivity. Plant Sci
Lett 4:293-299
Greenberg DM (ed) (1975) Metabolic pathways, vol VII. Metabolism of sulfur com-
pounds. Academic Press, London New York
Griffith OW, Anderson ME, Meister A (1979) Inhibition of glutathione biosynthesis
by prothionine sulfoximine (S-n-propyl homocysteine sulfoximine) a selective inhibitor
of y-glutamylcysteine synthetase. J BioI Chern 264: 1205-1210
Haines TW (1973) In: Erwin JA (ed) Lipids and biomembranes of eucaryotic microorgan-
isms. Academic Press, London New York
Harvey MJ, McLachlan J (eds) Chondrus crispus. Nova Scotian Inst Sci Halifax, Nova
Scotia
Harwood JL (1980) Sulfolipids. In: Stumpf PK, Conn EE (eds) The biochemistry of
plants, vol IV. Academic Press, London New York
Harwood JL, Nicholls RG (1979) The plant sulpholipid - a major component of the
sulphur cycle. Biochem Soc Trans 7:440-447
Hilz H, Lipmann F (1955) The enzymic activation of sulfate. Proc Nat! Acad Sci USA
41:880-890
Hodson R, Schiff JA (1971) The ubiquity of sulfate reduction to thiosulfate. Plant Physiol
47:296-299
Holmgren A, Ohlsson I, Grankvist M-L (1978) Thioredoxin from Escherichia coli. Ra-
dioimmunological and enzymatic determinations in wild-type cells and mutants defec-
tive in Phage T 7 DNA replication. J Bioi Chern 253: 430-436
Hoppe W, Schwenn JD (1981) In vitro biosynthesis of the plant sulpholipid: on the
origin of the sulphonate group. Z Naturforsch 36C: 820-826
Hoskin FCG, Kordik ER (1977) Hydrogen sulfide as a precursor for the synthesis of
isethionate in the squid giant axon. Arch Bioch Biophys 180:583-586
Ishimoto M, Fujimoto D (1959) Adenosine 5' phospho sulfate as an intermediate
in the reduction of sulfate by a sulfate-reducing bacterium. Proc Jpn Acad 35:243-
245
Jones-Mortimer MC (1968a) Positive control of sulphate reduction in Escherichia coli.
Isolation, characterization and mapping of cysteine-less mutants of E. coli K12. Bio-
chern J 110:589-595
Jones-Mortimer MC (1968b) Positive control of sulphate reduction in Escherichia coli.
Biochem J 110:597-602
Kredich N, Tomkins G (1966) The enzymic synthesis of L-cysteine in Escherichia coli
and Salmonella typhimurium. J Bioi Chern 241 :4955--4965
Leinweber FJ, Monty K (1963) The metabolism of thiosulfate in Salmonella typhimurium.
J BioI Chern 238: 3775-3780
Levinthal M, Schiff JA (1968) Studies of sulfate utilization by algae. 5. Identification
of thiosulfate as a major acid-volatile product formed by a cell-free sulfate-reducing
system from Chlorella extracts. Plant Physiol 43: 555-562
Martin WG, Truex CR, Tarka SM, Hill LJ, Groby WG (1974) The synthesis of taurine
III.1 Reduction and Other Metabolic Reactions of Sulfate 419

from sulfate. VIII. A constitutive enzyme in mammals. Proc Soc Exp BioI Med
147:563-565
Mayer AM (1967) Subcellular location of sulphite reductase in plant tissues. Plant Physiol
42:324-326
McLachlan KD (ed) (1975) Sulphur in Australasian agriculture. Sydney Univ Press,
Sydney
Moller ME, Evans LV (1976) Sulphate activation in the unicellular red alga Rhodella.
Phytochemistry 15: 1623-1626
Onajobi FD (1975) Effects of adenine nuc1eotides on rice-root adenosine triphosphate
sulfurylase activity in vitro. Biochem J 149: 301-304
Onajobi FD, Cole CV, Ross C (1973) Adenosine 5'-triphosphate-sulfurylase in corn
roots and its partial purification. Plant Physiol 52: 580-584
Pasternak CA, Ellis RJ, Jones-Mortimer MC, Crichton CE (1965) The control of sulphate
reduction in bacteria. Biochem J 96:270-275
Pate JS (1965) Roots as organs of assimilation of sulfate. Science 149:.547-548
Paul JS, Bassham JA (1978) Effects of sulfite on metabolism in isolated mesophyll cells
from Papaver somniferum. Plant Physio162:210-214
Peck HD Jr (1959) The ATP-dependent reduction of sulfate with hydrogen in extracts
of Desulfovibrio desulfuricans. Proc N atl Acad Sci USA 45: 701-708
Peck HD Jr (1961) Enzymatic basis for assimilatory and dissimilatory sulfate reduction.
J Bacteriol 82:933-939
Peck HD Jr (1962a) The role of adenosine 5' phosphosulfate in the reduction of sulfate
to sulfite by Desulfovibrio desulfuricans. J BioI Chern 237: 198--203
Peck HD Jr (1962b) Symposium on metabolism of inorganic compounds. V. Comparative
metabolism of inorganic sulfur compounds in microorganisms. Bacteriol Rev
26:67-94
Peck HD Jr (1970) Sulfur requirements and metabolism of microorganisms. In: Muth
OH, Oldfield JE (eds) Symposium: Sulfur in nutrition. Avi, Westport
Peck HD Jr, Doacon TE, Davidson JT (1965) Studies on adenosine 5'-phosphosulfate
reductase from Desulfovibrio desulfuricans and Thiobacillus thioparus. 1. The assay
and purification. Biochim Biophys Acta 96: 429-446
Percival E, McDowell RH (1967) Chemistry and enzymology of marine algal polysaccha-
rides. Academic Press, London New York
Pigiet V, Conley RR (1978) Isolation and characterization of phosphothioredoxin from
Escherichia coli. J BioI Chern 253: 1910-1920
Pitman MG (1976) Ion uptake by plant roots. Encycl Plant Physiol I1B:95-128
Plesnicar M (1977) The uptake and distribution of 35S02 in bean (Phaseolus vulgaris
L. cv. Pasuljica). Plant Sci Let 10:205-211
Porque PG, Baldesten A, Reichard P (1970) The involvement of the thioredoxin system
in the reduction of methionine sulfoxide and sulfate. J BioI Chern 245:2371-2374
Quatrano R (1978) Development of cell polarity. Annu Rev Plant PhysioI29:487-510
Ramus J (1974) In vivo molybdate inhibition of sulfate transfer to Porphyridium capsular
polysaccharide. Plant Physiol 54: 945-949
Ramus J, Groves ST (1974) Precursor-product relationships during sulfate incorporation
into Porphyridium capsular polysaccharide. Plant PhysioI54:434-439
Renosto F, Ferrari G (1975) Mechanism of sulfate transport inhibition by cycloheximide
in plant tissues. Plant Physiol 56:478-480
Reuveny Z (1977) Derepression of.ATP sulfurylase by the sulfate analogs molybdate
and selenate in cultured tobacco cells. Proc Natl Acad Sci USA 74:616-622
Reuveny Z, Filner P (1977) Regulation of adenosine triphosphate sulfurylase in cultured
tobacco cells. J BioI Chern 252: 1858-1864
Robbins PW, Lipmann F (1957) Isolation and identification of active sulfate. J BioI
Chern 229:837-851
Robbins PW, Lipmann F (1958) Separation of the two enzymatic phases in active sulfate
synthesis. J BioI Chern 233: 681-685
Roy AB, Trudinger PA (1970) The biochemistry of inorganic compounds of sulfur.
Cambridge Univ Press '
420 J.A. SCHIFF:

Schiff JA (1963) The metabolism of sulfur in algae. In: Lewin R (ed) Biochemistry
and physiology of the algae. Academic Press, London New York
Schiff JA (1972) The development, inheritance and origin of the plastid in Euglena.
In: Advances in morphogenesis, vol X. Academic Press, London New York
Schiff JA, Hodson RC (1973) The metabolism of sulfate. Annu Rev Plant Physiol
24:381-414
Schiff JA, Levinthal M (1968) Studies of sulfate utilization by algae. 5. Identification
of thiosulfate as a major acid-volatile product formed by a cell-free sulfate-reducing
system from Chlorella extracts. Plant PhysioI43:555-562
Schiff JA, Lymann H (eds) (1982) On the origins of chloroplasts. Elsevier, Amsterdam
Schmidt A (1973) Sulfate reduction in a cell-free system of Chlorella. The ferredoxin-
dependent reduction of a protein-bound intermediate by a thiosulfonate reductase.
Arch MikrobioI93:29-52
Schmidt A (1975a) Distribution of APS sulfotransferase activity among higher plants.
Plant Sci Lett 5:407-415
Schmidt A (1975b) A sulfotransferase from spinach leaves using adenosine 5' phospho sul-
fate. Planta 124:267-275
Schmidt A (1976) Development of the adenosine 5' phosphosulfate sulfotransferase in
sunflower (Helianthus annuus L.). Z Pflanzenphysiol 78: 164-168
Schmidt A (1977) Assimilatory sulfate reduction via 3' phosphoadenosine 5' phospho sul-
fate (PAPS) and adenosine 5' phosphosulfate (APS) in blue-green algae. FEMS Micro-
bioI Lett 1:137-140
Schmidt A (1982) Assimilation of sulfur. In: Schiff JA, Lyman H (eds) On the origins
of chloroplasts. Elsevier, Amsterdam
Schmidt A, Schwenn JD (1971) On the mechanism of photosynthetic sulfate reduction.
Proc 2nd Int Congr Photosynthesis. Junk, The Hague
Schmidt A, Abrams WR, Schiff JA (1974) Studies of sulfate utilization by algae. 12.
Reduction of adenosine 5' phosphosulfate to cysteine in extracts from Chlorella and
mutants blocked for sulfate reduction. Eur J Biochem 47:423-434
Schwenn JD, Trebst A (1976) Photosynthetic sulfate reduction by chloroplasts. In: Barber
J (ed) The intact chloroplast. Elsevier, Amsterdam
Siegel LM (1975) Biochemistry of the sulfur cycle. In: Greenberg DM (ed) Metabolism
of sulfur compounds. Academic Press, London New York
Siegel LM, Davis PS (1974) Reduced nicotinamide adenine dinucleotide phosphate-sulfite
reductase of enterobacteria. IV. The Escherichia coli hemoflavoprotein: subunitstruc-
ture and dissociation into hemoprotein and flavoprotein components. J BioI Chern
249: 1587-1598
Siegel LM, Murphy MJ, Kamin H (1973) Reduced nicotinamide adenine dinucleotide
phosphate-sulfite reductase of enterobacteria. J BioI Chern 248: 251-264
Siegel LM, Davis PS, Kamin H (1974) Reduced nicotinamide adenine dinucleotide phos-
phate-sulfite reductase of enterobacteria. III. The Escherichia coli hemoflavoprotein:
catalytic parameters and the sequence of electron flow. J BioI Chern 249: 1572-1586
Silvius JE, Baer CH, Dodrill S, Patrick H (1976) Photoreduction of sulfur dioxide by
spinach leaves and isolated spinach chloroplasts. Plant Physiol 57: 799-801
Singer TP (1975) Oxidative metabolism of cysteine and cystine in animal tissues. In:
Greenberg DM (ed) Metabolism of sulfur compounds. Academic Press, London New
York
Smith IK (1975) Sulfate transport in cultured tobacco cells. Plant PhysioI55:303-307
Smith IK (1976) Characterization of sulfate transport in cultured tobacco cells. Plant
Physiol 58: 358-362
Steber J, Schleifer KH (1975) Halococcus morrhuae: A sulfated heteropolysaccharide
as the structural component of the bacterial cell wall. Arch Microbiol1 05: 173-177
Tamura G, Iwasawa T, Masada M, Fukushima K (1976) Some properties of cysteine
synthase from radish roots. Agric BioI Chern 40:637-638
Thomas E, Davey MR (1975) From single cells to plants. Springer, Berlin Heidelberg
New York
III.1 Reduction and Other Metabolic Reactions of Sulfate 421

Thompson JF, Smith IK, Moore DP (1970) Sulfur requirement and metabolism in plants.
In: Muth OH, Oldfield JE (eds) Symposium: Sulfur in nutrition. Avi, Westport
Torii K, Bandurski RS (1967) Yeast sulfate-reducing system. III. An intermediate in
the reduction of 3' phosphoryl 5' adenosine phosphosulfate to sulfite. Biochim
Biophys Acta 136:286-295
Tsang ML-S (1981) Assimilatory sulfate reduction in Escherichia coli: Identification of
the alternate cofactor for adenosine 3' phosphate 5' phosphosulfate reductase as glu-
taredoxin. J BacterioI146:1059-1066
Tsang ML-S, Schiff JA (1975) Studies of sulfate utilization by algae. 14. Distribution
of adenosine 5' phosphosulfate (APS) and adenosine 3' phosphate 5' phosphosulfate
(PAPS) sulfotransferases in assimilatory sulfate reducers. Plant Sci Lett 4: 301-307
Tsang ML-S, Schiff JA (1976a) Studies of sulfate utilization by algae. 17. Reactions
of the adenosine 5' phosphosulfate (APS) sulfotransferase from Chlorella and studies
of model reactions which explain the diversity of side products with thiols. Plant
Cell Physiol17: 1209-1220
Tsang ML-S, Schiff JA (1976b) Studies of a sulfate-reducing pathway in Escherichia
coli involving bound intermediates. J Bacteriol 125: 923-933
Tsang ML-S, Schiff JA (1976c) Properties of enzyme fraction A from Chlorella and
copurification of 3' (2'), 5' bisphosphonucleoside 3' (2') phosphohydrolase, adenosine
5' phosphosulfate sulfohydrolase, adenosine 5' phosphosulfate cyclase activities. Eur
J Biochem 65: 113-121
Tsang ML-S, Schiff JA (1978a) Studies of sulfate utilization by algae. 18. Identification
of glutathione as a physiological carrier in assimilatory sulfate reduction by Chlorella
extracts. Plant Sci Lett 11 : 177-183
Tsang ML-S, Schiff JA (1978b) Assimilatory sulfate reduction in a mutant of Escherichia
coli lacking thioredoxin activity. J Bacteriol134: 131-138
Umbarger E (1978) Amino acid biosynthesis and its regulation. Annu Rev Biochem
47:533-606
White PR (1963) The cultivation of animal and plant cells. Ronald, New York
White PR, Grove AR (eds)'(1965) In: Proc Int Conf Plant Tissue Cult. McCutchan,
Berkeley
Wilson LG, Bierer D (1976) The formation of exchangeable sulphite from adenosine
3' phosphate 5' sulphatophosphate in yeast. Biochem J 158:255-270
Wilson LG, Asahi T, Bandurski RS (1961) Yeast sulfate-reducing system. 1. Reduction
of sulfate to sulfite. J BioI Chern 236: 1822-1829
Wilson LG, Bressan RA, Filner P (1978) Light-dependent emission of hydrogen sulfide
from plants. Plant Physiol61: 184-189
Ziegler I (1977) Subcellular distribution of 35S-sulfur in spinach leaves after application
of35S0~-, and 35S02. Planta 135:25-32
Ziegler I, Hampp R (1977) Control of 35SOi- and 35S0~- incorporation into spinach
chloroplasts during photosynthetic CO 2 fixation. Planta 137: 303-307
111.2 Physiology and Metabolism of Phosphate
and Its Compounds
R.L. BmLESKI and LB. FERGUSON

1 Introduction

The distribution of phosphorus 1 in the world, unlike that of all other elements
but carbon, is dominated by the present and past activities of living organisms.
Thus it was first isolated as an element from that preeminently biological fluid,
urine, by the Arabian alchemists in the 12th century and then by H. BRAND
in 1669; while the next source to be discovered was bone, in 1770 (CORBRIDGE
1978). It is widely distributed in the Earth's crust, where it comprises 0.1 %
by weight of the elements present. Igneous deposits are known, but most of
the phosphate used by man has been formed either as guano and its end-product,
phosphatized coral, or as sedimentary deposits laid down under marine condi-
tions in a combination of biological and physicochemical processes. In each
case, the key event in formation of the deposit has been the ability of living
organisms to scavenge phosphate from their surroundings, so that the concentra-
tion within the organism is increased one thousand fold or more (see Sect. 2).
With guano-based products, phosphate has passed through a long food chain
(marine microorganism - crustacean - fish - sea bird) and has finally been
drawn into one place as excreta and as fish and bird remains. With sedimentary
rocks, the process has been more direct, in that the microorganisms themselves
settle to the bottom of the sea where, as they die, they create a zone in which
the phosphate ion concentration is many hundreds of times higher than in
the surrounding waters. If the combination of pH, temperature, and concentra-
tion of Ca2+ and F- is right, the solubility product of various calcium phos-
phates is exceeded, and precipitates form. Alternatively, pre-existing calcium
carbonate deposits in the sea bed, arising from shells of marine organisms,
are progressively phosphatized, a process wherein the phosphate ion being
brought in by the microorganisms replaces the carbonate one of the shell. Typi-
cally, these events have taken place in the shallow waters of the eastern seaboards
of continental masses, and the subsequent uplifting of the seabeds has given
us our phosphate deposits. Now that the more accessible sources of phosphate
are disappearing, geologists are considering whether it might be economically
feasible to mine phosphate deposits directly from the sea bed in their formative
state as phosphate nodules (CORBRIDGE 1978).
Because it is part of the DNA and RNA molecule, P is part of every living
thing. In typical plant tissues it is present at about 0.04% of the fresh weight
1 The following abbreviations will be used throughout. P, phosphorus; Pi, inorganic
phosphate; P-lipid, phospholipid; P-esters, acid-soluble and water-soluble simple phos-
phate esters; RNA, ribonucleic acid; DNA, deoxyribonucleic acid; G6P, glucose-6-
phosphate; 3PGA, 3-phosphoglyceric acid. Standard nucleotide abbreviations are used
III.2 Physiology and Metabolism of Phosphate and Its Compounds 423

or 0.3% of the dry weight. If we exclude carbohydrate, however, P forms about

4% of the nutrient content, with N, K, Ca, Mg, P and S being present in
the molar ratio 20: 5: 2.5: 2: 1 : 0.5 (EpSTEIN 1972). At values below 1/3 and above
three times those quoted above, P deficiency and P toxicity respectively may
be operating. Usually, somewhere in excess of50% of the P present is ininorgan-
ic form, so that the Pi concentration of living plant tissues is typically around
5-20 mM (see Sect. 4).
With only very minor exceptions, all P in both geological and biological
systems is in the fully oxidized form; geologically, mainly as the various insoluble
calcium salts, and biologically as the free Pi ion and the various phosphate
esters (Sect. 4). Soils represent something of a half-way house in that, while
P coming from the breakdown of rock particles is inorganic (and held at very
low concentration by the presence of Fe, Al and Ca ions) much of the phosphate
deposited in plant remains can stay in its original organic form, or can be
rapidly taken up by soil microorganisms and converted to other organic forms.
Thus, soils high in humus can often contain over half their P in organic form
and as a consequence, those processes which lead to the recycling of organic
P in the soil can be of major importance to plants (COLE et aL 1978, GOULD
et aL 1979). Whether in inorganic or organic form, P is very easily locked up
and made unavailable to the plant, and so P deficiency is a common phenome-
non. Availability of P can be the single most important factor limiting plant
growth: for example, life in the world's seas is dominated by the upwelling
of phosphate-rich waters from the ocean deeps, particularly in Antarctic regions,
leading to massive phytoplankton blooms and the forging of the first links
in the food chain (HARVEY 1969).
Man causes similar processes to occur when he discards phosphate-rich mate-
rials such as detergents, sewage and processing wastes into inland waters. Again,
by supplying P (and/or nitrogen) we remove the existing limit, and the resulting
burst of plant growth is what we see as eutrophication and pollution. In general,
very little P is lost from the soil by leaching because phosphate is strongly
bound. Thus in a moderate rainstorm of 34 mm spread over 3 days, only
0.03 g P ha - 1 was removed from a cropping catchment, corresponding to an
annual loss of around 1 g ha - 1 year - 1. In a more intense storm of 63 mm
over 12 h, the loss was much greater, 0.8 g P ha -1 (corresponding to
12 g P ha -1 year- 1 ), about half being sorbed to fine particles of sediment
(KUNISHI et aL 1972). Thus the rate of loss is greatly increased when the soil
particles are lost, particularly from the upper soil layers where P tends to accu-
mulate (HEALY and MCCOLL 1974, MCCOLL et aL 1975). Severe erosion by
wind and water, as with loess soils in China and dust-bowl soils in the USA,
has caused P losses estimated at more than 20 kg P ha -1 year- 1 • Another factor
which can greatly increase rate of P loss is the application of P fertilizer:, under
particularly severe conditions of a 76 mm rainfall immediately following fertiliz-
er application, about a quarter of the applied P was lost (GILCHRIST and GIL-
LINGHAM 1970). A high level of animal activity, with much of the P excreted
in organic or bound form, can also lead to a higher rate of loss, estimated
at 0.5-2 kg P ha -1 year- 1 for stable high-yielding pasture/herd systems (COOKE
1981). Thus, P loss is extremely sensitive to land management.
424 R.L. BIELESKI and I.B. FERGUSON:

All these factors combine to make phosphate one of the key nutrients in
horticultural and agricultural management, and fertilizer use accounts for more
than 90% of total P use in the world. Countries with a highly-developed agricul-
ture use an average of 20-50 kg arable ha -1 year- 1 , developing countries as
little as a tenth of this level or in population terms, less than 1 kg head - 1 year - 1.
Apart from the requirements of crops and stock, man himself excretes some
0.4 kg P head -1 year- 1 , and this must be replaced by P in food.

2 Uptake and Transport of Phosphate

More than for any other mineral nutrient, the uptake of phosphate has to
be a metabolically driven process. This is apparent if we compare the nutrient
compositions of a typical plant tissue, its xylem sap, and the soil solution bathing
the roots (Table 1). If there were no accumulation, the xylem sap would have
the same concentration and composition as the soil water, whilst the tissue
would have the elements in the same ratio but at about 50-150 times higher
concentration (because of the concentrati9n of nutrients due to water loss
through transpiration). In practice, most of the elements show the observed
pattern but phosphate does not: its concentration in the xylem sap is much
higher than would be expected (Table 1). In the step from the soil solution
to the xylem sap, most elements do not increase their concentration markedly,
so that their uptake could be passive, but phosphate increases its concentration
400-fold both with respect to that in the soil solution and in relation to the
other elements. The phenomenon is also shown in model systems: the phosphate
concentration of xylem sap exuding from excised roots is at least two orders
of magnitude higher than in the surrounding nutrient solution (COLLINS and
REILLY 1968). More detailed experiments show that the root cortical cells are

Table 1. Comparison of nutrient concentrations in plant tissue, xylem sap and soil solu-
tion. Tissue nutrient concentrations (T, limol g-l fresh wt.) are given for the kiwifruit
vine (fruit plus leaf) (FERGUSON and EISEMAN 1983), and resemble those of many plant
tissues (EpSTEIN 1972). Xylem sap concentrations (X, limol ml- 1 ) are those obtained
for sap of kiwifruit vines, averaged over the main growing period (Dec-Apr, see FERGUSON
et al. 1983). Soil solution concentrations (S, limol ml- I) are those found for a range
of about 250 soils (BARBER et al. 1963, REISENAUER 1966)

Element Tissue conc. Sap conc. Ratio Soil conc. Ratio

(T) (X) (T/X) (S) (XIS)

Na 170 2.7 63 1-3 1.3

K 80 2.8 29 1-2 1.8
Ca 28 1.1 25 0.5-1.5 1.1
Mg 14 0.6 23 2-4 0.2
S 0.2 0.3-0.7 0.4
P 9 0.4 23 0.0005-0.002 400

a N in xylem sap is the sum of nitrate plus arginine plus amino acid
111.2 Physiology and Metabolism of Phosphate and Its Compounds 425

the major site of this phosphate accumulation step, and that phosphate is then
transported across the root in the symplast (FERGUSON and CLARKSON 1975).
Within most plant cells, the Pi concentration is 1-10 mM (BmLEsKl 1973) so
that the final gradient is as much as 10,000-fold. If we finally take into account
the negative membrane potential of the cell (in the region of -120 mV in
clover roots; BOWLING and DUNLOP 1978), then the effective gradient to be
overcome is about 106 -fold (BmLESKl 1976).
It is clear from the work of ASHER and EDWARDS (Chap. 1.3, this Vo1.)
that the soil water measurements used in Table 1 represent a realistic situation,
and that optimum plant growth is obtained at very low external phosphate
concentrations (3-13 J.1M) despite the high concentrations in the cell. Epiphytic
plants apparently encounter a similar situation, as the P concentration of rain-
water is in the range 0.7-3 J.1M (GROVES 1981). Thus it is not surprising that
phosphate uptake shows to a marked degree the features to be expected of
a metabolically driven accumulation process. It is very sensitive to temperature,
to anoxia, to metabolic inhibitors (SMITH 1966, RAVEN 1974b, FERGUSON and
CLARKSON 1975) and to competitive inhibition by the very similar anion, arse-
nate (JUNG and ROTHSTEIN 1965). A close relationship between the cell PD
and phosphate uptake suggests the operation of some type of electrogenic pump
(BoWLING and DUNLOP 1978); and recent work with fungi (BEEVER and BURNS
1980) and Lemna (ULLRICH-EBERIUS et a1. 1981) points to a proton extrusion
pump of the type envisaged by Mitchell.
There has been some consideration given to the ionic form entering the
plant. At normal soil pH, the H 2 PO';:- ion is predominant (Chap. 1.6, this Vo1.),
and there is a general agreement that this can be transported. However, at
higher pH's, still within the biological range, there is almost no H 2 PO';:-, and
HPO~- is the major ion, yet Pi is still taken up. One view is that there is
a separate uptake mechanism for the divalent anion (HAGEN and HOPKINS 1955).
A factor which tends to be overlooked in such discussions is that the pH in
the soil solution need not be anything like that existing at the uptake site itself.
The uptake system, by its very operation, must have a powerful effect on the
effective pH nearby. If, for example, Pi or any other anion is taken up without
the corresponding transport of an exactly balancing amount of cation, then
the pH at the uptake site must change. Indeed, if current theories on uptake
of Pi prove correct, an outward proton pump must lower the pH at the outer
surface of the membrane at the same time as it creates a potential difference;
Pi will actually enter as a positively charged complex (ULLRICH-EBERIUS et a1.
1981, BEEVER and BURNS 1980). The question has also been raised as to whether
it is energetically possible for Pi to overcome the major concentration and
electrical gradient if in the divalent form (ULLRICH-EBERIUS et a1. 1981). For
these reasons, the H 2 PO';:- /HPO~- ratio in the bulk solution is likely to have
limited significance. This is reflected in practice, as Pi uptake is often not mark-
edly influenced by pH over a range that drastically alters the H 2 PO';:- /HPO~­
ratio, and the pH relationship that occurs can never be simply interpreted by
having a HPO~- uptake mechanism operating additively.
A feature of Pi uptake systems is that the rate of uptake is seldom a simple
function of concentration over the full concentration range (10- 7-10- 2 M).
426 R.L. BIELESKI and I.B. FERGUSON:

Surprisingly often, the complex uptake kinetics that are observed can be satisfac-
torily interpreted in terms of the simultaneous presence and activity of two
systems showing Michaelis-Menten kinetics (uptake a rectangular hyperbolic
function of concentration), one having a high affinity (Km around 5 11M) and
one of low affinity (Km around 500 11M) (LEGGETT et al. 1965, CARTER and
LATHWELL 1967). This pattern, often called the "dual isotherm" or "double
hyperbola" or "two-phase kinetics", has been variously interpreted and is
beyond the scope of this chapter; but the most detailed study to date merits
separate attention as it brings out the importance to the plant of maintaining
a constant Pi intake in the face of a varying external concentration (BURNS
and BEEVER 1977, BEEVER and BURNS 1977). Neurospora spore germlings sup-
plied with Pi in the concentration range 1-3,200 11M have an uptake rate/concen-
tration behaviour that fits a clearly defined double hyperbola pattern (Fig. 1).
Depending on the prior history of the germlings - that is, the Pi concentration
in which they were grown prior to the actual uptake experiments - the kinetic
constants differed, but the overall double-hyperbola pattern was still maintained.
As the Pi concentration used to grow the germlings was increased from 50 JlM
to 10 mM, the affinity of the high affinity system remained fixed (Km about
2.5 11M) but the affinity of the low affinity system fell by a factor of 3 (Km
rising from 350 to 1,000 11M); the capacity .of the high affinity system decreased
8 x 01 max from 2.4 to 0.3 Ilmol g-l dry wt. min -1) and the capacity of the low
affinity system decreased 2 x 01 max from 6.8 to 3.4) (Fig. 1 B). The effect of
these changes was to adjust the total uptake capacity of the cell in such a
way that regardless of the growth conditions (Pi concentration range 50 IlM-
10 mM), the Pi uptake rate at that concentration was fixed at about 3.3 Ilmol
g - 1 dry wt. min - 1 (Fig. 1 B). That is, by relatively small modulations to two
uptake systems, the plant achieved a constant Pi supply regardless of the concen-
tration in the environment. Very much larger modulations would have been
needed if only a single Michaelis-Menten uptake system had been available
to the cell. It remains to be seen whether this explanation for two-phase kinetics
will also apply to higher plant tissues.
In a macroscopic land plant, uptake is only the first stage in utilizing P
from the environment. Accumulative uptake of Pi by the root cortical cells
must then be followed by transfer across the root to the xylem. This occurs
through the symplast; thus a barrier to apoplastic transport such as endodermal
suberization has little effect on Pi transport, whereas a barrier to cortical cell
uptake such as hypodermal suberization reduces it (FERGUSON and CLARKSON
1975, 1976). The vacuoles of the root cells are seen as being to one side of
the main pathway (GREENWAY and KLEPPER 1968).
Although a case has been made for a diffusive release of ions into the
xylem vessels (DUNLOP and BOWLING 1971) most current evidence supports
a release which is energy-dependent (PITMAN 1977), allowing a greater control
by xylem parenchyma cells of ion levels in the xylem fluid. There is certainly
a need for some secreting or accumulating mechanism in the xylem. Under
normal transpiration conditions with a steady xylem flow, the xylem Pi concen-
tration is about 2%-5% of that in the root cells. However, under conditions
of xylem sap immobility such as in a deciduous plant just before bud break,
the sap Pi concentration can rise to levels as high or higher than in the root
 8

c 6
'E
.....
~
.,;
~
'0
~4

200 400 600 800

UPTAKE RATE ( Jlmol/g d.w./min )
[j»HOSPHATE] ( mM )

Vmax values
1200
[[] Km values c
's
......
ii! 6
900 .,;
'"
600
:E
...
..
:0.
300
E
---------- -----------...-----~---------- ------ ---

:
... ...... '-total P uptake

~ .----.--------.--------------~ l;
.
t"". I
10-4
! !" ",I
10 -3
! " " ..
10-2
I
>
"E
10-4 10-2
PHOSPHATE IN GROWTH MEDIUM (M)

Fig. lA, B. Dual phosphate uptake mechanisms in Neurospora. (BEEVER and BURNS 1980).
A The Hofstee plot of the relationship between Pi concentration and uptake rate for
germlings previously grown in low Pi concentration, 50 J-lM. The data were analyzed
assuming simultaneous operation of two uptake systems, one of high affinity (H) and
one of low affinity (L). The three lines (which are the calculated lines based on the
kinetic constants) give the uptake due to the two systems operating individually (dotted
lines) and operating together (solid line). Note the goodness of fit of the model curve
to the experimental data. B Adaptive changes in the kinetic constants brought about
by growing the germlings in a range of Pi concentrations (50-10,000 J-lM). Left Km values;
right Vmax values; top and -.-low affinity system; bottom and -.- high affinity system.
The total calculated Pi uptake that would result at each Pi concentration from the com-
bined operation of both uptake mechanisms (each having the kinetic constants character-
istic of that concentration) is shown by -A-. Note that it is almost constant
428 R.L. BIELESKI and I.B. FERGUSON:

cells themselves (15 mM - FERGUSON et al. 1983). This property is shared by

other xylem tissues: excised shoot segments of a dormant plant, induced to
leaf in water in a warm room, can increase the Pi concentration in the xylem
fluid from 0.4 to 3.4 mM in the absence of roots (FERGUSON et al. 1983). Such
observations strongly suggest an energy-dependent secretion, although direct
evidence for this process is still lacking.
P transferred to the xylem stream is mostly or completely present as Pi,
and the occasional presence of phosphatyl choline still requires explanation
(BmLEsKI 1973). Once there, Pi is swept to all parts of the plant. Nor does
it stop moving when it reaches a leaf or growing point because P is, with
N, the most readily retranslocated nutrient (Chap. 1.6, this Vol.). Thus if Pi
is supplied to the leaf surface, or when the leaf ages, up to 60% of the P
can be retranslocated through the phloem to other plant parts, particularly
the growing points and developing fruits (BARRmR and LOOMIS 1957, BIDDULPH
et al. 1958, GREENWAY and GUNN 1966). When P enters the sieve tubes it is
rapidly metabolized to a wide range of P-esters, particularly A TP; but Pi is
also present, and it has been suggested (BmLESKI 1969, KLUGE et al. 1970) that
this is the form being transported in the mobile phase, with the P-esters being
associated with the metabolically active stationary cytoplasmic phase.
Thus uptake into the cell, secretion and transport in the xylem, and retranslo-
cation within the phloem, all seem to occur with P in the form of Pi, and
this seems to be the normal transport form. The only place where P-ester trans-
port is likely to be important is between organelles and the cytoplasm.

3 Efflux of Phosphate, and Aspects of Phosphate Deficiency

The uptake processes discussed in Section 2 create a steep Pi concentration

gradient between the inside of the living cell and the external medium. In the
cells near the root surface, where the apoplast is in direct diffusional contact
with the soil solution, the plasmalemma is the only barrier to the strong physical
forces tending to drive Pi out of the cell. Given this situation, it is to be expected
that there would be some actual outward movement or efflux occurring even
while there is a net uptake taking place. Whilst Pi efflux from plant roots
has been recognized for some time (HEVESY 1946, EMMERT 1959, WEIGL 1968,
MAZEL and FOKIN 1977) detailed study has been lacking until recently. In the
water plants Spirodela, Lemna and Azolla (MCPHARLIN 1981, BmLEsKI unpub-
lished data), P efflux occurs at a significant rate and shows the following fea-
tures. The rate is increased by raising the external Pi concentration (Table 2A),
by the presence of metabolic inhibitors such as CCCP, azide and cyanide, and
by lowering the temperature (Table 2B). When the plant starts to become P-
deficient, there is a marked decrease in efflux rate, accompanying the expected
increase in uptake rate (Table 2A). The P-containing material lost to the medium
as a consequence of efflux is entirely in the form of Pi.
III.2 Physiology and Metabolism of Phosphate and Its Compounds 429

Table 2. Effect of various conditions on Pi effiux and uptake in Spirodela (rates in

nmol g-l fresh wt. h- 1)

Control plants -P plants

Uptake Effiux Net change Uptake Effiux Net change

A. External Pi cone. (11M)

1000 500 40 460 1200 16 1184
100 220 29 191 950 12 938
10 100 24 76 280 10 270
1 20 .14 6 40 4 36
B. Temperature
25° 460 38 422
15° 180 58 122
5° 80 84 -4

The effect of external Pi concentration (A) was measured at 25°C; the effect of tempera-
ture (B) was measured at 1,000 11M external Pi concentration .." - P plants" are in inci-
pient deficiency, having been 4 days without external Pi, leading to a 30% reduced
growth rate but no marked visual symptoms of deficiency. (Data from MCPHARLIN
1981)

Most ofthe responses are opposite to those obtained with the uptake process,
but not all: for example, increased external Pi concentration increases both
uptake and efflux. The nature of the responses is consistent with at least part
of the efflux being under metabolic control, as would occur, for example, if
metabolism was needed to maintain the physical impermeability of the cell
membrane. The actual rates of loss measured have been in the range 4-50 nmol
g - 1 fresh wt.h - 1, which can be compared with uptake rates in the range
20--1,000 nmol g-l fresh wt.h- 1 .
We can compare the measured uptake and efflux rates with the change
in total P content of exponentially-growing Spirodela plants. Under exponential
growth, the instantaneous net P uptake rate U (nmolg- 1 fresh wt.h- 1 ) can
be calculated from U = C ln2/t where C is the P content of the tissue (in nmol
g-l fresh wt.), and t is the doubling time (h). For plants such as those described
in Table 2, growing under optimal conditions in medium containing 1 mM Pi,
t=53 hand C=37,000 nmol g-l fresh wt., so U =484 nmol g-l fresh wt.h- 1 .
The close agreement with the measured net uptake rate, 460 (Table 2) indicates
that the experimentally measured net phosphate uptake rates do have meaning
in terms of the growing plant.
In such optimally growing plants, the efflux rate is 8% of the uptake rate.
But if such plants are suddenly put into less favourable conditions (10wered
external Pi concentration, lowered temperature), efflux can equal or even exceed
uptake, leading to net loss of P from the plant. In contrast, a plant that has
been grown for a short time under P-deficient conditions, upon being returned
to optimal growing conditions, has a decreased efflux but is able to take up
Pi about 2.5 times faster than is required to maintain optimal growth, and
430 R.L. BIELESKI and I.B. FERGUSON:

so is able to restore its P status within about 12 h. Because P effiux remains

much lower than in the normal plant, the effiux rate is now only 1.3% of
the uptake rate. Thus when a plant responds to P deficiency, it not only increases
P uptake, it also decreases P loss, and the second effect is not insignificant.
The importance to a plant of maintaining a constant P intake over a wide
range of external P supply conditions has already been discussed (Sect. 2).
Equally as important is the ability of a plant to adjust the P intake to the
rate of utilization (CHAPIN and BIELESKI 1982). At a doubling time of 2.2 days,
Spirodela needs a net intake of P at 480 nmol g-l fresh wt.h -1. If a change
in conditions (e.g. low temperature, poor light, reduced N supply) slows growth
so that the doubling time is now 6.6 days, the net P intake must fall to
160 nmol g -1 fresh wt.h -1 if the P content of the tissue is to remain the same.
In the example quoted, an unchanged P intake would lead to a tissue P content
of 100 nmol g-l fresh wt. - about 2% of dry weight, and well into the region
of severe toxicity. If the plant is able to actively modulate the P loss through
the effiux mechanism as well as the P uptake itself, a much more effective
total control of net P intake would be achieved.
We believe that future research will confirm that Pi effiux is part of the
mechanism for maintaining the P balance of the plant. As yet, the signal the
plant uses in responding to deficiency and excess of P has not been identified,
but on present evidence it is most likely to be the Pi concentration in the
cytoplasm. For both Neurospora (BURNS and BEEVER 1979) and Spirodela and
Lemna (MCPHARLIN 1981), a depletion of cellular Pi is the fastest and largest
response to P deficiency conditions that has been detected in the various P
compounds. Much of the cell Pi in green plants is in the vacuole, and passage
of Pi from vacuole to cytoplasm may be slow and restricted (BIELESKI 1968b,
1973). The depletion of Pi arising out of P deficiency would be expected to
take effect in the cytoplasm, where Pi is being withdrawn for growth: thus
an 8% depletion of total P will represent a 12% depletion of cellular Pi, but
an 80% depletion of cytoplasmic Pi. These are the sorts of conditions that
obtain in Spirodela when, after 12 h of P deficiency, the capacity of the Pi
uptake mechanism has already increased by about 50% (MCPHARLIN 1981).
One final point emerges from these effiux experiments. It is apparent from
the data in Table 2 that lowering the Pi concentration decreases Pi uptake much
more markedly than Pi effiux. This means that eventually an equilibrium concen-
tration must be reached where effiux and uptake are equal: below this concentra-
tion, there must be a net loss of P from the plant.
The equilibrium point will not be immutable but will partly depend on
other factors (for example, in Spirodela it is below 1 IlM P at 25°C but approxi-
mately 1,000 IlM Pi at 5°C; see Table 2B). We can find the value of this equilib-
rium concentration either by separately measuring uptake and e£flux rates at
different Pi concentrations as in Table 2A, and finding where the uptake and
effiux lines cross; or by putting plants into a Pi-containing solution and allowing
uptake and effiux to proceed until no further concentration change occurs in
the medium, then measuring the concentration. Or, finally, we can supply Pi
at different concentrations to plants, and find the minimum concentration that
will support growth (AsHER and LONERAGAN1967). The limited information
III.2 Physiology ,and Metabolism of Phosphate and Its Compounds 431

to date from all these procedures suggests that for higher plants, the equilibrium
point is around 0.3 J.1M Pi for control plants; but that in P deficient plants,
the combination of increased uptake with decreased efflux lowers the equilibri-
um point to about 0.07 J.1M Pi.
The ecological implications of this are of interest, particularly in light of
the possibility that temperature, with its opposite effect on uptake and efflux,
will have a powerful influence on the equilibrium concentration. As they stand,
the results suggest that P deficiency will occur much more readily at suboptimal
temperatures.

4 Phosphorus Compartments and Pools

We can use the terms "pool" and "compartment" to describe the situation
where different parts of the cell are physically isolated from each other and
behaving in somewhat different ways, and also to describe the situation where
different portions of a single compound are metabolically isolated from one
another, whether or not this is due to their actual physical separation in the
organelles. The situation is of particular importance with the P-esters. Metabolic
compartmentation shows up in the following ways. (a) When tissues supplied
with 32Pi are returned to non-radioactive medium, the rate of 32p reappearance
in the medium can show two or more distinct phases, due to differences in
the rates of movement from different compartments. The most obvious is the
distinction between the cell wall (apoplast) and the rest of the cell inside the
plasmalemma (symplast) (MACRoBBIE 1971). (b) When subcellular fractions are
isolated and analyzed we detect differences in P-ester composition and concen-
tration. Thus phosphatidyl glycerol is largely restricted to the chloroplast (BIE-
LESKI 1973, see Sect. S) and RNA is absent from the amyloplast (Lm and
SHANNON 1981). (c) By comparing the rates of movement of various solutes
into isolated organelles, we can uncover selective permeability behaviour in
the membrane. Thus the chloroplast membrane is readily permeable to 3-PGA,
but not to ribulose 1,S-diphosphate (JOHNSON and BRUFF 1967, BASSHAM et al.
1968). (d) There can be radioactive labelling anomalies with unequal labelling
in different parts of the cell, or unexpected patterns of specific activity in differ-
ent compounds. Thus when 32Pi is supplied for a short time to a root, the
specific activity of Pi recovered in the xylem sap is higher than that in the
root tissues, showing that the 32Pi entering the root did not mix with all of
the Pi already present in the root tissue (GREENWAY and KLEPPER 1968). Within
the tissue itself, the P-esters have a specific activity about 10 times higher than
the Pi extracted from the tissue (BIELESKI 1968b, WEIGL 1963, LOUGHMAN"1960).
Again, the anomalous labelling suggests that much of the Pi within the tissue
does not equilibrate readily with 32Pi that is entering, and does not contribute
to P-ester synthesis or to xylem transport. This inactive portion has been called
the "non-metabolic pool" and is estimated to contain 8S%-9S% of the total
Pi present (BIELESKI 1973). The small Pi pool, S%-lS%, which receives the
entering 32Pi and which contributes to P-ester synthesis is termed the "metabolic
432 R.L. BIELESKI and I.B. FERGUSON:

Fig. 2. Changes in cytoplasmic and

vacuolar Pi concentration in the cell
of Hydrodictyon. (From data of
RAVEN 1974a)

cytoplasm ~
~---~---

pool". Although the two pools could conceivably be in different types of cell
(a "storage" cell and an "ester-producing" cell), the accepted view is that
the non-metabolic pool represents Pi in the vacuole, and the metabolic pool
is the Pi in the cytoplasm plus its organelles. Thus when ULLRICH et al. (1965)
pulse-labelled leaves with 32Pi then isolated various subcellular fractions, 75%
of the Pi was found to be in the vacuole, and very slow to label; 15% was
in the cytoplasm, and rapidly labelled; and 10% was in the chloroplast, rapidly
labelled. Recently, high-resolution NMR studies have shown that 90% of the
Pi in living maize root tips is at a pH characteristic of the vacuole, while the
remaining 10% of Pi, along with all the ATP and glucose-6-phosphate, is at
a much higher pH characteristic of the cytoplasm (LOUGHMAN personal commu-
nication). Finally, in an elegant experiment with the giant alga Hydrodictyon,
RAVEN (1974a) has separately analysed samples of the vacuole and cytoplasm,
and shown that the P-esters are confined to the cytoplasm; that about 10%
of the Pi is in the cytoplasm and 90% in the vacuole; and that when 32Pi
is supplied to the alga, it labels the cytoplasmic Pi and P-esters long before
the vacuolar Pi becomes significantly labelled (Fig. 2).
Such compartmentation is to be expected. The various P-esters are centrally
involved in many biochemical pathways, and often control the activities of
those paths. The same P-ester can take part in an anabolic pathway in one
part of the cell and a catabolic pathway in another (ROWAN 1966). The need
for separate regiIlation of the paths creates the need for separate pools. It
is of some interest that P-deficient Spirodela plants behave experimentally as
though they have lost their vacuolar Pi pool but have retained their metabolic
Pi (BIELESKI 1968 b). In view of the central involvement of Pi as an end-product
of many enzyme reactions, a substrate of many others, and as a controller
of others again, it is logical to expect that the cell would use compartmentation
111.2 Physiology and Metabolism of Phosphate and Its Compounds 433

either to maintain cytoplasmic Pi concentration constant at the expense of that

in the vacuole, or by changing Pi fluxes between compartments to create Pi
concentration changes as triggers for some metabolic events. An example of
the latter might be the triggering of phosphofructokinase activity in Jhe cyto-
plasm of fruit cells as a prelude to the respiratory burst of the climacteric.
This activity can respond to increased Pi levels in the cytoplasm which might
result from changes in Pi flux across both the tonoplast and the plasmalemma
(CHALMERS and ROWAN 1971, WOODROW and ROWAN 1979).
There needs to be separate mention about the compartmentation of the
phosphatases (phosphomonoesterase, RNAase, DNAase, P-lipase). Most of the
activity in normal cells appears to be localized in the vacuole (MATILE 1978),
but under the stress of P deficiency, there can be a considerable increase in
phosphatase activity which may now be located in the apoplast or outer cell
membrane (RIDGE and ROVIRA 1971, BIELESKI and JOHNSON 1972). Not much
is known about the significance of these different pools of phosphatase: by
analogy with the animal lysosome, the normal plant vacuole may supply a
salvage mechanism whereby structural P-containing compounds such as P-lipids
and RNA are hydrolyzed to their basic units (MATILE and WIEMKEN 1976).
Phosphatase located in the outer apoplast is demonstrably able to release Pi
from externally supplied P-esters, making Pi available for growth (KNYPL 1978).
Thus the apoplastic phosphatase could potentially release Pi from humus materi-
als in the soil. However, the roles of the phosphatases and the significance
of their location need a lot more study.

5 The Form of Phosphorus in the Cell

Though aminoethyl phosphonate is known from protozoa (BEEVER and BURNS

1976), all P in higher plants is as phosphate, either free as Pi, or esterified
through a hydroxyl group to a carbon chain (C-O-P) as in the simple phos-
phate esters, or attached to another phosphate by a pyrophosphate bond, as
in the nucleoside di- and triphosphates and in polyphosphate (Fig. 3). Individual
compounds where phosphate is differently linked (e.g. by N-P bonding in carba-
myl phosphate) are biologically important but present in very small amounts.
It is a characteristic of the ester phosphate bond (particularly when associated
with a carboxyl or enol group), and even more so the pyrophosphate bond,
that it has a high negative free energy of hydrolysis and yet is reasonably
stable under biological temperatures and pH. Under the control of various
enzyme systems, the breaking of such a bond, which is highly exergonic, can
be coupled to various endergonic steps, driving on otherwise unfavourable syn-
thesis in what is now an overall exergonic and favourable reaction. An example
here is the reaction, catalyzed by PEP carboxylase, where phosphoenolpyruvate
plus CO 2 gives oxaloacetate plus Pi. Similarly, where there are existing ester
bonds to cleave, phosphorolysis by Pi can replace hydrolysis by H 2 0, conserving
some of the bond energy in the newly created phosphate ester bond, at the
434 R.L. BIELESKI and LB. FERGUSON:

(~C.)

CH20-Acyl
I
Acyl-OCH
I
OH
I
CH20-P-O-Alcohol
ttf"oI
O-P-OH
I
OH

l> oI

OH OH
I I
OH
I
HO-P-O-P-O- P-O-CH2 0
l()r oI OH

bAbQ"" O-P-OH

?
(Etc.)
OH OH
Fig. 3 a-d. General structures of phosphorus-containing compounds. The unit structure
of RNA is shown in d, where "base" can be adenine, guanine, cytosine or uracil. In
DNA the free OH on the pentose is replaced by H, and uracil by thymine. The structure
of a phospholipid is shown in b, where Acyl can be one of several fatty acids (e.g.
linolenic acid), and Alcohol can be one of the hydroxyl-bearing compounds choline,
ethanolamine, inositol, serine or glycerol. The monoester structure of (X-glycerophosphate
(1-phosphoglycerol) is shown in a, as an example of a simple phosphate ester. The pyro-
phosphate structure of ATP is shown in c

expense of making the reaction less exergonic. An example here is the phospho-
rylase-catalyzed formation of glucose-i-phosphate from starch. Thus energy
which is carried in the phosphate ester and pyrophosphate bond (the so-called
"energy-rich bond") is used to drive many of the anabolic processes in the
cell, and to conserve the energy released by many of the catabolic ones. In
evolution, the phosphate ester and pyrophosphate bonds have become the ener-
gy currency of the cell, to the extent that almost all the metabolic pathways
involve one or another ofthe nuc1eotides and phosphate esters (ROWAN 1966).
There is a second, contrasting role that phosphate can play, based on the
relative stability of its diester state, where there are two C chains linked to
two of the hydroxyls of a single phosphate. In this state, phosphate forms
a bridging group, connecting units together and thus helping to build up macro-
molecular structures. Note that the third hydroxyl, which remains free, is still
readily ionizable, with a pKa around 5, so that even as a bridge, the phosphate
remains a hydrophilic, moderately acidic centre.
In analytical fractionation procedures, the P-containing material in the tissue
tends to fall into five groups; and conveniently, these groups match the rather
distinct roles that are served by the different compounds. Two, DNA and RNA,
remain in the tissue residue when the tissue is extracted with simple solvents;
111.2 Physiology and Metabolism of Phosphate and Its Compounds 435

P-lipid is soluble in and extracted by lipid solvents such as chloroform; while

the simple P-esters plus Pi are extracted by ordinary aqueous solvents. Pi can
then be selectively precipitated as its barium salt or trimethyl ammonium com-
plex (SUGINO andMIYOSID 1964), or selectively dissolved in butanol and assayed
as its phosphomolybdate complex (MARSH 1959). The absolute and relative
amounts of each fraction will depend on the nature of the tissue. Over the
years there have been many studies of varying degrees of thoroughness on
plant tissue P fractions, and we have noted an overall pattern in these data.
Thus tissues which have a high proportion of cytoplasm and very small vacuoles
(e.g. root tips) have high levels of RNA, P-lipid and particularly the P-esters
(TSUJI 1964); tissues that are metabolically active but vacuolate (young to
mature leaves, young fruits) have moderately high amounts of all five groups
(BARKER et al. 1962, WESTE et al. 1974, CHAPIN and BIELESKI 1982); storage
tissues (seeds, tubers) may have high amounts of Pi, or phytic acid (MAKOWER
1969, MUKHERJI et al. 1971), or in the case of fungi, polyphosphate (HAROLD
1966); while woody tissues of low metabolic activity will have a generally low
P level with a relatively high proportion of Pi. Taking the young leaf as a
representative tissue, the proportions of the different groups are (values in llg
atoms P g-l fresh wt.): Pi, 10; RNA, 2; DNA, 0.15; P-lipid, 1.5; P-ester, 1.0
(BIELESKI 1973). Values for actively growing fungi are similar in proportion
but about ten times higher on a fresh weight basis: Pi (plus polyphosphate),
100; RNA, 45; DNA, 2.5; P-lipid, 10; P-ester, 10 (BEEVER and BURNS 1980).
It is possible to define the roles of the different P fractions reasonably confi-
dently. DNA is of course macromolecular, with a molecular weight exceeding
10 6 , and is responsible for carrying the genetic information of the cell. In theory,
the amount of DNA in a cell is fixed, being that amount required to equip
one nucleus with one set of chromosomes. Although the amount of DNA per
cell has been measured and used to indicate whether the cells are polyploid,
the situation is rather a special one. There are at least two reasons why plant
tissues do not have a fixed amount of DNA per cell; firstly, polyploid cells
can develop quite easily during cell differentiation and are common in differen-
tiated organs; secondly, DNA is not confined to the nucleus. There is now
ample evidence that the chloroplasts and mitochondria separately contain their
own DNA, responsible for carrying part of the genetic information required
by the organelles themselves (KUNG 1977, BORST 1972). In a photosynthetic
cell, this can be a substantial proportion of the total DNA in the cell, e.g.
about 3% in Chlamydomonas (SAGER and ISIDDA 1963). The main function
of the P in DNA is to form a structural bridge between the deoxyribonucleosides
that are the letters in the genetic code.
RNA is similar to DNA in structure, macromolecular (23,000-1.3 x 106
m.w.), and again the role of P is to form a bridge between the ribonucleoside
units. (An important exception is the terminal phosphate of tRNA, which has
an energy-transfer function as a carrier and activator of amino acids.) The
RNA is involved in the translation of the genetic information (via mRNA)
and in the synthesis of protein, both as a structural part of the ribosomes
(rRNA, 25S and 18S) and as the amino acid carrier which recognizes the nucleo-
tide triplet code (tRNA, 4S). Thus the amount of RNA in the cell tends to
436 R.L. BIELESKI and I.B. FERGUSON:

be much higher when the tissue is actively synthesizing protein. RNA is distrib-
uted throughout the cytoplasmic space of the cell, but each of the different
organelles isolated by cell fractionation techniques is characterized by its own
particular pattern and proportion of the various RNA's: thus the nucleus, the
mitochondrion, the "general cytoplasm", the amyloplast and the chloroplast
are all distinctive. In particular, the chloroplast apparently possesses its own
protein-synthesizing machinery, can contain 5%-50% of the cellular RNA, and
has a ribosomal RNA fraction (23S RNA and 16S RNA) which can be distin-
guished from the RNA of the main cell (25S and 18S) (LOENING and INGLE
1967).
The P-lipid' fraction is recovered in the various analytical procedures by
its ability to dissolve in a lipid solvent such as chloroform or ether. The phos-
phate, though in the diester or "bridging" form, is readily ionized and highly
hydrophilic, and so for the whole molecule to be lipid-soluble in this way,
there needs to be a lipid-like part also present. This is in fact what occurs:
a phospholipid can be thought of as a lipid which has one of its three fatty
acids replaced by a phosphate-containing group, the esterification being through
the phosphate itself (Fig. 3). The glycerol group can be thought of as the centre
of the molecule. Each of the two fatty acids attached to it can come from
eight or so long-chain fatty acids, C 12 -C1S ' saturated or mono-, di- or tri-
unsaturated, the most important being palmitic, linoleic and linolenic, and giving
in excess of 50 possibilities. The phosphate-containing group esterified to the
third hydroxyl is one out of only five main possibilities. In all classical separation
techniques, the nature of this phosphate-containing group dominates the separa-
tion, the nature of the fatty acids playing very little part, so that we recognize
five main "phospholipids" - phosphatidyl choline, phosphatidyl ethanolamine,
phosphatidyl glycerol, phosphatidyl inositol and phosphatidyl serine. There is
also a sixth of more complex structure, diphosphatidyl glycerol, which has two
diacylated glycerols, or fatty parts, connected by two phosphates to a glycerol
which now forms a bridge. Strictly speaking, each of these phospholipids so
defined is a whole series of compounds depending on the fatty acids present,
and though the nature of these has little effect on the broad physical properties,
there is not much doubt that it has considerable biological significance, since
different membranes yield phospholipids with characteristically different fatty
acid patterns (DONALDSON and BEEVERS 1977).
There are other compounds encountered in plant extracts which are related
to the phospholipids, and which can be obtained from them as a result of
hydrolytic cleavage by one of the phospholipases found in plant tissues. Taking
phosphatidyl choline as an example, if one fatty acid is removed from the
C1 position (by phospholipase B), we obtain lyso-phosphatidyl choline; if the
second fatty acid is also removed from the C2 position (by phospholipase A)
we have glycerophosphatyl choline; if the whole glycerol-plus-two-fatty-acid
unit is split off (by phospholipase C) we obtain phosphatyl choline; if on the
other hand, the choline is split from the rest of the phospholipid (by phospholi-
pase D) we obtain phosphatidic acid. (Phosphatidic acid is also obtained when
the other four phospholipids have their characteristic end-group removed in
III.2 Physiology and Metabolism of Phosphate and Its Compounds 437

the same way.) Phosphatidic acid is one of the main precursors in phospholipid
synthesis (MAZLIAK 1973).
The P-lipid compositions of the various plant tissues are relatively uniform,
the biggest single factor being whether or not the tissue is photosynthetic. In
both green and non-green tissues, phosphatidyl choline is predominant
(40%-50% of total P-lipid), and phosphatidyl ethanolamine is also important
(20%-30%), but the proportion of phosphatidyl glycerol is very much higher
in photosynthetic tissues (15%-25%) than in non-green ones (below 5%) (AB-
DELKADER 1968, BIELESKI 1972, ASHWORTH et al. 1981). The remaining P-lipids
are present in generally small amounts; phosphatidyl inositol (5%-10%), phos-
phatidyl serine (1 %-4%), diphosphatidyl glycerol (1 %-3%) and phosphatidic
acid (1 %-3%). Because phospholipases can so easily create artefacts during
tissue extraction, it is uncertain whether some P-lipid related compounds such
as the lysophospholipids are normally present in tissues, but labelling patterns
and other data make it clear that others such as phosphatidic acid (0.5%-3%
of total P-lipid), phosphatyl choline (0.3%-1.5%) and oc-glycerophosphate
(0.1 %-1 %) are genuine constituents. The various cell organelles have more
clearly distinguished P-lipid compositions than the bulk tissues (FALK and
STOCKING 1976). The most characteristic pattern is that of the chloroplast frac-
tion, which has very little phosphatidyl ethanolamine (4%-8% of total P-lipid),
and where the amount of phosphatidyl choline (30%-40%) is exceeded by that
of phosphatidyl glycerol (35%-45%) (ONGUN et al. 1968, BIELESKI 1972). The
phosphatidyl glycerol is itself unusual in having a very high proportion of its
fatty acid present as trans hexadec-3-enoic acid, rare in or absent from the
other P-lipids. The microsome fraction appears to be high in phosphatidyl ino-
sitol (DONALDSON et al. 1972, FALK and STOCKING 1976) and to have P-lipid
with a relatively high proportion of linolenic acid (ABDELKADER and MAZLIAK
1970). However, there is a great deal yet to be done in relating specific phospho-
lipids to specific organelles and to specific functions.
While the detailed function of specific phospholipids is uncertain, the broad
function is clear. The structure of the phospholipid is very like that of a deter-
gent: there is a hydrophilic and a lipophilic region combined in one molecule,
so that where there is a lipid-water interface, the phospholipid molecules can
orient themselves across the boundary and stabilize it. This can lead to the
formation of a stable emulsion. More important to living systems is the situation
where an oriented layer of phospholipid molecules folds back on itself to create
a laminate or bilayer with two hydrophilic outer surfaces sandwiching an inner
hydrophobic region; this is the basic structure of the various cell membranes
(FALK and STOCKING 1976). It is no surprise that the chloroplast, with its highly
developed membrane (thylakoid) system, can contain over 40% of the total
P-lipid in photosynthetic cells (BIELESKI 1972). The proteins which perform the
various functions of metabolism and transport are thought to be held in place
(whether on one face or crossing the membrane from one side to the other)
by the interaction of their own lipophilic regions with that of the membrane.
The interaction should be made more precise and specific by the nature of
the phospholipid (and the nature of its fatty acids) located in a given region.
438 R.L. BIELESKI and I.B. FERGUSON:

This is one way in which the specific structure of each individual phospholipid
could be of significance in the living cell, despite its great similarity to other
phospholipids.
The simple P-esters form the fourth of our five phosphate fractions. Together
they represent the metabolic machinery of the cell. Upwards of 50 individual
esters have been identified at various times in various tissues and by various
methods: amongst the compounds recognized are the 4 nucleoside mono-, di-
and triphosphates; about 8 nucleotide diphosphosugars and related compounds,
zeatin derivatives, about 12 pentose, hexose and heptose phosphates and diphos-
phates, about 6 polyol phosphates, 4 triose phosphates, 3PGA, phosphoenolpy-
ruvate, phosphatyl choline, and 6 phosphogluconate amongst them, but there
is no point in making a full list. The following general conclusions emerge.
P-esters that have been found in animal tissues are, in general, also found
in plant tissues (creatine phosphate may be an exception), but at much lower
concentrations. Even though the individual enzymes may differ somewhat, the
central metabolic pathways involving energy transfer and carbohydrate transfor-
mation are the same in plants and animals, and involve the same P-esters
(ROWAN 1966), presumably reflecting a very early evolutionary development
of the various roles for P-esters. Within different plants there is again a very
conservative pattern in the P-esters, to the extent that it is possible to write
down a generalized P-ester pattern for most plant tissues (see Table 1 in BIELESKI
1973). About 70% of the P in the P-ester fraction is usually contained in nine
compounds: G6P, fructose-6-P and mannose-6-P (20%, 6% and 4%), ATP
and ADP (10% and 3%), UTP, UDP and UDPG (4%, 5% and 9%) and
3PGA (8%). [A special situation can arise in seeds and tubers where phytic
acid - i.e. inositol hexaphosphate - can be the major P-ester present: functional-
ly, it needs to be considered as a storage compound like polyphosphate rather
than as a metabolically active P-ester (SAMOTUS and SCHWIMMER 1962, MATHE-
SON and STROTHER 1969)]. In normal intact tissues, the ATP/ADP/AMP ratio
is around 10/3/1 and the "energy charge" remains in the region 0.8-0.9 (PRADET
and RAYMOND 1981). The significance of the energy charge as a prime controller
of plant cell processes is discussed in a forthcoming review (PRADET 1983: Annu
Rev. Plant Physiol.).
There have been very few attempts to compare total P-ester patterns for
different subcellular fractions, partly because of the difficulty in measuring the
very small amounts of the many P-esters present, and partly because the rapid
turnover of P-esters (see Sect. 6) must inevitably allow for drastic changes in
P-ester pattern during isolation of the organelles. Special rapid-freezing and
non-aqueous isolation techniques have been used to try and avoid this problem
(WALKER 1976). There is some hope that new 31p_NMR techniques will allow
us to estimate the concentrations of some P-esters in different intracellular com-
partments within an intact, living tissue (ROBERTS et al. 1980, LOUGHMAN per-
sonal communication). Useful information has also come from studying the
behaviour of individual P-esters in isolated fractions: not surprisingly, mito-
chondria show a major capability for ATP synthesis and are readily permeable
to Pi, ADP and ATP (HELDT 1976); amyloplasts lack RNA but contain relative-
ly high amounts of the triose and hexose phosphates plus Pi (Lru and SHANNON
111.2 Physiology and Metabolism of Phosphate and Its Compounds 439

1981); chloroplasts contain Pi (10-15 mM), 3PGA and the two triose phos-
phates, and though not permeable to Pi (MOURIOUX: and DOUCE 1981), readily
exchange it across the chloroplast membranes for triose-P, suggesting that this
is the mechanism by which carbon fixed in photosynthesis passes into the cyto-
plasm to be further elaborated to carbohydrate (FLIEGE et al. 1978). The vacuole
is another recognizable cell compartment: here, Pi is the main P-compound
and P-esters are absent (Sect. 4).
Inorganic phosphate, Pi, is the last of the five fractions to be considered.
Though the most important single compound, it is by far the most variable
in proportion and amount of all the P fractions, ranging from about 1 limol g - 1
fresh wt. and 15% of total P in a P-deficient tissue to more than 40 limol g-1
fresh wt. and 70% of total P in a tissue approaching phosphate toxicity condi-
tions (BIELESKl 1968a, b, WESTE et al. 1974). There are two points of interest
about this: firstly the presence of a significant amount of Pi even under condi-
tions of extreme P stress implies that some Pi is required for the continued
metabolic activity of the cell; secondly, the variability in amount of Pi and
the very high tissue contents reached suggest that much of the Pi in a normal
cell is surplus to requirements, and serves a storage function (see Sect. 7 for
a fuller discussion). In a normal tissue, much of the Pi appears to be in the
vacuole, and as already noted, it is apparently the only form in which P is
transported in the higher plant (Sect. 2). There is evidence that individual cell
organelles (chloroplast, mitochondrion) can transport Pi, and require Pi at con-
centrations around 10 mM for their normal functioning (SANTARIUS and HEBER
1965, WALKER 1976, WISKICH 1977).
Two inorganic P compounds that are related to Pi require separate comment,
pyrophosphate and polyphosphate (in which many Pi units are linked in a
chain by pyrophosphate bonds). Pyrophosphate has only occasionally been de-
tected in plant cells, but is a known control agent for some plant enzyme systems
and a product of others (BEEVER and BURNS 1980). It must therefore be present,
but in functional terms it should be considered as one of the P-esters, along
with the nucleotides. The situation with polyphosphate is different. Claims have
been made for its presence in higher plant tissues (COLEMAN and SPECHT 1981),
but we feel that, because of possible contamination by mycorrhizal or other
fungi, the situation is still a "not proven" one. There is no doubt, however,
about its importance in the fungi (BEEVER and BURNS 1980) and some algae
(RAVEN 1974a). A survey of the earlier literature on polyphosphates can be
found in by KUHL (1960) and KULAEV (1979). Polyphosphate can be virtually
absent from fungal cells that are starved of P or that are actively growing,
but can comprise over 60% of the total P and as much as 3% of the dry
weight (250 limol g-l fresh wt.)in mature hyphae (BEEVER and BURNS 1980).
Polyphosphate has been found in both the vacuole and the cytoplasm, possibly
in different forms: it can sometimes be soluble in the cell, but it readily becomes
associated with positively charged macromolecules as microscopically visible
granules under various conditions of study (BEEVER and BURNS 1980). It may
be the main form in which phosphate is transported in the fungal hypha (Cox
et al. 1980), though in mycorrhizas, the transfer from hypha to host root is
probably as Pi (MARX et al. 1982). In its sensitivity to Pi supply, its transport
440 RL. BIELESKl and I.B. FERGUSON:

role and much of its general behaviour, polyphosphate can be thought of as

a special form of Pi. In most plant cells, Pi can be segregated from the cytoplasm
by being stored in the vacuole; polyphosphate, however, represents a form
in which Pi can be segregated chemically within the cytoplasm itself as well
as physically in the vacuole. Phytic acid, though a P-ester of inositol, also
serves a storage role, particularly in seeds where it can form 50% or more
of the total P. As germination of the seed begins, phytase is formed, phytic
acid is hydrolyzed and Pi is released (GUARDIOLA and SUTCLIFFE 1971, FERGUSON
and BOLLARD 1976). Most of the P used during early shoot and root growth
is derived from this source; only later does Pi uptake from the medium take
over.
Thus, a reasonable amount is known about the function and distribution
of the various P-containing compounds. We are helped by the relative insolubili-
ty of four of them - DNA, RNA, P-lipid and polyphosphate - and the specific
staining procedures that can be used to recognize them under both the optical
and electron microscopes. Further effort is now needed to obtain more precise
information on the distribution of the remaining fractions - Pi and the various
P-esters. Enzyme-coupled stains, selective precipitation methods, electron probe
and NMR techniques will all play their part in the near future.

6 Synthesis and Turnover of Phosphorus Compounds

A frequent procedure for studying the metabolic pathways followed by a partic-

ular compound is to supply it in radioactive form to the tissue or organelle
preparation for a very brief time, to follow this up by the same compound
in non-radioactive form, and to identify the radioactive products - that is,
to carry out a "pulse-chase" experiment. There are some basic difficulties in
using this tactic to discover what happens to Pi in intact plants. Firstly, Pi
is metabolized very rapidly and the pulse time needs to be very short. Secondly,
there is a very large amount of Pi already present in the tissue, and it is divided
between two pools which behave very differently (see Sect. 4). Thirdly, in stan-
dard pulse-chase experiments, the procedure is to study how the labelling of
a particular intermediate falls to zero: this cannot be done with the P-esters
because an inevitable end-product of the reactions is the original precursor,
Pi itself, and so the P-ester radioactivity never falls close to zero, but comes
to equilibrium. This "time to equilibrium" can be used as an alternative al-
though less satisfactory estimate of the rate and course of reactions, and despite
its poorer pedigree has given some useful information. The actual parameter
involved is the "half-time of labelling" (the time taken for a given ester to
reach half the equilibrium radioactivity), and related to it, the" turnover time"
(the time taken for the amount that is present of that esterto be broken down
and resynthesized). In a simple case, turnover time = half-time/In 2 (see BIELESKI
and LATIES 1963, BIELESKI 1973 for further discussion).
We can look at what happens when 32Pi enters intact plant tissues (LOUGH-
MAN 1960, WEIGL 1963, BIELESKI 1968b, 1973). The most rapidly labelled com-
111.2 Physiology and Metabolism of Phosphate and Its Compounds 441

pounds detected are ATP and UTP, supporting the view that in plants, as
in animals, the main entry for P into organic compounds is through esterification
of Pi as the y-phosphate group of A TP. It further suggests that the exchange
reaction between ATP and UTP, catalyzed by nucleoside diphosphate kinases,
is very rapid, so that the effective pool size of A TP is in fact (ATP + UTP).
The estimated turnover time for these two nucleotide triphosphates is always
very short, 10-40 s: that is to say, the entire supply of ATP+ UTP in the cell
is broken down (to ADP + UDP) and resynthesized in less than a minute. From
this turnover time and the pool size, we can then calculate the rate at which
the y-phosphate link is synthesized. Data obtained for Spirodela are given in
Table 3, showing the relative rates of synthesis of the different P-esters. It has
not been possible to directly measure a turnover time for the cytoplasmic Pi
pool, but a maximum value (i.e. minimum rate of Pi utilization) can be estimated
on the reasonable basis that one Pi is used for each y-phosphate link synthesized
(see below). The P-lipid, turnover time is an average value for all five major
P-lipids (BIELESKI 1972). The data available on RNA turnover take no recogni-
tion of different RNA fractions and are based on "non-extractable P" rather
than P that has been shown to be in RNA (BIELESKI unpublished data); but
a turnover time of 15 h is indicated. Again, a maximum time of 47 h for RNA
and DNA can be deduced on the basis that in the absence of true turnover
(i.e. cycling) there will still be an increase in the amount of each during growth,
and this will give an apparent turnover time (Table 3).
Such data are direct estimates of P-ester production. It is also possible to
indirectly estimate what the rate of ATP production should be from the respira-
tion rate. If the tissue produces all its y-phosphate as ATP by normal aerobic
respiration, then the utilization of 1 ~mole of oxygen yields 6.2 ~mol ATP.
The data presented in Table 3 are subject to a number of assumptions and
approximations, and should be viewed in this light. However, they do allow
us to make some useful general comments on the probable behaviour of P
in the various metabolic pathways of a plant cell. Firstly, y-phosphate produc-
tion as A TP or UTP measured in Spirodela is very similar to that expected
if it all came from respiration through the normal pathway. The reasonable
match between the two sets of figures allows us to hope that the turnover
rates which have been measured do have some significance. Secondly, the rela-
tively high rate of p-phosphate turnover suggests that about one in four ATP
or UTP molecules is utilized in reactions that take it through to the mononucleo-
tide. One important pathway of this type is involved in carbohydrate synthesis
and metabolism: UTP + G6P -+ UDPG -+ UMP + carbohydrate synthesis.
Thirdly, only 1 in 6 G6P molecules that are produced is used in respiration:
the rest are presumably drawn off in processes that do not yield ATP, such
as starch synthesis. A significant part of cellular G6P has been shown to be
localized in the amyloplast, along with other esters of starch synthesis, as a
separate pool (Lm and SHANNON 1981). Fourthly, 3PGA has a high rate of
turnover, which we can take as an index of the relative importance of the
photosynthetic pathway in terms of total metabolism in the cell. (Note that
Spirodela used in this study was grown on medium containing glucose and
so carbohydrate metabolism could exceed photosynthesis.) Fifthly, the turnover
of phosphatyl choline is consistent with it being largely involved in P-lipid
442 R.L. BIELESKI and LB. FERGUSON:

Table 3. Turnover times and rates of synthesis of various P fractions in Spirodela. (Data
from BIELESKI 1968 a, b, 1972 and unpublished data)

Phosphorus Amount Turnover Synthesis rate

fraction (nmol g-1 fresh wt.) time (min) (nmol
P g-1 min-I)

From tissue analysis:

ATP+ UTP (yp)a 170 0.50 340
NTP + NDP (fJP) b 300 3.5 85
G6P 670 7 95
3PGA 330 6.5 50
Adenosine ape 200 ca. 160 1e
Uridine aP 320 ca. 160 2e
Phosphoryl choline 70 6 12
P-lipid 2700 130 20
RNA (est) 4900 ca. 900 5e
RNA (min. rate)d 4900 ·2800 2
DNA (min. rate)d 560 2800 0.2
From respiration rate: f
"ATP" 170 0.25 650
G6P 670 40 17
Pig (3400) 5 650

a "y" is the yP of the nucleotide pool. The turnover time is given by the labelling
pattern of ATP and UTP, and the amount is 1/3 (total P in ATP + UTP)
b "PP" is the PP of the nucleotide pool. The turnover time is given by the labelling
pattern of ADP and UDP, and the amount is 1/3 (p in ATP +UTP) +t (p in ADP+
UDP)
e "Adenosine ocP" is the P directly attached to adenosine. The turnover time is given
by the labelling pattern of AMP, and the amount is 1/3 (p in ATP)+t (p in ADP)+(p
in AMP). "Uridine ocP" is calculated in a similar way
d Minimum synthesis rates for RNA and DNA are calculated on the basis that, as
the amount of each must increase with growth, the maximum turnover time (minimum
synthesis rate) is the same as the doubling time for the tissue, 1.9-2.0 days. The esti-
mated RNA time is from unpublished data
e If all the "adenosine ocP" and "uridine ocP" synthesis was used to replace nucleotide
lost in RNA synthesis, the synthesis rate of each should be about 25% of the RNA
value (as 1 in 4 P of RNA, approximately, is contributed by adenylic and uridylic
acids)
f Values can be calculated from the known respiration rate of Spirodela, 2.3 III O 2 g-1
fresh wt. min - 1 (105 nmol O 2 g -1 fresh wt. min - 1) assuming normal aerobic respira-
tion rate and 6.2 mol ATP produced (or 6.2 mol Pi utilized) per 1 mol oxygen consumed
a~d _1/ 6 mol glucose utilized. The phosphate uptake rate is 7.5 nmol g-1 fresh wt.
mm
g The size of the Pi fraction assumes that 12% of the tissue Pi is in the metabolic
pool and turning over, and that the remaining 88% is non-metabolic

synthesis; and the turnover rate of adenosine- and uridine-aP is similarly consis-
tent with the turnover rate found for RNA. Taken all together, the data indicate
the following rough balance sheet: for every 1,000 P atoms involved in A TP
(and UTP) production, about 250 are involved in carbohydrate synthesis, 150
in photosynthesis, 60 in P-lipid turnover and synthesis, 15 in RNA turnover
111.2 Physiology and Metabolism of Phosphate and Its Compounds 443

and synthesis, and 1 in DNA turnover and synthesis. About 15-25 new atoms
of P are imported into the cell to replace P withdrawn for growth and storage.
But for every P atom taken up, 40-80 are cycled through ATP - that is, each
Pi molecule which enters the cell makes on average 60 passages through ATP
and the metabolic cycle before it is withdrawn in growth, storage or transport.
We can conclude that the entire Pi pool in the cytoplasm must tum over every
5 min or so. This is a remarkably short time, and indicates the very important
role played by Pi in cellular metabolism.

7 Dynamics of Phosphate Use in the Plant

At this point it is instructive to look at the dynamics of cellular and plant

utilization of P. We can consider the plant in terms of four compartments.
The apoplast, which is the non-living tissue lying outside the plasmalemma,
can be divided into the outer apoplast, in ready communication with the soil,
and the inner apoplast (e.g. the xylem) in communication with the main body
of the plant but separated from the outer apoplast by a diffusion barrier (endo-
dermal suberin or the casparian band). The region inside the plasmalemma
can be divided into the cytoplasm or metabolic space (the symplast) and the
vacuole or non-metabolic space, inside a second membrane, the tonoplast. Pi
moves readily between these four compartments, and Fig. 4 summarizes the
nature of the interplay.
Pi in the soil is released into solution by the dissolving of soil minerals
or by the death and decomposition of other organisms (BEEVER and BURNS
1976). It diffuses into the outer apoplast. The uptake mechanism at the plasma-
lemma pumps Pi into the cytoplasm where it is rapidly esterified into ATP.
This is the metabolic function of P. A small portion is drawn off in the synthesis
of P-lipid, DNA and RNA for growth - this is the structural function of P
- but most passes through the metabolic paths and reappears as Pi in the
metabolic pool. If the amount withdrawn for growth is not as great as the
uptake, the Pi concentration in the cytoplasm will rise. An active homeostatic
mechanism at the tonoplast comes into play, restoring the Pi concentration
in the cytoplasm by transferring Pi to the vacuole. Though this process has
not been directly demonstrated in higher plants, it is well shown in the data
for Hydrodictyon (RAVEN 1974a) (Fig. 2). This movement into the vacuole repre-
sents the storage function of P in the cell (incorporation into polyphosphate
may playa similar role in fungi). A second option is that Pi may instead be
moved to other cells far away from the source of supply - this is the transport
function of P. The major movement upwards in the plant occurs by Pi being
unloaded at the plasmalemma into the inner apoplast, thence to be taken
upwards in the xylem stream. An opposite situation arises if the Pi demands
of the cell exceed the supply from the environment. Pi must now be returned
to the cytoplasm from the vacuole, or brought in by retranslocation from distant
tissues: both are known responses to P deficiency. Such retranslocation typically
444 R.L. BIELESKI and LB. FERGUSON:

,
_ SYM'l ... SMIC _ _ .... .. O'OPlAS. IC
CYTOPLASM " MOVEMENT SlUND

I
~--~0~-~5~--~-- ~gC~~~N
I ~ EFFlUX
: I'OSSI
I
11"'Un

SOIL Pi ~U'T"'I(E~_.....;.1;0;...;.;.....;..;;;..... Pi 0 - 10 T.... NS' O lT

~II :~ M""~ ~~~

(WI'"OIAWAL)

lV-
-
(CYWHO. ' -Ul'II I , -U"O)

"1-
SO il I MEM ..... NE
, ... n , ... n 0 .ESY NTH ES IS

l j

/ -Em. ; "- Pi / '

SOil , - ... n ' -... SE ' - "'SE

c (S" OIAOI... S",VAO.)

' l... NT DE ... TH
•
~
SO IL MINER ... lS

O UTER INNER
APOPLAST ~_ _ ....Iiiiiiiii_L ____A_PO=-P..::LA:...::..ST_ _

Fig. 4. Dynamics of phosphorus use in the cell. Numbers relating to arrows on the diagram
give the relative significance of the different pathways of P movement and in a plant
like Spirodela, are approximately equivalent to (nmol g- l fresh wt. min - 1). P-ase is
"phosphatase", and in the context of this diagram, includes non-specific acid and alkaline
phosphomonoesterases, plus ribonucleases, plus phospholipases

occurs in the phloem, and apparently involves both apoplastic and symplastic
transport steps.
In summary: the cytoplasmic pool of Pi appears to be the hub of all cellular
P metabolism and of the whole P economy of the plant. It is a major controller
of cell processes, but what we now need to know is how this control is achieved
and regulated. That will be a major research area for the future.
111.2 Physiology and Metabolism of Phosphate and Its Compounds 445

8 Conclusions

With a subject as broad as the P nutrition of plants, it is too much to hope

for a compact body of information without any large gaps of knowledge. In
this chapter, we used scattered data on P metabolism for relatively few plants,
in order to draw a picture for "plants" in general. We do not apologize for
doing so: we had a choice of either presenting the subject in a very broken
fashion, reviewing only those parts where a hundred researchers have gone
before, or else extending the little knowledge we do have to fill in those gaps,
and present a coherent if tentative and sketchy picture. For this volume, which
aims to synthesize current knowledge rather than review the literature, we have
chosen the second course. There is some hope that our picture will not be
too far wrong, partly because a lot of the data, though scattered, do correspond
quite well and can be accounted for by our interpretations. A second factor
is that the behaviour of plants towards P seems to be very conservative, going
back to the fundamental importance of P in all three of the major life processes
- reproducing oneself (DNA), segregating oneself from the hotch-potch world
outside (P-lipid), and acquiring and using energy to beat the second law of
thermodynamics in creating the ordered state of living things (P-esters). Thus
the same compounds are present in all tissues, plant and animal, they serve
the same role, and they usually seem to be present in similar proportions, so
there is a lot of common ground in the processes that occur. But despite these
reassuring thoughts, the chapter we have presented must be taken as a starting
point for future research, not a museum full of past knowledge.

References

Abde1kader AB (1968) La lipogem:se dans Ie tubercule de pomme de terre (Solanum

tuberosum L.). 1. Analyse et biosynthese des lipides dans Ie parenchyme central. In-
fluence de la "survie" (ageing) de rondelles de parenchyme sur cette biosynthese.
Physiol Veg 6:417-442
Abdelkader AB, Mazliak P (1970) Echanges de lipides entre mitochondries, microsomes
et surnageant cytoplasmique de cellules de pomme de terre ou de chou-fleur. Eur
J Biochem 15: 250-262
Asher CJ, Loneragan JF (1967) Responses of plants to phosphate concentration in solu-
tion culture: I. Growth and phosphorus content. Soil Sci 103:225-233
Ashworth EN, St John JB, Christiansen MN, Patterson GW (1981) Characterization
of the phospholipid composition of wheat roots using high-performance liquid chro-
matography. J Agric Food Chem 29: 879-881
Barber SA, Walker JM, Vasey EH (1963) Mechanisms for the movement of plant nu-
trients from the soil and fertilizer to the plant root. J Agric Food Chem 11 :204-207
Barker J, Isherwood FA, Jakes R, Solomos T, Younis ME (1962) Determination of
certain phosphate compounds in plant extracts. Nature 196: 1115
Barrier GE, Loomis WE (1957) Absorption and translocation of 2,4-dichlorophenoxy-
acetic acid and p 32 by leaves. Plant PhysioI32:225-231
446 R.L. BIELESKI and I.B. FERGUSON:

Bassham JA, Kirk M, Jensen RG (1968) Photosynthesis by isolated chloroplasts. I. Diffu-

sion of labelled photosynthetic intermediates between isolated chloroplasts and sus-
pending medium. Biochim Biophys Acta 153:211-218
Beever RE, Burns DJW (1976) Microorganisms and the phosphorus cycle: some physio-
logical considerations. In: Blair GJ (ed) Prospects for improving efficiency of phos-
phorus utilization. Reviews in rural science, vol III. Univ New England, Armidale
Beever RE, Burns DJW (1977) Adaptive changes in phosphate uptake by the fungus
Neurospora crassa in response to phosphate supply. J Bacteriol132: 520-525
Beever RE, Burns DJW (1980) Phosphate uptake, storage and utilization by fungi. Adv
Bot Res 8:127-219
Biddulph 0, Biddulph S, Cory R, Koontz H (1958) Circulation patterns for phosphorus,
sulfur and calcium in the bean plant. Plant PhysioI33:293-300
Bieleski RL (1968 a) Levels of phosphate esters in Spirodela. Plant Physiol 43: 1297-1308
Bieleski RL (1968 b) Effect of phosphorus deficiency on levels of phosphorus compounds
in Spirodela. Plant Physiol43: 1309-1316
Bieleski RL (1969) Phosphorus compounds in translocating phloem. Plant Physiol
44:497-502
Bieleski RL (1972) Turnover of phospholipids in normal and phosphorus-deficient Spiro-
dela. Plant Physio149:74o-745
Bieleski RL (1973) Phosphate pools, phosphate transport, and phosphate availability.
Annu Rev Plant PhysioI24:225-252 . .
Bieleski RL (1976) Passage of phosphate from soil to plant. In: Blair GJ (ed) Prospects
for improving efficiency of phosphorus utilization. Reviews in rural science, vol III.
Univ New England, Armidale
Bieleski RL, Johnson PN (1972) The external location of phosphatase activity in phospho-
rus-deficient Spirodela oligorrhiza. Aust J Bioi Sci 25: 707-720
Bieleski RL, Laties GG (1963) Turnover rates of phosphate esters in fresh and aged
slices of potato tuber tissue. Plant Physiol 38: 586-594
Borst P (1972) Mitochondrial nucleic acids. Annu Rev Biochem 41: 333-376
Bowling DJF, Dunlop J (1978) Uptake of phosphate by white clover. I. Evidence for
an electrogenic phosphate pump. J Exp Bot 29: 1139-1146
Burns DJW, Beever RE (1977) Kinetic characterization of the two phosphate uptake
systems in the fungus Neurospora crassa. J Bacteriol132:511-519
Burns DJW, Beever RE (1979) Mechanisms controlling the two phosphate uptake systems
in Neurospora crassa. J Bacteriol 139: 195-204
Carter OG, Lathwell DJ (1967) Effects of temperature on orthophosphate absorption
by excised corn roots. Plant Physiol42: 1407-1412
Chalmers DJ, Rowan KS (1971) The climacteric in ripening tomato fruit. Plant Physiol
48:235-240
Chapin FS, Bieleski RL (1982) Mild phosphorus stress in barley and a related low-
phosphorus-adapted barleygrass: phosphorus fractions and phosphate absorption in
relation to growth. Physiol Plant 54: 309-317
Cole CV, Elliott ET, Hunt HW, Coleman DC (1978) Trophic interactions in soils as
they affect energy and nutrient dynamics. V. Phosphorus transformations. Microb
EcoI4:381-387
Coleman RG, Specht RL (1981) Mineral nutrition of heathlands: The possible role of
polyphosphate in the phosphorus economy of heathland species. In: Specht RL (ed)
Ecosystems of the world, vol 9 B. Heathlands and related shrublands. Analytical
studies. Elsevier, Amsterdam
Collins JC, Reilly EJ (1968) Chemical composition of the exudate from, excised maize
roots. Planta 83:218-222
Cooke JG (1981) Pollution from our pastures. Soil Water 17:13-15
Corbridge DEC (1978) Phosphorus. Elsevier, Amsterdam
Cox G, Moran KJ, Sanders F, Nockolds C, Tinker PB (1980) Translocation and transfer
of nutrients in vesicular-arbuscular mycorrhizas. III. Polyphosphate granules and
phosphorus translocation. New Phytol 84:649-659
111.2 Physiology and Metabolism of Phosphate and Its Compounds 447

Donaldson RP, Beevers H (1977) Lipid composition of organelles from germinating

castor bean endosperm. Plant Physiol 59: 259-263
Donaldson RP, Tolbert NE, Schnarrenberger C (1972) A comparison of microbody
membranes with microsomes and mitochondria from plant and animal tissue. Arch
Biochem Biophys 152: 199-215
Dunlop J, Bowling DJF (1971) The movement of ions to the xylem exudate of maize
roots. I. ProfIles of membrane potential and vacuolar potassium activity across the
root. J Exp Bot 22: 434-444
Emmert FH (1959) Loss of phosphorus-32 by plant roots after foliar application. Plant
Physiol 34: 449-454
Epstein E (1972) Mineral nutrition of plants. Principles and perspectives. Wiley and
Sons, New York
Falk RH, Stocking CR (1976) Plant membranes. In: Stocking CR, Heber U (eds) Trans-
port in plants III Encyclopaedia of plant physiology new ser, vol 3. Springer, Berlin
Heidelberg New York
Ferguson AR, Eiseman JA (1983) Estimated annual removal of macro nutrients in fruit
and prunings from a kiwifruit orchard. NZJ Agric Res 26: 115-117
Ferguson AR, Eiseman JA, Leonard JA (1983) Xylem sap from Actinidia chinensis:
seasonal changes in composition. Ann Bot (in press)
Ferguson IB, Bollard EG (1976) The movement of calcium in germinating pea seeds.
Ann Bot 40: 1047-1055
Ferguson IB, Clarkson DT (1975) Ion transport and endodermal suberization in the
roots of Zea mays. New Phytol 75: 69-79
Ferguson IB, Clarkson DT (1976) Ion uptake in relation to the development of a root
hypodermis. New Phytol 77:11-14
Fliege R, Fliigge U-I, Werdan K, Heldt HW (1978) Specific transport of inorganic phos-
phate, 3-phosphoglycerate, and triosephosphates across the inner membrane of the
envelope in spinach chloroplasts. Biochim Biophys Acta 502:232-247
Gilchrist AN, Gillingham AG (1970) Phosphate movement in surface run-off water.
NZ J Agric Res 13:225-231
Gould WD, Coleman DC, Rubenk AJ (1979) Effect of bacteria and amoebae on rhizo-
sphere phosphatase activity. Appl Environ Microbiol 37:943-946
Greenway H, Gunn A (1966) Phosphorus retranslocation in Hordeum vulgare during
early tillering. Planta 71 : 43-67
Greenway H, Klepper B (1968) Phosphorus transport to the xylem and its regulation
by water flow. Planta 83: 119-136
Groves RH (1981) Nutrient cycling in heathlands. In: Specht RL (ed) Ecosystems of
the world, vol 9B. Heathlands and related shrublands. Analytical studies. Elsevier,
Amsterdam
Guardiola JL, Sutcliffe JF (1971) Mobilisation of phosphorus in the cotyledons of young
seedlings of the garden pea (Pisum sativum L.). Ann Bot 35:809-823
Hagen CE, Hopkins HT (1955) Ionic species in orthophosphate absorption by barley
roots. Plant Physiol 30: 193-199
Harold FM (1966) Inorganic polyphosphates in biology: structure, metabolism and func-
tion. Bacteriol Rev 30: 772-794
Harvey HW (1969) The chemistry and fertility of sea waters. Cambridge Univ Press,
Cambridge
Healy WB, McColl RHS (1974) Clay particles as sources of phosphorus for Lemna
and their role in eutrophication. NZJ Sci 17:409-420
Heldt HW (1976) Transport of metabolites between cytoplasm and the mitochondrial
matrix. In: Stocking CR, Heber U (eds) Transport in plants III. Encyclopaedia
of plant physiology, new ser, vol 3. Springer Berlin Heidelberg New York
Hevesy G (1946) Interaction between phosphorus atoms of the wheat seedling and the
nutrient solution. Ark Bot 33:1-16
Johnson EJ, Bruff BS (1967) Chloroplast integrity and ATP-dependent CO 2 fixation
in Spinacia oleracea. Plant Physiol 42: 1321-1328
448 R.L. BIELESKI and I.B. FERGUSON:

Jung C, Rothstein A (1965) Arsenate uptake and release in relation to the inhibition
of transport and glucolysis in yeast. Biochem PharmacoI14:1093-1112
Kluge M, Becker D, Ziegler H (1970) Untersuchungen iiber ATP und andere organische
Phosphorverbindungen im Siebrohrensaft von Yucca flaccida und Salix triandra.
Planta 91 : 68-79
Knypl JS (1978) Reversal of the symptoms of phosphate deficiency in Spirodela by RNA
and adenosine monophosphates. Z Pflanzenphysiol 90: 265-277
Kuhl A (1960) Die Biologie der kondensierten organischen Phosphate. Ergeb BioI
23:144-186
Kulaev IS (1979) The biochemistry of inorganic polyphosphates. Wiley, Chichester New
York Brisbane Toronto
Kung S-D (1977) Expression of chloroplast genomes in higher plants. Annu Rev Plant
Physiol28 :401-437
Kunishi HM, Taylor AW, Heald WR, Gburek WJ, Weaver RM (1972) Phosphate move-
ment from an agricultural watershed during two rainfall periods. J Agric Food Chem
20:90(}-905
Leggett JE, Galloway RA, Gauch HG (1965) Calcium activation of orthophosphate
absorption by barley roots. Plant Physiol 40: 897-902
Liu T-TY, Shannon JC (1981) Measurement of metabolites associated with nonaqueously
isolated starch granules from immature Zea mays L. endosperm. Plant Physiol
67:525-529
Loening UE, Ingle J (1967) Diversity of RNA components in green plant tissues. Nature
215:363-367
Loughman BC (1960) Uptake and utilization of phosphate associated with respiratory
changes in potato tuber slices. Plant PhysioI35:418-424
MacRobbie EAC (1971) Fluxes and compartmentation in plant cells. Annu Rev Plant
PhysioI22:75-96
Makower RU (1969) Changes in phytic acid and acid-soluble phosphorus in maturing
pinto beans. J Sci Food Agric 20: 82-84
Marsh BB (1959) The estimation of inorganic phosphate in the presence of adenosine
triphosphate. Biochim Biophys Acta 32: 357-361
Marx C, Dexheimer J, Gianinazzi-Pearson V, Gianinazzi S (1982) Enzymatic studies
on the metabolism of vesicular-arbuscular mycorrhizas. IV. Ultracytoenzymological
evidence (ATPase) for active transfer processes in the host-arbuscle interface. New
Phytol 90: 37-43
Matheson NK, Strother S (1969) The utilization ofphytate by germinating wheat. Phyto-
chemistry 8: 1349-1356
Matile Ph (1978) Biochemistry and function of vacuoles. Annu Rev Plant Physiol
29: 193-213
Matile Ph, Wiernken A (1976) Interactions between cytoplasm and vacuole. In: Stocking
CR, Heber U (eds) Transport in plants III. Encyclopaedia of plant physiology new
ser, vol 3. Springer, Berlin Heidelberg New York
Mazel YT, Fokin AD (1977) Excretion of ions by roots of plants. Sov Plant Physiol
24:805-810
Mazliak P (1973) Lipid metabolism in plants. Annu Rev Plant PhysioI24:287-310
McColl RHS, White E, Waugh JR (1975) Chemical run-off in catchments converted
to agricultural use. NZJ Sci 18: 67-84
McPharlin IR (1981) Phosphorus transport and phosphorus nutrition in Lemna (Lemna
major L.) and Spirodela (Spirodela oligorrhiza (Kurz.) Hegelm.). Ph D thesis, Univ
Auckland, NZ
Mourioux G, Douce R (1981) Slow passive diffusion of orthophosphate between intact
isolated chloroplasts and suspending medium. Plant Physiol 67: 4 7(}-473
Mukherji S, Dey B, Paul AK, Sircar SM (1971) Changes in phosphorus fractions and
phytase activity of rice seeds during germination. Physiol Plant 25: 94-97
Ongun A, Thomson WW, Mudd JB (1968) Lipid composition of chloroplasts isolated
by aqueous and nonaqueous techniques. J Lipid Res 9:409-415
Pitman MG (1977) Ion transport into the xylem. Annu Rev Plant Physiol28 :71-88
111.2 Physiology and Metabolism of Phosphate and Its Compounds 449

Pradet A, Raymond P (1982) Adenylate energy charge, an indicator of energy metabolism.

In: Physiology and biochemistry of plant respiration. Palmer JM (ed) Soc Exp BioI.
Cambridge Univ Press, Cambridge
Raven JA (1974a) Phosphate transport in Hydrodictyon africanum. New Phytol
73:421-432
Raven JA (1974b) Energetics of active phosphate influx in Hydrodictyon africanum. J
Exp Bot 25:221-229
Reisenauer HM (1966) Mineral nutrients in soil solution. In: Altman PL, Dittmer DS
(eds) Environmental biology. Fed Am Soc Exp BioI, Bethesda
Ridge EH, Rovira AD (1971) Phosphatase activity of intact young wheat roots under
sterile and non-sterile conditions. New Phytol 70: 1017-1026
Roberts JKM, Ray PM, Wade-Jardetzky N, Jardetzky 0 (1980) Estimation of cytoplas-
mic and vacuolar pH in higher plant cells by 31p NMR. Nature 283:870-872
Rowan KS (1966) Phosphorus metabolism in plants. Int Rev CytoI19:301-390
Sager R, Ishida MR (1963) Chloroplast DNA in Chlamydomonas. Proc Natl Acad
Sci USA 50: 725-730
Samotus B, Schwimmer S (1962) Phytic acid as a phosphorus reservoir in the developing
potato tuber. Nature 194: 578-579
Santarius KA, Heber U (1965) Changes in intracellular levels of ATP, ADP, AMP,
and Pi and regulatory function of the adenylate system in leaf cells during photosyn-
thesis. Biochim Biophys Acta 102:39-54
Smith FA (1966) Active phosphate uptake by Nitella translucens. Biochim Biophys Acta
126:94-99
Sugino Y, Miyoshi Y (1964) The specific precipitation of orthophosphate and some
biological applications. J BioI Chern 239:2360-2364
Tsuji H (1964) Acid-soluble phosphate ester contents of developing rice and oat shoots.
Bot Mag 77: 247-252
Ullrich W, Urbach W, Santarius KA, Heber U (1965) Die Verteilung des Orthophos-
phates aufPlastiden, Cytoplasma und Vacuole in der Blattzelle und ihre Veranderung
im Licht-Dunkel-Wechsei. Z Naturforsch 20B:905-910
Ullrich-Eberius CI, Novacky A, Fischer E, Liittge U (1981) Relationship between energy-
dependent phosphate uptake and the electrical membrane potential in Lemna gibba
G 1. Plant PhysioI67:797-801
Walker DA (1976) Plastids and intracellular transport. In: Stocking CR, Heber U (eds)
Transport in plants III. Encyclopaedia of plant physiology new ser, vol 3. Springer,
Berlin Heidelberg New York
Weigl J (1963) Die Bedeutung der energiereichen Phosphate bei der lonenaufnahme durch
Wurzeln. Planta 60: 307-321
Weigl J (1968) Austauschmechanismus des Ionentransports in Pflanzen am Beispiel des
Phosphat- und Chloridtransports bei Maiswurzeln. Planta 79: 197-207
Weste JE, Rowan KS, Chalmers DJ (1974) The distribution of phosphorus-containing
compounds in tomato plants during the development of phosphorus deficiency. In:
Bieleski RL, Ferguson AR, Cresswell MM (eds) Mechanisms of regulation of plant
growth. Bull 12. R Soc NZ, Wellington
Wiskich JT (1977) Mitochondrial metabolite transport. Annu Rev Plant Physiol28 :45-69
Woodrow IE, Rowan KS (1979) Change of flux of orthophosphate between cellular
compartments in ripening tomato fruits in relation to the climacteric rise in respiration.
Aust J Plant PhysioI6:39-46