See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/11210586

Evidence for protection of nitrogenase from O2 by colony structure in the

aerobic diazotroph Gluconacetobacter diazotrophicus

Article in Microbiology · September 2002

DOI: 10.1099/00221287-148-8-2293 · Source: PubMed

CITATIONS READS

25 186

9 authors, including:

Zhongmin Dong Baozhi Pan

Saint Mary's University Jilin University
40 PUBLICATIONS 2,094 CITATIONS 85 PUBLICATIONS 1,069 CITATIONS

SEE PROFILE SEE PROFILE

Randolph S Faustino Joseph Kevin Vessey

University of South Dakota Saint Mary's University
71 PUBLICATIONS 2,027 CITATIONS 83 PUBLICATIONS 6,508 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Hydrogen fertilization View project

Protein Subdomain Enrichment of NUP155 Variants Identify a Novel Predicted Pathogenic Hotspot View project

All content following this page was uploaded by Zhongmin Dong on 14 March 2014.

The user has requested enhancement of the downloaded file.

Microbiology (2002), 148, 2293–2298 Printed in Great Britain

Evidence for protection of nitrogenase from O2

by colony structure in the aerobic diazotroph
Gluconacetobacter diazotrophicus
Z. Dong,1† C. D. Zelmer,2 M. J. Canny,1‡ M. E. McCully,1‡ B. Luit,2 B. Pan,2
R. S. Faustino,3 G. N. Pierce3 and J. K. Vessey2

Author for correspondence : J. K. Vessey. Tel : j1 204 474 8251. Fax : j1 204 474 7528.
e-mail : kIvessey!umanitoba.ca

1 Department of Biology, Gluconacetobacter diazotrophicus is an endophytic diazotroph of sugarcane

Carleton University, which exhibits nitrogenase activity when growing in colonies on solid media.
Ottawa, ON, Canada
K1S 5B6 Nitrogenase activity of G. diazotrophicus colonies can adapt to changes in
2
atmospheric partial pressure of oxygen (pO2). This paper investigates whether
Department of Plant
Science, University of colony structure and the position of G. diazotrophicus cells in the colonies are
Manitoba, Winnipeg, MB, components of the bacterium’s ability to maintain nitrogenase activity at a
Canada R3T 2N2 variety of atmospheric pO2 values. Colonies of G. diazotrophicus were grown
3 Department of Physiology, on solid medium at atmospheric pO2 of 2 and 20 kPa. Imaging of live, intact
University of Manitoba, colonies by confocal laser scanning microscopy and of fixed, sectioned colonies
Winnipeg, MB, Canada
R3E 3J7 by light microscopy revealed that at 2 kPa O2 the uppermost bacteria in the
colony were very near the upper surface of the colony, while the uppermost
bacteria of colonies cultured at 20 kPa O2 were positioned deeper in the
mucilaginous matrix of the colony. Disruption of colony structure by physical
manipulation or due to ‘ slumping ’ associated with colony development
resulted in significant declines in nitrogenase activity. These results support
the hypothesis that G. diazotrophicus utilizes the path-length of colony
mucilage between the atmosphere and the bacteria to achieve a flux of O2 that
maintains aerobic respiration while not inhibiting nitrogenase activity.

Keywords : Acetobacter diazotrophicus, diffusion resistance, N fixation, nitrogen,

oxygen #

INTRODUCTION diazotrophs nitrogenase activity requires substantial

amounts of ATP and reductant derived from aerobic
Gluconacetobacter diazotrophicus (Yamada et al., respiration (Hunt & Layzell, 1993). Diazotrophs there-
1997), formerly Acetobacter diazotrophicus (Gillis et al., fore need to protect nitrogenase from O inactivation by
1989), is an N -fixing bacterium that inhabits inter- regulating intracellular concentrations #of free O . The
cellular spaces #of sugarcane (Dong et al., 1994). An mechanisms by which diazotrophs reduce free # O
unusual feature of this bacterium is the ability to fix N concentrations while permitting aerobic respiration has#
in vitro on semi-solid (Cavalcante & Do$ bereiner, 1988)# been the subject of much research (Robson & Postgate,
and solid media (Dong et al., 1995 ; Pan & Vessey, 2001) 1980 ; van Cauwenberghe et al., 1993 ; Oresnik &
in the presence of relatively high (approx. 20 kPa) partial Layzell, 1994).
pressures of O (pO ) in the atmosphere. The nitro-
genase enzyme # is oxygen
# labile ; however in aerobic Sugarcane does not form any specialized structure to
host G. diazotrophicus (Dong et al., 1994) that may aid
in the regulation of O flux as root nodules do in legume
.................................................................................................................................................
plants (Hunt & Layzell,# 1993). Nitrogenase activity by
† Present address : Department of Biology, St Mary’s University, Halifax,
NS, Canada B3H 3C3. G. diazotrophicus in liquid medium was optimized
‡ Present address : Division of Plant Industry, CSIRO, GPO 1600, Canberra, when the dissolved oxygen content of the medium was
ACT 2601, Australia. equilibrated with 0n2 kPa O in the gas phase (Reis &
Abbreviations : CLSM, confocal laser scanning microscopy ; DAI, days after Do$ bereiner, 1998). However,# G. diazotrophicus is able
inoculation ; pO2, partial pressure of oxygen. to use N as the its sole nitrogen source under 21 kPa O
# #
0002-5400 # 2002 SGM 2293
Z. Dong and others

on a semi-solid medium (Cavalcante & Do$ bereiner, maximum penetration depth of the laser (120 µm). A montage
1988). Under these conditions, distinct colonies are not of the Z-series was produced using Bio-Rad’s Confocal
formed ; rather, the bacteria grow just below the surface Assistant v. 4.02.
of the media. This behaviour may help to optimize the For light microscopy, a freeze-substitution method was used
O flux to the bacterium as seen in other aerotactic because standard fixatives flooded over the agar surface caused
#
diazotrophs (Zhulin et al., 1996). On solid medium, the colonies to rupture. Five-day-old colonies of G. diazotro-
distinct, superficial colonies of G. diazotrophicus form phicus JO-2, supported by small pieces of the subtending LGI-
thick, mucilaginous matrices and are able to grow on N P agar, were plunged into the freezing mixture (isopentane\
as the sole nitrogen source at 20 kPa O (Dong et al.,# methyl cyclohexane, 1 : 1, at the melting point). Freeze sub-
1994). Pan & Vessey (2001) showed #that bacterial stitution in dry acetone resulted in the formation of sucrose
respiration and nitrogenase activity by G. diazotrophi- crystals (from the medium), which damaged the structure of
cus in colonies adapted over long-term exposures (i.e. the colonies. The colonies were therefore freeze-substituted in
methanol\acrolein (10 : 1) at k80 mC for 7 days. Under these
several days) to different atmospheric pO (10, 20 and
30 kPa). Optimal nitrogenase activity by G.#diazotrophi-
conditions, sucrose crystals formed at the bottom of the vial or
on the surface of the agar, from which they were easily
cus colonies occurred at or slightly above (i.e. j10 kPa) removed.
the O concentrations at which they were grown (Pan &
# 2001).
Vessey, Freeze-substituted colonies were gradually warmed (k20 mC
overnight, then j5 mC for 24 h), rinsed (3i10 min each) in
Since nitrogenase is active in G. diazotrophicus colonies methanol on ice, post-fixed with 1 % OsO in methanol for 1 h
%
grown on solid media, the bacterium must have means on ice, then rinsed again with three changes of methanol. The
of ensuring an appropriate concentration and flux of O methanol was replaced by the transition solvent acetone in a
to balance aerobic respiration and nitrogenase activity.# graded series (5, 10, 20, 50, 70, 90 and 100 % acetone ; two
In this study, we test the hypothesis that G. diazotro- 10 min changes per step). Colonies were gradually infiltrated
phicus positions itself within the mucilaginous matrix of in Spurr’s resin monomer mixture (Spurr, 1969), with the
concentration of the resin in acetone reaching 5 % at 90 min,
its colony to achieve an appropriate O environment for
nitrogenase activity, and that an intact# colony structure
10 % at 150 min, 25 % at 210 min and 75 % at 330 min. The
vials were then covered with perforated foil to allow evap-
is required to maintain this nitrogenase activity. This oration of the remaining acetone overnight. The next day,
hypothesis was tested by comparing G. diazotrophicus resin in the vials was replaced with 100 % resin and poly-
colony structure when grown on solid medium under 2 merized at 70 mC overnight. Mid-colony transverse sections
and 20 kPa pO , correlating nitrogenase activity to were cut with glass knives, stained with toluidine blue O
# and observing nitrogenase activity
colony development, (0n05 % in benzoate\borate buffer at pH 4n4), mounted in
in response to disruption of colony structure. immersion oil and viewed with an Olympus Vanox micro-
scope using phase-contrast and brightfield optics. Trans-
mission electron microcopy was performed to confirm that
METHODS stained bodies seen in the light microscopy corresponded to
bacteria (data not shown).
Assessment of colony structure. G. diazotrophicus strain JO-
2 was originally isolated from a Cuban line of sugarcane Observation of nitrogenase activity. Nitrogenase activity was
(Dong et al., 1995) and strain PAL 5 (ATCC 49037) was assayed for developing G. diazotrophicus colonies and for
originally isolated from sugarcane in Brazil (Gillis et al., 1989). mature colonies before and after physical disruption. Nitro-
Colonies were cultured at 30 mC in Petri dishes on a modified genase activity was measured by H evolution in the presence
#
version of LGI-P medium the same as that described in Pan & of Ar\O (Hunt & Layzell, 1993) in a flow-though gas-
#
Vessey (2001) except that sugarcane extract was not added to exchange system (Pan & Vessey, 2001). To assess the effect of
the medium. This medium is free of mineral nitrogen. Colonies physical disruption of colony structure on nitrogenase activity,
were grown for 4–5 days at either 2 or 20 kPa O , and N was G. diazotrophicus PAL 5 was grown on solid, modified LGI-P
# # medium (Pan & Vessey, 2001). At 6 days after inoculation
used to bring the gas blends to 100 kPa (Pan & Vessey, 2001).
At this stage of development, colonies were of similar size in (DAI), 20 Petri dishes containing 100–150 colonies per dish
both pO treatments. were placed in the gas-exchange system. A gas mixture of
# Ar\O (80 : 20) was passed through the chamber at the rate
Colony structure was examined in both live, intact colonies #
of 500 ml min−". Hydrogen evolution of the colonies was
and fixed, sectioned colonies. Cells of G. diazotrophicus recorded after 1 h, then the plates were removed from the
accumulated the pH indicator bromothymol blue from the chamber. The colonies on each plate were gently disrupted by
LGI-P medium. Fluorescence of the pH indicator permitted using a glass rod to smear the colonies on the surface of the
the visualization of cells within live colonies using confocal agar to approximately twice their original surface area. The
laser scanning microscopy (CLSM). A minimum of six plates were returned to the chamber along with the bent glass
randomly selected 4-day-old colonies of G. diazotrophicus rod, and once again exposed to the Ar\O mixture. Hydrogen
PAL 5 from each of the two pO treatments were examined #
# evolution by the disrupted colonies was recorded after 1 h and
using a Bio-Rad MRC600 CLSM equipped with a 514 Argon reported as nmol H h−" per colonyp.
laser utilizing a GHS filter block (514 nm DF excitation\ #
550 nm LP emission). An inverted stage and 32i open air To assess the effect of colony development and morphology
objective were used to view the colonies, which were removed on nitrogenase activity, G. diazotrophicus PAL 5 was inocu-
with a thin layer of subtending agar from the plates on which lated onto Petri dishes containing solid, modified LGI-P
they were grown. Optical sections (Z-series) were initiated at medium (Pan & Vessey, 2001) and incubated at 30 mC. Discrete
the top of the colony mucilage and collected at 5 µm intervals colonies were visible at 2 DAI. Nitrogenase activity of the
down through each colony toward the agar substrate, to the intact colonies was measured daily from 3 to 8 DAI by H
#
2294
 Colony structure of G. diazotrophicus

evolution in the flow-through gas-exchange system. Nitro-

(a)
genase measurements were performed daily on four replicates
of 20 Petri dishes each (100–150 colonies per dish) as described
above. Because of increasing colony size over time, bacterial
titre per colony was quantified and nitrogenase activity is
reported as H evolution rate per cell number (i.e. µmol H per
# #
10"! cells h−"). Concentrations of bacteria per colony were
assessed by plate counting as detailed by Pan & Vessey (2001).
Colonies were visually assessed daily from 3 to 8 DAI for
breakdown of colony structure. Breakdown of colony struc- (b)
ture was indicated by a ‘ slumping ’ of the upper layer of
mucilage to one side of the colony (easily visible due to the
brightly yellow stained bacteria) and a flattening of the colony
profile.

RESULTS AND DISCUSSION

.................................................................................................................................................
Dong et al. (1995) demonstrated that G. diazotrophicus
colonies on solid media have nitrogenase activity at 2 Fig. 2. Mid-colony, transverse sections through fixed colonies of
G. diazotrophicus grown at an atmospheric pO2 of 20 kPa (a)
and 20 kPa O . Further to this, Pan & Vessey (2001)
recently showed# that nitrogenase activity by G. diazotro- and 2 kPa (b) for 5 days. Cells were stained with bromothymol
blue absorbed from the culture medium and viewed using
phicus in colonies adapts to long-term changes in darkfield microscopy. Arrows indicate the level of the upper-
atmospheric pO . The current study supports the most population of bacteria within the colonies. Note that
#
hypothesis that colony structure is important in the the highest density of the uppermost population is deeper in
the colony grown at 20 kPa (a) than in that at 2 kPa O2 (b).

ability of nitrogenase activity by G. diazotrophicus to

(a)
adapt to long-term changes in pO .
#

Influence of pO2 on colony structure

Colonies grown for 4 days under 2 and 20 kPa pO were
#
lens-shaped in transverse section. In both treatments,
two bacterial populations were seen embedded in the
matrix of each colony, one adjacent to the agar medium,
the other in a layer closer to the upper surface of the
colony, with a low density of cells in between the two
(Figs 1 and 2). The toluidine-blue-stained colonies (Fig.
(b)
1) clearly demonstrate that the position of the upper
population varied with pO treatment. At 20 kPa O , the
#
uppermost population of bacteria was midway between #
the surface and the base of the colony (Fig. 1a). In
contrast, the upper population of bacterial cells in
colonies grown at 2 kPa O was positioned just below
the surface of the colony (Fig.# 1b). The direction of the
knife blade sectioning these colonies was from the top of
the colony to the bottom. The vertical striations in these
sections (particularly Fig. 1a) were due to fine crystals of
sucrose remaining in the colony after fixation.
The darkfield images of G. diazotrophicus colonies (Fig.
................................................................................................................................................. 2), at slightly lower magnification than those in Fig. 1,
confirm that pO affects the relative position of the
upper population# of bacteria in the colonies. Bromo-
Fig. 1. Mid-colony, transverse sections through fixed colonies of
G. diazotrophicus grown at an atmospheric pO2 of 20 kPa (a)
and 2 kPa (b) for 5 days. Sections were stained with toluidine thymol blue (displaying bright yellow in colour) is
blue and viewed using phase-contrast microscopy. Arrows accumulated by the bacterial cells, but not the mu-
indicate the level of the uppermost population of bacteria
within the colony matrix. Note that the highest density of the
cilaginous matrix, and clearly indicates that the up-
uppermost population is deeper in the colony grown at 20 kPa permost population of bacteria in colonies grown at
(a) than in that at 2 kPa O2 (b). 20 kPa O were located deeper in the colony matrix (Fig.
#
2295
Z. Dong and others

(a) (b)

.................................................................................................................................................................................................................................................................................................................
Fig. 3. Serial optical sections (Z series) at low magnification down through intact, live colonies of G. diazotrophicus
grown at an atmospheric pO2 of 20 kPa (a) and 2 kPa (b) for 4 days. Reading from left to right and top to bottom, optical
sections start at the colony surface (section 1) and each image of the montage is 5 µm deeper into the colony to a depth
of 100 µm (section 20). Cells were stained with bromothymol blue absorbed from the culture medium and viewed by
CLSM. Because of the fluorescence of the bromothymol blue, the bacterial cells fluoresce white. The relative intensity of
the whitish fluorescence indicates the relative density of cells. The diameter of each colony was approximately 0n8 mm.
Note that the highest density of cells is located 85–100 µm from the surface (sections 17–20) of the colony grown under
20 kPa O2 (a) and only 45–60 µm from the surface (sections 9–12) of the colony grown under 2 kPa O2 (b).

2a) than the uppermost population of the colonies dome-shaped colony. Because the penetration of the
grown at 2 kPa O (Fig. 2b). These images also more CLSM system was limited to just beyond the upper
# population of bacteria at the base
clearly show the lower 100 µm of the colony, the lower population of cells near
of the colonies in both treatments. It is unknown if these the base of the colonies (Fig. 2) could not be imaged.
lower populations of bacteria represent dead or live The microscopic imaging of G. diazotrophicus colonies
cells. Some smearing of stain and vertical striations in (Figs 1, 2 and 3) clearly indicates that the uppermost
the images (especially Fig. 2b) are scratches in the population of bacteria of colonies grown at 20 kPa O is
embedding resin due to fine sucrose crystals, as seen in #
located deeper in the colony matrix than that of colonies
Fig. 1. grown at 2 kPa O . Since nitrogenase activity by G.
Because fixation processes have the potential of disrupt- #
diazotrophicus colonies is known to adapt in the long
ing the structure of what is being observed, live colonies term to changes in atmospheric pO (Pan & Vessey,
grown at 2 and 20 kPa O were also observed by CLSM. #
2001), the microscopic evidence presented here suggests
The difference in location# of the upper population of that the bacteria use the path-length of mucilage between
bacteria between the pO treatments was also evident in the surface of the colony and the site of nitrogenase
#
intact colonies of G. diazotrophicus (Fig. 3). Each panel activity to affect the rate of O diffusion and achieve a
in these images represent a 5 µm optical section through proper flux of O for aerobic # respiration without
a living colony, starting at the top of the dome of the #
inhibiting nitrogenase activity. Bacterial mucilage is
colony (upper left panel) and moving down towards known to decrease the rate of O diffusion to cells
the base of the colony (lower left panel). The whitish (Brown, 1970). The presence of #extracellular poly-
florescence indicates the presence of bacteria due to saccharide surrounding Beijerinckia derxii cells is neces-
laser-induced excitation of bromothymol blue bound sary to maintain nitrogenase in this organism (Barbosa
to bacterial capsular material. For colonies grown at & Alterthum, 1992). Derxia gummosa forms small non-
20 kPa O , the highest density of the upper population fixing colonies if grown at 20 kPa O ; however if grown
# located at a depth of 85–100 µm below the
was typically at 5 kPa O , the bacterium forms# large, highly mu-
highest point of the colony surface (Fig. 3a). At 2 kPa #
cilaginous colonies which fix N (Hill, 1971 ; Hill et al.,
O , the majority of cells of the upper population was 1972). The motile diazotroph #Azospirillum brasilense
#
typically located only 45–60 µm below the top surface of (Zhulin et al., 1996) displays aerotaxis within suspen-
the mucilage (Fig. 3b). The haloing effect seen in the sions to achieve the appropriate O environment for N
sections from 60 to 95 µm (Fig. 3b) shows that this upper fixation. G. diazotrophicus is also# motile (Gillis et al.,#
population of bacteria is following the contour of the 1989) ; however it is unknown at this time whether, if

2296
 Colony structure of G. diazotrophicus

.....................................................................................................
Fig. 4. Time series of a G. diazotrophicus
colony illustrating the morphological
changes in the colony as it develops from
4 to 7 days (d) after inoculation (DAI).
Colonies were grown on solid medium at
30 mC at an atmospheric pO2 of 20 kPa. Note
that by 7 DAI, the colony has ‘ slumped ’ and
breakdown in colony structure is coincident
with a decline in nitrogenase activity (see
text for details).

colonies were switched between 20 and 2 kPa O , the 100

( ímol H2 per 1010 cells h–1 ) Percentage of ‘slumped’ colonies

#
upper population of bacterial cells in colonies would (a)
a
migrate upwards or a new population of bacteria 80
would grow nearer the surface of the colony. Con-
tinuous or near-continuous imaging of intact, living 60 b
colonies over many hours after switching between the
two pO conditions would be necessary to conclusively
# if migration or differential growth of bacteria
determine
40

within the colony were responsible for such a change in

20 c
location of the upper bacterial population.
d

Nitrogenase activity as influenced by physical (b)

disruption of colonies and by colony development
Nitrogenase activity 0·8 a a
Nitrogenase activity of intact G. diazotrophicus colonies
ab
grown at 20 kPa O at 6 DAI was 1n14p0n07 0·6 bc
nmol H h−" per colony. # After physical disruption of c
#
colony structure by smearing colonies across the agar 0·4
surface with a glass rod, nitrogenase activity was
decreased by 96n7 % to 0n038p0n006 nmol H h−" per d
colony. Likewise, breakdown in colony structure # due to 0·2

development\ageing (Figs 4 and 5) also resulted in

declines in nitrogenase activity. 3 4 5 6 7 8
Colony structure of G. diazotrophicus changes as Days after inoculation
colonies develop (Fig. 4). Starting at 3 DAI, the .................................................................................................................................................
accumulation of the pH indicator bromothymol blue by Fig. 5. Proportion of slumped colonies (a) and nitrogenase
the cells from the medium resulted in a disc-shaped, activity (b) of G. diazotrophicus colonies from 3 to 8 DAI.
stained (yellow) area inside the translucent, hemispheri- Colonies were grown on solid media at 30 mC at an atmospheric
cal colony mucilage (Fig. 4, colony images from 4 to 6 pO2 of 20 kPa. Bars representpSEM and values in columns marked
by different letters are significantly different at P  0n05.
DAI). However, as colonies continued to develop, their
structure began to break down, taking on a ‘ slumped ’
morphology (Fig. 4, colony image from 7 DAI). The first
slumped colonies were recorded at 5 DAI, and by 7 DAI
the proportion of collapsed colonies had increased the breakdown in colony structure due to ageing (Fig.
significantly (Fig. 5a). At 8 DAI, about 75 % of the 5a) was coincident with the decline in nitrogenase
colonies on each plate had slumped (Fig. 5a). Breakdown activity per cell within the colonies (Fig. 5b).
of bacterial colony structure with age is not uncommon
Disruption of colony structure due to either manipu-
and may be caused by enzymic depolymerization and
lation or ageing would compromise all spatial relation-
loss of viscosity of the exopolysaccharide matrix of the
ships between the bacterial cells and path-length of
colony (Sutherland, 1999).
mucilage to the open atmosphere. Decline in the path-
Nitrogenase activity of the colonies grown at 20 kPa O length between bacteria and the atmosphere would
(Fig. 5b) was detectable at 3 DAI, although at a very low# result in a dramatic increase in O flux to the bacteria
rate of 0n2 µmol H per 10"! cells h−". Nitrogenase and in the concentration of free# O at the sites of
#
activity increased daily to a maximum value of # demonstrated
nitrogenase activity. Pan & Vessey (2001)
0n697 µmol H per 10"! cells h−" at 6 DAI. After this that G. diazotrophicus displays a rapid switch-off
#
time, nitrogenase activity decreased, declining to only protection phenomenon when O flux rapidly increases
76 % of the maximum at 8 DAI. Hence, the increase in #
to G. diazotrophicus within colonies. Alternatively, a

2297
 Z. Dong and others

rapid increase in O flux to G. diazotrophicus with Brown, D. E. (1970). Aeration in the submerged culture of micro-
#
disruption of colony structural integrity could result in a organisms. Methods Microbiol 2, 125–174.
reversible inhibition of nitrogenase activity (Burris, Burris, R. H. (1991). Nitrogenases. J Biol Chem 266. 9339–9342.
1991). Cavalcante, V. A. & Do$ bereiner, J. (1988). A new acid-tolerant
nitrogen-fixing bacterium associated with sugarcane. Plant Soil
Physical disturbance of the colonies by manipulation 108, 23–31.
with a glass rod resulted in a much greater decline in Dong, Z., Canny, M. J., McCully, M. E., Roboredo, M. R., Cabadilla,
nitrogenase activity (96n7 %) than that caused by the C. F., Ortega, E. & Rodes, R. (1994). A nitrogen-fixing endophyte
‘ slumping ’ of colonies due to ageing between 6 and 8 of sugarcane stems. A new role for the apoplast. Plant Physiol 105,
DAI (Fig. 5b ; a 24 % decline in nitrogenase activity). 1139–1147.
However, this is not unexpected as the smearing of the Dong, Z., Heydrich, M., Bernard, K. & McCully, M. E. (1995).
colony with the glass rod is a much more severe physical Further evidence that the N fixing endophytic bacterium from
#
disturbance than that induced by colony slumping. the intercellular spaces of sugarcane stems is Acetobacter
diazotrophicus. Appl Environ Microbiol 61, 1843–1846.
Gillis, M., Kersters, K., Hoste, B., Janssens, D., Kroppenstedt,
R. M., Stephan, M. P., Teixeira, K. R. S., Do$ bereiner, J. & De Ley, J.
Conclusion (1989). Acetobacter diazotrophicus sp. nov., a nitrogen-fixing
acetic acid bacterium associated with sugarcane. Int J Syst
The results of both the imaging of colonies grown at Bacteriol 39, 361–364.
different pO values and the correlation of nitrogenase Hill, S. (1971). Influence of oxygen concentration on the colony
activity with# colony intactness are consistent with the type of Derxia gummosa grown on nitrogen free medium. J Gen
hypothesis that the mucilaginous matrix of G. diazo- Microbiol 67, 77–83.
trophicus colonies is important in the protection of Hill, S., Drozd, J. W. & Postgate, J. R. (1972). Environmental
nitrogenase activity by the bacterium from excessive O effects on the growth of nitrogen-fixing bacteria. J Appl Chem 22,
flux. Likewise, the position of G. diazotrophicus within# 541–558.
the colony appears to be a component of the bacterium’s Hunt, S. & Layzell, D. B. (1993). Gas exchange of legume nodules
long-term adaptation to changes in pO in the sur- and the regulation of nitrogenase activity. Annu Rev Plant Physiol
rounding atmosphere. However, the actual # concen- Plant Mol Biol 44, 483–511.
tration of free and dissolved O within the colonies at Oresnik, I. J. & Layzell, D. B. (1994). Composition and distribution
the sites of nitrogenase activity #is as yet unknown. Reis of adenylates in soybean (Glycine max L.) nodule tissue. Plant
& Do$ bereiner (1998) found that nitrogenase activity of Physiol 104, 217–225.
G. diazotrophicus in liquid cultures was maximal when Pan, B. & Vessey, J. K. (2001). Response of the endophytic
the culture was at equilibrium with 0n2 kPa O in the gas diazotroph Gluconacetobacter diazotrophicus on solid media
phase. However, the actual concentration of# dissolved to changes in atmospheric pO . Appl Environ Microbiol 67,
#
O at the site of nitrogenase activity in any medium is 4694–4700.
#
dependent upon the concentration of O in the gas Reis, V. M. & Do$ bereiner, J. (1998). Effect of high sugar
phase, the diffusion rate through the medium, # and the concentration on nitrogenase activity of Acetobacter diazotro-
rate of O consumption by bacterial respiration (Hunt phicus. Arch Microbiol 171, 13–18.
& Layzell, # 1993). The current study supports the
Robson, R. L. & Postgate, J. R. (1980). Oxygen and hydrogen in
hypothesis that G. diazotrophicus utilizes the path- biological nitrogen fixation. Annu Rev Microbiol 34, 183–207.
length of colony mucilage between the atmosphere and Spurr, A. R. (1969). A low viscosity epoxy resin embedding
the bacteria to achieve this optimal flux and con- medium for electron microscopy. J Ultrastruct Res 26, 31–43.
centration of O for respiration and nitrogenase activity.
# Sutherland, I. W. (1999). Polysaccharases for microbial exopoly-
saccharides. Carbohydr Polymers 38, 319–328.
van Cauwenberghe, O. R., Newcomb, W., Canny, M. J. & Layzell.
D. B. (1993). Dimensions and distribution of intercellular spaces
ACKNOWLEDGEMENTS in cryoplaned soybean nodules. Physiol Plant 89, 252–261.
This study was funded by grants from Cargill Ltd, Agriculture Yamada, Y., Hoshino, K. & Ishikawa, T. (1997). The phylogeny of
and AgriFood Canada, and the Natural Sciences and En- acetic acid bacteria based on the partial sequences of 16S
gineering Research Council (NSERC) of Canada. ribosomal RNA : the elevation of the subgenus Gluconoaceto-
bacter to generic level. Biosci Biotechnol Biochem 61, 1244–1251.
Zhulin, I. B., Bespalov, V. A., Johnson, M. S. & Taylor, B. L. (1996).
Oxygen taxis and proton motive force in Azospirillum brasiliense.
REFERENCES J Bacteriol 178, 5199–5204.
Barbosa, H. R. & Alterthum, F. (1992). The role of extracellular .................................................................................................................................................
polysaccharide in cell viability and nitrogenase activity of Received 5 December 2001 ; revised 11 March 2002 ; accepted 29 April
Beijerinckia derxii. Can J Microbiol 38, 986–988. 2002.

2298

View publication stats