PhD thesis_Opt

Plant Major Intrinsic Proteins
Natural Variation and Evolution
Jonas Danielson
Lund 2010

Plant Major Intrinsic Proteins
Natural Variation and Evolution
© 2010 Jonas Danielson
All rights reserved
Printed in Sweden by Media-Tryck AB, Lund, 2010
Department of Biochemistry and Structural Biology
Center for Molecular Protein Science
Lund University
P.O. Box 124
SE-221 00 Lund, Sweden
ISBN 978-91-7422-259-3

Abstract
Major Intrinsic Proteins (MIPs, also called Aquaporins, AQPs) are channel forming
membrane proteins. Although initially functionally characterized and named after
their water channeling property in human red blood cells, it has become increasingly
evident that MIPs are present in all types of organisms and transport a variety of
small, uncharged molecules besides water. MIPs have a highly conserved structure
with a constriction region and electrostatic repulsion filter allowing the combination
of high transport rate and selectivity, characteristic of MIPs.
In plants, MIPs form a large and varied protein family, with roughly three times as
many isoforms as found in animals. Even though the abundance implies that MIPs
have important functions in plants, the roles of individual MIPs have so far only
been described for a handful of isoforms. Variations in the filter regions as well as
experimental data, suggest differences in substrate specificities and localization for
different MIP subfamilies. However actual functions of different isoforms remain
largely unknown as traditional knock out/knock down experiments to a large extent
have failed to reveal any clear phenotypes.
Using another approach, we used naturally occurring genetic variants (accessions) of
the model plant Arabidopsis thaliana to see if differences in traits, such as drought
tolerance, could be linked to differences in MIP regulation. We also looked at the
evolution of the MIP family in plants, to see if this could for example be linked to
events such as the emergence of a vascular system or the development of a terrestrial
lifestyle. Therefore we identified the whole MIP family in the moss Physcomitrella
patens and in nine different green algae and investigated how these relate to those of
higher plants. An amazing diversity of MIPs was found and surprisingly some of the
higher plant subfamilies were present also in these simple plants.
Keywords: MIP, aquaporin, plant, natural variation, evolution, Arabidopsis,
Physcomitrella patens, NIP, XIP

List of Publications
This thesis is based upon the following papers, which in the text are referred to by
their Roman numerals:
I. Unexpected complexity of the Aquaporin gene family in the moss
Physcomitrella patens.
Danielson JÅH and Johanson U
BMC Plant Biology, 8:45, 2008
II. Transcriptional regulation of aquaporins in accessions of Arabidop-
sis in response to drought stress.
Alexandersson E, Danielson JÅH, Råde J, Moparthi VK, Fontes M,
Kjellbom P and Johanson U
Plant Journal, 61(4):650-60, 2010
III. Phylogeny of major intrinsic proteins.
Danielson JÅH and Johanson U
Advances in Experimental Medicine and Biology, 679:19-31, 2010
IV. Algae MIPs.
Danielson JÅH, Anderberg H and Johanson U
Manuscript
Papers II and III are reprinted with kind permission of John Wiley & Sons (II),
Springer Science and Business Media (III) and Landes Bioscience (III).

Additional Publications
Persistent LTP without triggered protein synthesis.
Abbas AK, Dozmorov M, Li R, Huang FS, Hellberg F, Danielson JÅH, Tian Y, Ekström J,
Sandberg M, Wigström H.
Neuroscience research, 63(1):59-65, 2009
Fibrinogen gene variation and ischemic stroke.
Jood K, Danielson JÅH, Ladenvall C, Blomstrand C, Jern C.
Journal of thrombosis and haemostasis, 6(6):897-904, 2008

Contribution to the Papers
I. JD took part in designing the study. JD did the acquisition, analysis and
interpretation of data and drafting of the manuscript. JD took part in
revising the manuscript.
II. JD took part in all the experiments except the GUS experiment. JD took part
in analyzing the data and drafting of the manuscript. JD took part in revision
of the manuscript
III. JD took part in designing the study. JD did the acquisition, analysis and
interpretation of data. JD took part in drafting and revising the manuscript.
IV. JD took part in designing the study. JD did the acquisition of data. JD took
part in the analysis. JD took part in the drafting of the manuscript.

Table of Contents
Abbreviations 8
1 Introduction 9
2 Plants 11
2.1 The Plant Cell 12
2.2 Water Transport in Plants 13
3 Major Intrinsic Proteins 15
3.1 Phylogeny 16
3.2 Structure 17
3.3 Substrate Specificity 20
3.4 MIP Regulation 22
4 Plant MIPs 25
4.1 The Land Plant MIP Family 26
4.2 Subfamilies 29
4.2.1 PIPs 29
4.2.2 TIPs 32
4.2.3 NIPs 33
4.2.4 SIPs 35
4.2.5 GIPs 36
4.2.6 HIPs 36
4.2.7 XIPs 36
4.3 Evolution 37
4.4 Physiological Role 42
5 Conclusions and Future Perspectives 43
Populärvetenskaplig sammanfattning 45
Acknowledgements 47
References 48

Abbreviations
AQP Aquaporin
ar aromatic
bb backbone
cRNA copy RNA
DNA Deoxyribonucleic Acid
ER Endoplasmic Reticulum
EST Expressed Sequence Tag
FRET Fluorescence Resonance Energy Transfer
GFP Green Fluorescent Protein
GIP GlpF‑like Intrinsic Protein
GlpF Glycerol Uptake Facilitator
HGT Horizontal Gene Transfer
HIP Hybrid Intrinsic Protein
MIP Major Intrinsic Protein
miRNA micro RNA
NIP NOD-26 like Intrinsic Protein
NJ Neighbor Joining
PBM Peribacteroid Membrane
PIP Plasma membrane Intrinsic Protein
PM Plasma Membrane
RNA Ribonucleic Acid
RNAi RNA interference
sc sidechain
SDS-PAGE Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis
SIP Small and basic Intrinsic Protein
TIP Tonoplast Intrinsic Protein
TM Trans Membrane
WUE Water Use Efficiency
XIP X Intrinsic Protein
YFP Yellow Fluorescent Protein
8

1 Introduction
In a way, life is confinement, as all life as we know it is cellular, existing only in
that unit of space, the cell. Membranes, consisting of lipid bilayers, define not only
the cell but also many of the organelles within the cell and effectively isolate what
is inside from the outside. This makes it possible to create the controlled environ-
ment necessary for all the chemical reactions that is life to occur. However, having a
lipid bilayer only confines the space, and in order to make it controlled, transports
across the membrane must be regulated. In cells, transports over the membranes are
maintained by three different kinds of proteins, pumps that use energy to actively
transport substances, channels that facilitate the passive diffusion of substances over
the membrane and carriers that can either be active or passive transporters.
For a long time it was believed that cellular water transport occurred entirely by passive
diffusion through the lipid bilayer, and that differences in permeability between
different membranes were depending only on differences in lipid composition. In
the 1950s researchers did experiments showing that membranes from certain tis-
sues (e.g. kidney and red blood cells) were not only highly permeable to water, but
also that this water transport could be blocked in a reversible way by mercury (Stein
and Danielli, 1956). This indicated that proteins were somehow involved in the
permeabilities of these membranes, but how this occurred and which proteins were
involved would not be revealed until almost 40 years later.
Major Intrinsic Proteins (MIPs, also referred to as aquaporins, AQPs) is a group of
channel forming proteins of ancient origin, predating the emergence of the three
domains of life (Pao et al., 1991, Park and Saier, 1996, Kozono et al., 2003). Although
identified early on as major constituents of cellular membranes, they first gained
massive attention in the early 1990s when CHIP28 (a MIP later renamed AQP1)
was identified as the long sought water channel protein (Preston et al., 1992). This
did not only solve the old ongoing debate on the involvement of proteins in cellular
water transport, but also resulted in a burst of MIP related research. It turns out
MIPs are not only water channels, but rather they are involved in transport of several
other small uncharged molecules (e.g. glycerol, urea, silicic acid, ammonia, hydrogen
peroxide) across cellular membranes (which is why the use of MIPs is preferred over
the more catchy, but misleading, AQPs in this thesis).
9

The first plant MIP was identified in 1987 (Fortin et al., 1987) as a membrane protein
of root nodules, but it was not until 1993 (Maurel et al., 1993) that a plant MIP was
for the first time being experimentally determined to be a water channel. Since then,
numerous different plant MIPs have been identified and the transport properties have
been experimentally determined for several of them. It turns out that plants contain
a very large MIP family with more than 30 isoforms, almost three times as many as
found in mammals. The abundance of isoforms suggests that MIPs have important
roles in plant life, but largely the functions of the different subfamilies and individual
isoforms is still unknown.
It was in this context that the subject of my doctoral project was formulated, to expand
on the knowledge of plant MIPs, the diversity of the superfamily, the roles of different
subfamilies and isoforms. In my project I used both traditional molecular biology
approaches (such as GFP/GUS tagging of proteins, knock out/knock down plants,
transcriptional profiling and so on) and more genetics based approaches, looking
at differences between naturally occurring variants (accessions) in one species, or at
variation between more distantly related species.
10

2 Plants
Even though water is essential for all living organisms, it plays an extremely large and
crucial role in the life of plants. For the process of photosynthesis, plants have to take
up carbon dioxide from the air, but in doing so they also lose water, and they lose a
lot of water. For every gram of matter produced by photosynthesis, approximately
500 g of water has to be absorbed by the roots, transported throughout the plant and
lost to the atmosphere (Taiz and Zeiger, 2006). Water transport in the plant is passive
in the way that it is driven by this transpiration, and water uptake and transport
throughout the plant is a spontaneous movement of water from regions of higher to
lower free energy. This, together with the crucial role of water and the fact that plants
are immobile, makes it of utmost importance for plants to be able to regulate water
flow in response to environmental conditions.
Water has many different functions in the plant, at a cellular scale it is the solvent
of the cell and therefore influencing everything from chemical reactions to protein
stabilities, but at a macroscopic level it also constitutes the plant long-distance transport
system, being both the transporter of mineral nutrients from roots to shoots in the
xylem, and the transporter of sugars made in the photosynthesis from shoot to root
in the phloem. Water also has a fundamental role in the very first steps of photo-
synthesis in plants, where water is oxidized to oxygen by the use of light, oxygen
that ultimately all organisms depending on aerobic respiration (such as ourselves)
depend on for energy conversion.
11

2.1 The Plant Cell
There are numerous unique features of the plant cell compared to that of other
eukaryotes, such as animals. One of the more striking is that the plant cell is enclosed
by a cell wall, a rigid cellulosic structure giving strength and toughness but also restrict-
ing morphological changes of the mature cell. The cell wall allows the plant cell to
build up a positive hydrostatic pressure (called turgor or turgor pressure), pressing
the plasma membrane (PM) of the cell to the wall, resulting in among other things
further rigidity. This hydrostatic pressure is mainly created by large water filled vacuoles
that occupy up to 95 % of the total cell volume of mature plant cells (Taiz and
Zeiger, 2006). By actively accumulating solutes in the vacuole, an osmotic gradient
is created over the tonoplast (the vacuolar membrane), resulting in water uptake and
increased turgor of the cell.
Even though the vacuole occupies a large volume of the plant cell, the remaining part
of the cytosol is highly compartmentalized, containing many different organelles
consisting of at least one membrane type as seen in Figure 1. Most plant cells are
interconnected via plasmodesmata, microscopic tubular extensions of the endoplas-
mic reticulum (ER) surrounded by PM that traverse the cell walls of adjacent cells.
12
CELL WALL
PLASMA MEMBRANE
CHLOROPLAST
MITOCHONDRIA
CYTOPLASM
VACUOLE
GOLGI
PLASMODESMATA
ENDOPLASMIC RETICULUM
NUCLEUS
OUTER
INNER
OUTER
INNER
THYLAKOID
TONOPLAST
MITOCHONDRIAL
MEMBRANES
CHLOROPLAST
MEMBRANES
VACUOLAR
MEMBRANE
Figure 1. Schematic picture of the plant cell.Top part shows the whole cell with different organelles.
Below are enlargements of three different organelles displaying the naming of their membranes.

2.2 Water Transport in Plants
Water can move through the plant via three different pathways: the apoplastic, the sym-
plastic and the transmembrane pathway, illustrated in Figure 2. The apoplast consists of
the continuous system of cell walls, intercellular airspaces and lumens of dead cells (e.g.
xylem and fibers) found in the plant and in the apoplastic pathway, water moves in the
plant without having to cross any plasma membranes. In the symplastic pathway water
moves through the symplast, that is the network of cell cytoplasm interconnected by
plasmodesmata. In order for the water to enter and exit the cytoplasm it needs to pass the
plasma membrane, but when it moves in the symplast, no membranes need to be crossed.
In the third, transmembrane, pathway water moves from cell to cell via the cell wall and
therefore needs to cross the plasma membrane at least twice for each cell it passes. Most
likely the relative importance of the three pathways differs in different tissues and species
and although this has not yet been clearly established some general implications can be
seen. For example the existence of the casparian strip, a radial band of cell walls impreg-
nated with the hydrophobic substance suberin, in roots prevent water movement in the
apoplast and force water to cross the plasma membrane of these cells in order to reach the
vascular system of mature roots. This enables plants to regulate water permeability of the
root at a local level by changing the permeability of the plasma membrane at the caspar-
ian strip, thereby affecting the water uptake of the root system globally.
13
VASCULAR TISSUE
SPONGY MESOPHYLL
WATER
WATER VAPOR
CARBON DIOXIDE
STOMATA
SYMPLASTIC PATHWAY
ROOT HAIR
APOPLASTIC PATHWAY
TRANSCELLULAR PATHWAY
VASCULAR TISSUE
CASPARIAN STRIP
Figure 2. Water Transport Pathways in plant. The Upper picture shows water flow in the leaf,
from the xylem to the atmosphere via the stomata and the countercurrent carbon dioxide flow.
The lower part shows a cross section of a root with the three water transport pathways and the
location of the vascular tissues and casparian strip.

Water flux through the plants can also be regulated at the leaf level, where the aperture
of the openings to the atmosphere can be regulated (see Figure 2). These openings,
stomata, are composed of two guard cells, forming a pore between them. The guard
cells have a special cell wall organization that allows them to change shape depending
on the turgor pressure, resulting in stomatal opening/closure. This allows the plant
to change the rate of transpiration to environmental cues such as light intensity,
temperature, leaf water status and CO2
concentrations. In a similar way, other cells
can also change shape depending on water uptake/loss leading to changes in cell tur-
gor influencing for example cell elongation in newly formed cells or leaf movement.
Some of these changes are quite fast, requiring water fluxes across the plasma mem-
brane and the tonoplast membrane that cannot be explained by the permeability of
the tonoplast and plasma membrane on their own.
14

3 Major Intrinsic Proteins
Major Intrinsic Proteins (MIPs) are integral membrane proteins that form channels,
permeable to water and other small uncharged molecules, in cell membranes. Unlike
membrane transport-proteins that take part in directional, energy demanding, active
transport of substrates, MIPs form discriminating channels which affect the perme-
ability of the membrane for certain substrates and therefore the rate of passive transport
of these.
Even though it was shown as early as in the mid 1950s that diffusion of water across
the lipid bilayer of cellular membranes was facilitated by some membrane compo-
nent (Stein and Danielli, 1956), it was still very much the common view that passive
diffusion across membranes was the only cellular transport mechanism for water,
when in the early 1990s the first water channel protein, Aquaporin-1 (AQP1) was
reported (Preston et al., 1992). This work, carried out in Peter Agres group, started
out by the identification of an unknown protein in red blood cells in the late 1980s
(Smith and Agre, 1991, Denker et al., 1988, Preston and Agre, 1991) and ultimately
lead to the publication of the 3.8 Å resolution structure of a water channel MIP in
2000 (Murata et al., 2000). These results changed the view of how water was trans-
ported in the cell and the impact of this research lead to the Swedish Royal Academy
of Science awarding Peter Agre with a Nobel Prize in chemistry 2003 for “discoveries
concerning channels in cell membranes”.
This was in fact not the first MIP known, it was not even the first MIP believed to
transport something, but it was the first MIP believed and proven to transport water.
Prior to this other MIPs had been postulated to transport glycerol in E. coli and small
molecules in soybean root nodules (Richey and Lin, 1972, Sandal and Marcker,
1988, Baker and Saier, 1990), but it was really the water channeling properties that
got the attention and lead to the boom in MIP related research in the early 2000s.
Even though the water transporting properties of MIPs was for a long time the main
focus of research, lately MIPs have been showntobepermeabletomanyothersubstrates
and the importance and role of this substrate diversity is a growing research field.
15

3.1 Phylogeny
In order to evaluate the evolutionary relationship, or phylogeny, of genes or proteins
their corresponding DNA or amino acid sequences must be known. Today this is
rather easily achieved, but in the days before automated sequencing this was very
labor intensive. It is therefore not surprising that, even though MIPs were early on
discovered as major constituents of certain membranes, it was not until in 1984 that
the first sequence of a MIP, isolated from bovine lens fiber cells, was reported (MIP-
26, later renamed to AQP0) (Gorin et al., 1984). This was followed in 1987 by the
sequence of another MIP protein isolated from the peribacteroid membrane (PBM)
of soybean root nodules (Nod-26, later NIP1;1) (Fortin et al., 1987). At the time
it was not realized that these two proteins were related to each other, but in 1988
two groups, independently of each other, found the two sequences to be surprisingly
similar (Shiels et al., 1988, Sandal and Marcker, 1988). Sandal and Marcker also
concluded that the proteins contained 6 hydrophobic regions large enough to be
membrane spanning and suggested that Nod-26 might transport small molecules
across the PBM by forming channels. One year later the sequence of E. coli glycerol
uptake facilitator (GlpF) was released (Muramatsu and Mizuno, 1989) and the fol-
lowing year it was realized that also this, bacterial, protein belonged to the same
ancient super family of proteins, the MIP super family (Baker and Saier, 1990).
In 1991 the first phylogenetic analysis of MIPs was presented (Pao et al., 1991), but
since only six MIPs were known at the time, the result was rather ambigous. This
was however more clear in the analysis of 52 MIPs presented five years later (Park
and Saier, 1996). In this, 12 subfamilies could be seen and based on the finding that
some prokaryotes had two types of MIPs, it was postulated that all MIPs originate
from two ancestral prokaryotic genes, giving rise to the GLP and AQP cluster.
The most thorough phylogenetic analysis up till now was presented in three articles
by Zardoya and coworkers between 2001 and 2005 (Zardoya, 2005, Zardoya et al.,
2002, Zardoya and Villalba, 2001). They found that MIPs cluster in 8 major groups
(GLPs, animal AQPs, bacterial AQPs, PIPs, TIPs, NIPs, SIPs and AQP8s) but also
that the relation between these groups could not be resolved.
Since then many more groups have been discovered, but still the phylogeny between
these groups remain obscured. It is therefore worth noting that, from a phylogenetic
point of view, the commonly used grouping of MIPs into AQPs and GLPs is not cor-
rect. For even though the GLPs form a phylogenetically well defined, monophyletic,
group, the AQP cluster is in fact only “everything that is not a GLP”, and different
groups within this cluster might not be more closely related to each other than they
are to the GLPs. The present phylogeny of some major MIP subfamilies is presented
in Figure 3.
16

3.2 Structure
Not only the primary structure (as reflected in the phylogeny), but also the second-
ary, tertiary and quaternary structure of MIPs are highly conserved. Until now (Nov
1st
2010) high resolution structures of 15 different MIPs have been solved, all sharing
the same overall structure with some minor variations.
The MIP structure can be seen in Figure 4 and consists of six transmembrane (TM)
helices (H1 to H6) connected by five loops (LA to LE) and is situated in the mem-
brane so that both the amino and the carboxy termini are at the cytoplasmic side. In
loop B and E there are two “half-membrane-spanning” helices (HB and HE), which
are inserted from opposite sides, making up a seventh transmembrane “broken” helix
in the structure.
17
0.2
AQP8S
ANIMAL
AQPS
ARCHAEAL
MIPSAQPZS
NIPS
ALGAL
MIPS
HIPS
TIPS
FUNGAL
AQPS
PIPS
XIPS
”SUPERAQUAPORINS”
GLPS
SIPS
AQP
11&12
BACTERIA
ARCHAEA
ANIMAL
PROTIST
FUNGI
PLANT
ALGAE
Figure 3. An overview of the phylogeny of the MIP family. A Neighbor Joining (NJ) tree show-
ing the phylogeny of some major MIP subfamilies. Species of MIP sequences are indicated by
different line styles and subfamilies are shown by bracketing.The gray shading in the central
part of the tree illustrates the uncertainty of the phylogeny of this region indicated by low
bootstrap supports in the original analysis.

When looking at the primary and secondary structure of MIPs it is clear that they are
built up from two repeats, each consisting of 2 TM helices followed by a half TM helix
and then a final TM helix. In an analysis by Pao and coworkers (Pao et al., 1991) the
relationship between the repeats was investigated and it was shown that, for all MIPs
in the analysis, the first halves were more closely related to each other than to any of
the second halves and vice versa, indicating a very ancient origin of the repeats and the
gene duplication leading to this MIP structure.
The helices are inserted in the membrane, tilted at a slight angle and in a right-hand
fold, giving rise to what is often referred to as the hourglass fold, with two conical
vestibules meeting with their tips in the middle of the protein. However, the narrowest
region of the pore is actually 8 Å more to the extracellular side, where four amino acids
from H2, H5, LE and LE make up the aromatic/arginine (ar/R) filter, a size restric-
tion region largely responsible for the substrate specificity of the MIP. Another region
of utmost importance for MIP function is found in the middle of the pore where two
highly conserved NPA box motifs are found at the ends of HB and HE. The positions
of the ar/R filter and the NPA boxes can be seen in the stereo pictures in Figure 5.
18
H1
REPEAT 1
SIDE
REPEAT 2
TOP
LA LB LC LD LE
H2 HB H3 H4 H5 HE H6
Figure 4. An overview of the MIP Structure. Upper part shows the secondary structure of MIPs,
arrows indicate helices. Naming of helices and loops, as well as the positions of the repeats, are indi-
cated. Lower part shows the three-dimensional structure of MIPs from side and top, colored accord-
ing to the secondary structure cartoon above (The structure is human AQP4, PDBID 3GD8).

19
Figure 5. Stereo pictures showing the ar/R filter and NPA motifs. The MIP (HsAQP4, 3GD8)
is shown in a green cartoon representation with the ar/R filter and the NPA box motifs shown
as sticks in magenta and blue respectively. The upper part shows a top view (from extracellular
side) while the lower part shows a side view were parts of helices 4 and 6 are transparent to
make the ar/R and NPA regions more visible. The red dots show the positions of waters in the
crystal structure.

3.3 Substrate Specificity
Even though some MIPs transport only water, others have physiological roles trans-
porting a whole range of other solutes (reviewed in Carbrey and Agre, 2009). Glyc-
erol and urea were among the first substrates shown for MIPs (such as the aquaglyc-
eroporins AQP 3, 7, 9 and 10), but other substrates include anions such as nitrate
(AQP6), ammonia and hydrogen peroxide (AQP8). Some MIPs have been shown
to be permeable to the gases carbon dioxide and nitrous oxide, but the physiological
relevance for this is still debated. Recently, MIPs have also been shown to be perme-
able to metalloids (e.g. B, Si, Sb, As) and the physiological importance of this can for
example be seen in the interesting case of the protozoan parasites of the genus Leish-
manias. These parasites cause the disease leishmaniasis, a disease usually treated with
antimony and arsenite. It was recently shown that some drug resistant variants have
increased arsenite tolerance, achieved by them altering their MIP expression (Lin et
al., 2008). Metalloid transport is also of great importance in the context of arsenic
contamination of ground water which is a major health problem for hundreds of
millions of people around the world, threatened either directly through contami-
nated drinking water, or indirectly through crops grown in farmlands irrigated with
contaminated water.
20
+
+ +
H
H
O H
H
O H
H
O H
H
H
O H
H
O H
H
O H H
H
O H
H
O H
H
O H
Figure 6. A schematic picture of the Grotthuss mechanism showing how proton movement in
water can be achieved by rearrangement of bonds.
Proton pumping across membranes is a fundamental part of energy production in
all organisms and it is therefore crucial that transport proteins in these membranes
do not disrupt this proton gradient. Proton movement in bulk water is extremely
fast, due to the “proton wiring” effect of the Grotthuss mechanism (see Figure 6 and
Cukierman, 2000), and for quite some time after the water transport properties of
MIPs were first reported (Preston et al., 1992) it was puzzling how the pore could let
water through and at the same time block the passage of protons. The answer turned
out to have two parts. First, hydrogen bonding between the water molecules is bro-
ken when the water, which runs through the pore in a single file, is forced to reorient
at the NPA boxes. Secondly, a positive charge at the NPA boxes, created by the dipole
moment of the half α-helices, results in an electrostatic repulsion of the positively
charged protons, effectively preventing them from passing, see Figure 7.

As previously mentioned, MIPs are per-
meable to a wide range of substrates
(water, ammonia, urea, glycerol, arsen-
ite, antimony to name a few) and Froger
and coworkers early found out that
amino acids at five different positions
in MIPs seemed to be indicative of sub-
strate specificity for two groups of MIPs,
transporting either water or small neutral
solutes. These positions, called P1 to P5,
where identified by their conservation
in amino acid alignments (Froger et al.,
1998). However, when high-resolution
structures became available it became
obvious that the P1 to P5 positions could
not interact with any substrate directly
and therefore perhaps did not have any
functional role, but rather the invariabil-
ity is just reflecting a closer evolutionary
distance of MIPs with the same substrate
specificity. In the structures, an intricate
hydrogen-bonding network across the
pore could be seen and it also became apparent that there was a size constriction
region of importance for substrate specificity, the ar/R filter (Sui et al., 2001, Murata
et al., 2000, Fu et al., 2000). The ar/R filter consists of four amino acids, encircling
the pore of the MIP. Whereas three of the amino acids (H2, H5 and LE2) have their
side chains facing the pore, the fourth (LE1) interacts with the substrate through the
backbone carbonyl group thereby allowing a greater variation at this position. The
effects of the side chains of the other amino acids have been evaluated by point mu-
tations in AQP1, showing that water selectivity can be modulated to allow ammo-
nia, urea and glycerol transport just by changing the ar/R filter (Beitz et al., 2006).
But, even though the ar/R filter is important for substrate specificity, it is not solely
responsible for it, as was shown in a study where a water permeable PIP was point
mutated to mimic ammonia and urea transporting TIPs and NIPs, requiring not
only point mutations in the ar/R filter but also at an extra position (Dynowski et al.,
2008a). In a recent paper Savage and coworkers (Savage et al., 2010) sequentially
engineered the three signature amino acids of glycerol conducting GLPs into the
water transporting AqpZ from Escherichia coli. Crystal structures and transport rate
experiments showed that mutants had a pore size similar to that of wild type AqpZ
and that even though water transport was diminished, no glycerol transport could
be detected. This suggests that the ar/R filter region is important for water transport
21
HB
+
-
HE
+
-
o=c
o=c
o=c
N
o=c
o=c
o=c
N
Figure 7. Schematic picture showing the reorien-
tation of water molecules in the pore. On the left
side are the half TM helices, showing how their
dipole moments result in the positive charge of
the electrostatic repulsion filter. Grey lines show
how the water molecules hydrogen bond to each
other and to the backbone. A reorientation at the
NPA boxes breaks this chain of interactions.

(and possibly other small molecules) whereas transport of larger molecules such as
glycerol seem to depend on channel cross-section size. There have also been several
simulation studies supporting both the role of the NPA boxes and the importance of
the ar/R filter and to various degrees explaining the experimental permeabilities seen
(see for example Hub and de Groot, 2008, Phongphanphanee et al., 2010).
Mainly three types of transport assays have been used for experimental determina-
tion of MIP permeation; the oocyte system, stopped flow measurements on proteo-
liposomes and yeast complementation assays. In the oocyte system, eggs from the
frog Xenopus laevis are injected with cRNA encoding the MIP under investigation.
After the eggs have been given time to express the protein (which hopefully localizes
to the plasma membrane) they are either transferred to hypo/hyper osmotic media
to observe swelling/shrinking, or they are transferred to a media containing labeled
substrate to observe uptake of this. The proteoliposome assay is also based on shrink-
ing/swelling, but in this case, purified protein is reconstituted in lipid vesicles (lipo-
somes). Due to the small size of liposomes (and therefore the high surface/volume
ratio) the swelling/shrinking is very fast. Measurements therefore have to be made
in a stopped flow spectrophotometer, ensuring rapid mixing and fast measurements.
Finally, complementation assays in yeast has proven to be useful for identifying a
range of substrates, some of which would have been difficult to measure with the
other methods. The rationale here is to look at substrates that are either beneficial or
detrimental for the yeast and by deleting a naturally occurring transporter, making
the yeast less or more responsive to the substrate. Different MIPs are then expressed
in the deletion mutant and MIPs transporting the substrate can be identified by
identifying clones with restored wild-type sensitivity.
3.4 MIP Regulation
To only increase the permeability of a membrane for a certain substrate might not
be very useful for an organism, but to be able to regulate the permeability gives the
organism the possibility to adapt to different conditions, which is beneficial. It is
therefore perhaps not very surprising that MIP mediated permeabilities can be regu-
lated in many different ways.
At a supracellular level, expression of MIPs can be controlled by different transcrip-
tion factors and mRNA stability can be affected by different miRNAs. This is likely
to be part of the explanation to why some species have so many MIP isoforms, but
it has also turned out that posttranslational regulation of MIP activity is common,
making the regulatory possibilities far more complex than can be achieved simply by
a large number of MIP genes.
22

Posttranslationally, MIPs can either be regulated at a structural level, where alteration
in a protein structure mediates a change in permeability of that protein, or at a local-
ization level, where permeabilities of membranes are affected by changes in the sub-
cellular localization of MIPs. MIP gating (opening and closing of the pore) has been
shown to occur for both animal, plant and yeast MIPs (Gonen et al., 2004, Harries
et al., 2004, Tornroth-Horsefield et al., 2006, Fischer et al., 2009) and to respond to
a whole range of cues such as pH, divalent cations and phosphorylation. It has also
been suggested that there is a “solute gating” mechanism operating by a cohesion/
tension mechanism, in which solute exclusion from the pore leads to negative pres-
sures in the channel resulting in a collapse of the MIP channel (Ye et al., 2004).
Some MIPs have been shown to have a polar localization in the cell. This allows
MIPs to work in series with active transporters, coupling passive diffusion to active
transport of substrates. MIPs have also been shown to exhibit a dynamic subcellular
localization where MIPs translocate from vesicles or the ER to the membrane upon
some signal. This has for example been shown for AQP2, where cyclic AMP acti-
vates protein kinase A, which in turn phosphorylates AQP2 residing in vesicles, then
fusing with the cell membrane (reviewed in Nedvetsky et al., 2009). Finally, MIP
abundancy can be regulated by degradation and it has been shown that MIPs can
be targets for E3 ubiquitin ligases, suggesting that targeted degradation via the 26S
proteasome is an additional regulatory pathway (Lee et al., 2009).
23

4 Plant MIPs
The diversity and abundance of MIPs found in plants suggests that they have
important functions in whole plant water and nutrient status. However, it was not in
a search for proteins involved in these functions that MIPs were first identified, but
rather their discovery was a consequence of them being major constituents of differ-
ent membranes (see Figure 8). As previously mentioned, one of the very first MIPs
identified was isolated from the peribacteroid membrane of soybean root nodules,
but plant MIPs were also identified as major components in plasma membranes and
tonoplast (the vacuolar membrane) (e.g. Kjellbom and Larsson, 1984, Johansson et
al., 1996). A MIP from tonoplast, Arabidopsis γ-TIP, was the first plant MIP dem-
onstrated to have water-channel activity. This experiment was published just a year
after the report on the first water channel from Peter Agres group, using the same
technique with the oocyte system (Maurel et al., 1993). Since then, a great number
of plant MIPs have been shown to be permeable to water, but also to a whole range of
other substrates including glycerol, carbon dioxide, urea, ammonia, different metal-
loids and hydrogen peroxide.
Plant MIPs are primarily expressed in
two types of tissues. One of these is the
vascular tissue, where plant MIPs are
expressed in cells in or surrounding the
vascular bundle and likely take part in
transport to and from the phloem and
xylem. The other is in cells where regu-
lated rapid water flows occur, such as in
elongating cells, guard cells of the sto-
mata, motor cells controlling leaf move-
ment or in seeds (Maurel et al., 2008).
25
97
66
43
31
22
PM
A B C
Figure 8. SDS-PAGE gel picture of total PM
proteins from poplar. The arrow points at the
band identified as a MIP. Poplar PM was isolated
from xylem (A), cambium/phloem (B) and leaf
(C). (Kindly provided by Katja Bernfur).

4.1 The Land Plant MIP Family
Already in one of the first phylogenetic analysis of the 18 MIP sequences known at
the time two subfamilies of plant MIPs were identified, the tonoplast intrinsic pro-
tein (TIP) and the NOD, later renamed to nodulin-26 like intrinsic protein (NIP),
subfamilies (Reizer et al., 1993). During the following years, several more phylo
genetic analyses were published and it soon became clear that plants also had a third
group of MIPs, the plasma membrane intrinsic protein (PIP) subfamily, and that
these three subfamilies were specific for plants (Kammerloher et al., 1994).
Even though it was already in these early analyses obvious that plants contained
many different isoforms of MIPs, belonging to several different subfamilies, the com-
plete identification of all MIPs within one species was just not possible until when
the first plant genome became available in the year 2000 (The Arabidopsis Genome
Initiative, 2000). When this important identification was done in 2001, it did not
only became clear that plants had nearly three times as many isoforms as there were
in mammals (Arabidopsis has 35 MIPs compared to 13 in humans), but also a fourth
“new” subfamily of plant MIPs was discovered, the small and basic intrinsic pro-
teins (SIPs) (Johanson et al., 2001). The largest subfamily was the PIPs (with 13
isoforms), followed by the TIPs and the NIPs (10 and 9 members respectively) and
the smallest was the SIPs (with only 3 isoforms). It was also evident that within the
subfamilies there were MIPs with higher degrees of similarity, forming subgroups,
and when naming the MIPs they were named after MIP subfamily, subgroup num-
ber and number within subgroup (i.e. PIP1;1, belongs to the PIP subfamily, to the
first subgroup of PIPs and is the first MIP in that subgroup). The identification of
all Arabidopsis MIPs and their phylogeny forms a framework for classification of
other plant MIPs and its usefulness is made clear by the fact that, since then, several
other higher plant species have been shown to contain a comparable number of MIP
isoforms (e.g. 33, 36 and 37 MIPs in rice, maize and tomato respectively) with a
division into subfamilies and subgroups consistent with that of Arabidopsis (Sakurai
et al., 2005, Chaumont et al., 2001, Sade et al., 2009). Recently some higher plants
with very large number of MIPs have been reported (55 and 71 for poplar and cotton
respectively) (Park et al., 2010, Gupta and Sankararamakrishnan, 2009), however
this is also consistent with the Arabidopsis findings, since poplar have gone through
a recent whole genome duplication and cotton is polyploid.
Interestingly, even though the total number of MIPs is rather conserved in higher
plants, it does not seem possible to find specific pairs of corresponding genes (or-
thologs) of MIPs in different species. At the same time, all higher plants seem to
contain the same subfamilies and to a large extent also the same subfamily groups.
Taken together these findings implies that there are certain restrictions for the sub-
families and subfamily groups, but that apart from these constraints, MIPs are rap-
26

idly evolving. It also suggests that the specific functions of subfamilies and subgroups
are conserved in all these plants, hence the conserved MIP family structure. This
provoked an interest in us to find out to what extent MIP subfamilies and subgroups
would be conserved in a more primitive plant, that is a plant with a more ancient
shared ancestry with higher plants thus lacking many of the features of higher plants.
Therefore, in paper I we identify all MIPs in the moss Physcomitrella patens, a plant
that diverged from the lineage leading to higher plants around 450 million years ago.
We found that the P. patens MIP family shows an even greater variation with seven
subfamilies, including the four found in Arabidopsis. One of the three subfamilies
not found in Arabidopsis was previously identified and found to belong to the GLP
group of MIPs and were therefore named GLP like intrinsic protein (GIP) (Gustavs-
son et al., 2005). The remaining two subfamilies had not been reported previously.
One of them showed similarities with both the PIP and the TIP subfamilies and was
therefore named hybrid intrinsic protein (HIP) subfamily, whereas the other did not
show similarity to any known MIP family and to emphasize this lack of information
it was named the X intrinsic protein (XIP) subfamily. After extensive searches it ap-
pears to be that the GIP and HIP subfamilies are absent in higher plants, but mem-
bers of the XIP subfamily was found also in several higher plants (such as tomato,
tobacco and poplar) although the subfamily seems to have been lost in monocots and
some eudicots, such as Arabidopsis.
The present phylogeny of land plant MIPs is shown in Figure 9. Note that not only
the subfamilies, but even some of the subgroups are conserved in mosses. As can be
seen, the NIP subfamily is the most diverse of the three larger subfamilies (PIPs, TIPs
and NIPs) and with the more relaxed subgroup criteria, first used in maize and rice,
three subgroups of NIPs are formed (Chaumont et al., 2001, Sakurai et al., 2005).
The extraordinary homogeneity of the PIP1 and PIP2 subgroups also stand out. As
of yet it is not clear why these subgroups show this conservation, but one possibility
is that they are subject to some extra constraints, making the neutral evolutionary
space more confined than for other MIPs. One such constraint might be interactions
with other proteins, for example it could be an effect of PIP heterotetramerization.
However, this ability of PIP1s and PIP2s to form tetramers that include both iso-
forms is still debated and the physiological function of this is largely unknown, but
this will be discussed more in detail in the next subchapter.
27

28
0.2
0.2
PIP
MIPA
MIPB
MIPC
MIPD
MIPE
TIP
NIP
SIP
XIP
GIP
HIP
HIGHER PLANT
MOSS
ALGAE
Figure 9. The phylogeny of plant MIPs. A NJ tree constructed using the complete set of MIPs
from Arabidopsis, tomato, P. patens, and the nine algal species described in paper IV. Origin of
different MIPs are indicated by line type and subfamilies are indicated by colored background.
The algal specific subfamilies are indicated by a dotted outline of the colored area. The SIP/
MIPC part is shown at a different scale in order to fit in the same picture. This is indicated by
the dashed box and the smaller size of the scale bar.

4.2 Subfamilies
The original division of plant MIPs into subfamilies was a strict phylogenetic clas-
sification and as such only based on sequence similarity. However, as more and more
experimental data has become available it seems like this classification also largely re-
flect subcellular localization and substrate specificity, making the subfamily concept
much more useful. A lot of the research on plant MIPs have focused on the PIPs,
TIPs and NIPs and still very little is known about the SIP, GIP, XIP and HIP sub-
families. In the following paragraphs, I will try to give an overview of what is known
for each of these seven subfamilies.
4.2.1 PIPs
The plasma membrane intrinsic proteins (PIPs) were first identified as major constitu-
ents of plant cell plasma membrane, hence the subfamily name. As previously men-
tioned, PIPs form two highly conserved subgroups, PIP1s and PIP2s (see Figure 10)
and it seems that each subgroup has its own unique features. Ultrastructural studies
have shown that some PIP2s have a polar localization in root epidermal cells (Hachez
et al., 2006) whereas some PIP1s in mesophyll cells seem to be localized to struc-
tures called plasmalemmasomas, where the plasma membrane folds inwardly to create
a pouch-like structure (Robinson et al., 1996). PIP1s have also been localized to the
chloroplast inner membrane by immunogold localization (Uehlein et al., 2008) and
shown to have a high degree of retention in the ER (Zelazny et al., 2007).
Whereas PIP2s seem to be highly efficient water transporters, increasing the water
permeability 10-20 fold, PIP1s are al-
most impermeable to water (Moshelion
et al., 2002, Fetter et al., 2004). Carbon
dioxide have been suggested as an alter-
native substrate for both PIP1s (Ueh-
lein et al., 2003, Uehlein et al., 2008)
and PIP2s (Hanba et al., 2004), and al-
though there have been both transport
assay experiments, showing that PIPs do
transport CO2
, and experimentsshowing
that PIP1 overexpression in planta leads
to increased biomass (Sade et al., 2010,
Uehlein et al., 2003), the physiological
relevance for this transport is still debat-
ed. Finally, PIPs have also been suggest-
ed both to transport and to be regulated
by hydrogen peroxide (Dynowski et al.,
2008b, Boursiac et al., 2008a, Henzler et
29
PIP2
PIP1
ARABIDOPSIS
TOMATO
RICE
MOSS
(P. PATENS)
Figure 10. Subtree of the PIPs, showing the con-
servation of PIP1s and PIP2s. Line types repre-
sent species origin of PIPs.

al., 2004, Boursiac et al., 2008b). The function of hydrogen peroxide transport and
regulation of PIPs still have to be investigated more in detail, but could potentially
be very interesting as hydrogen peroxide is known to work as a signaling molecule in
for example pathogen response and thermo tolerance in plants.
That PIPs are gated, i.e. able to change between open and closed conformations,
was suggested early on but exactly how this was achieved remained unclear until the
structures of open and closed SoPIP2;1 was solved in 2006 (Tornroth-Horsefield
et al., 2006). From these structures it became clear that it is the D-loop that blocks
the pore in the closed state and that the stabilities of the two different states can be
affected by many different signals such as pH, Ca2+
levels and different kinds of ki-
nases. This is illustrated in Figure 11.
30
S274
S115
D28
E31
H193
R190
R118
D191
S36
V263
L264
P199
L200
A B
Figure 11. Picture showing the gating of SoPIP2;1. Picture A shows an overview of a tetramer
of closed SoPIP2;1s, while B shows an enlargement of the enclosed area of picture A. The D-
loop (obstructing the pore) is colored orange, the proposed position of Ca2+
is purple while the
N- and C-termini are colored yellow and red respectively. Gating can be affected by phospho-
rylation of S115 and S274 (colored cyan in picture B), pH sensing (through protonation of
H193) and via Ca2+
levels. All these will affect the intricate hydrogen bonding network shown
by dashed lines in B. In short the C termini and D-loop of two adjacent monomers interacts
through the sidechain (sc) of S274 to the backbone (bb) of P199 and L200 and through the sc
R190 and bb V263 and bb L264. The N termini and D loop (in the same monomer) interacts
though sc D191 to sc S36 and through the bb of R190 and bb D191 via three water molecules
to sc R118 and sc E31. Also in the same monomer, sc S115 interacts with sc E31 (Tornroth-
Horsefield et al., 2006, Khandelia et al., 2009).
On a slower timescale the permeability of membranes can be affected by changing
the localization of PIPs. It has been shown that there is a diacidic motif present in
some PIP2s that is important for trafficking (Zelazny et al., 2009) and that in some
PIPs one of the residues in the motif can be methylated, possibly altering the traffick-
ing (Santoni et al., 2006). Still, other PIPs lack the diacidic motif entirely but locates
properly to the PM anyway (Zelazny et al., 2009). It has also been shown that export

of AtPIP2;1 from the ER is depending on phosphorylation of the C-termini, and
that this phosphorylation is changed upon salt stress, resulting in AtPIP2;1 being
localized to internal structures (Prak et al., 2008). Changes in PIP mediated perme-
ation is also likely to occur via targeted degradation, as it has recently been shown
that AtPIP2;1 can be targeted by an E3 ubiquitin ligase, and subsequent degradation
by the 26S proteasome (Lee et al., 2009).
Interactions between PIP1s and PIP2s have been an intensive research area the last
couple of years. In 2004 it was shown that, whereas PIP1 expression hardly leads to
any increased water permeability, coexpression of PIP1s and PIP2s dramatically in-
crease water permeability, suggesting some form of interaction between the isoforms
(Fetter et al., 2004). Since then experiments using various PIP1/PIP2 isoforms have
given comparable results, but also shown that coexpression not only affect water
transport but also pH sensitivity and shutdown efficiency of the gating, as well as
CO2
transport (Alleva et al., 2010, Bellati et al., 2010, Otto et al., 2010). Although it
was unclear in the beginning how the interactions occurred, an increasing amount of
experimental data using techniques such as FRET, split-YFP, size exclusion chroma-
tography and gel-electrophoresis now strongly suggest that the isoforms can interact
in heterotetramers containing both PIP1 and PIP2 monomers (Otto et al., 2010,
Zelazny et al., 2007). The changes in permeabilities found can partially be explained
by the finding that coexpression leads to an improved PIP1 PM localization (as
PIP1s otherwise show a strong retention in the ER) (Zelazny et al., 2007) but it
could also be that physical interactions between the isoforms change the behavior
of the proteins, so that for example composition of the tetramer could influence
the transporting properties of the mono-
mers. One possibility is that monomers
in the tetramer can affect the gating of
each other, by stabilizing/destabilizing
the D-loop in the open form, something
that could potentially change gating
properties such as pH sensitivity for the
whole heterotetramer. To conclude, PIP
mediated permeation can be regulated
in response to many cues and on several
different levels, some of which are sum-
marized if Figure 12.
31
PHOSPH.
NUCLEUS
ER
GOLGI
PM
ENDOSOME
26S
pH
Ca2+
PHOSPH.
PHOSPH.
METHYL.
PIP1/2 INTERACT.
E3 UBIQ.
Figure 12. Overview of proposed PIP regulation
mechanisms discussed in the chapter.

4.2.2 TIPs
Just as with PIPs, TIPs (Tonoplast Intrinsic Proteins) were named after their main
localization which in this case is the vacuolar membrane, the tonoplast. This mem-
brane is extremely permeable to water, with osmotic water permeabilities up to 100
fold higher than those of plasma membrane (Maurel et al., 1997). TIPs are respon-
sible for the high permeability and they can make up as much as 40 percent of the
total tonoplast proteins (Higuchi et al., 1998).
Several different kinds of vacuoles exists in plants and in some early localization
studies it was found that two types of vacuoles, the protein storage vacuoles of seeds
and the large central vacuole of elongating cells, contained different subgroups of
TIPs (TIP3 and TIP1 respectively) (Johnson et al., 1989, Ludevid et al., 1992).
The concept of multiple vacuolar compartments, originating from different targeting
machineries, and residing in one cell has been a source for controversy for the last
15 years or so. In the late 1990s, Jauh and coworkers did some localization studies
using confocal microscopy and immunofluorescence labeling with antibodies raised
against the three different subgroups of TIPs known at the time (TIP1-3) to see if
TIPs could be used as markers for the different vacuolar types. They found mul-
tiple types of vacuoles that were differentially labeled and argued that plant cells had
the ability to generate at least three different types of vacuoles (Jauh et al., 1999).
However, the subject of co-existing, functionally diverse vacuoles have recently been
questioned since two independent localization studies, using immunogold and YFP-
fusion proteins respectively, both failed to discriminate between vacuoles as all TIP
isoforms labeled the same compartment (Hunter et al., 2007, Olbrich et al., 2007).
The general conclusion drawn from this is that the expression of TIP isoforms is
mostly controlled by developmental cues and that as a consequence of this, some
TIPs are associated with special types of vacuoles existing at a specific developmental
stage. The multivacuole hypothesis might still be valid for a number of cases, but not
universally (Frigerio et al., 2008). Just recently it was reported that some TIPs local-
ize to the plasma membrane, specifically in the seed maturation and the early seed
germination stages (Gattolin et al., 2010a). The physiological function of TIPs in the
PM is unclear, but they might be involved in imbibition (the rapid initial uptake of
water) of dry seeds.
A detailed expression mapping of all ArabidopsisTIPs expressed in root tissue showed
that TIPs were mainly localized in mature tissue as well as in the elongation zone,
but not in the root meristem (Gattolin et al., 2010b, Gattolin et al., 2009). The two
exceptions to this was AtTIP1;2 and AtTIP2;1 which were present in the root caps
and in lateral root primordia respectively.
32

Apart from transporting water, TIPs have also been shown to be permeable to glyc-
erol, urea, ammonia and hydrogen peroxide (reviewed in Wudick et al., 2009). The
physiological relevance for non-water transport in TIPs remains unclear. A previ-
ously reported cell-death phenotype of an Arabidopsis TIP1;1 RNAi line (Ma et al.,
2004) has been reevaluated as due to an off-target silencing effect, since two stud-
ies on knock-out and knock-down AtTIP1;1 plants failed to reproduce any severe
phenotype (Schussler et al., 2008, Beebo et al., 2009). However, both AtTIP1;1 and
AtTIP1;2 have been found to be permeable to hydrogen peroxide (Bienert et al.,
2007) and double knock-outs of these TIPs showed minor signs of increased oxida-
tive stress under high light conditions, indicating that they might have roles during
plant stress (Schussler et al., 2008). AtTIP1;1 knockout plants were also recently
shown to have a reduced root growth when grown in glycerol containing media
(Beebo et al., 2009) which is puzzling since AtTIP1;1 was found not to transport
glycerol in oocyte experiments (Maurel et al., 1993).
Even though lack of TIP isoforms seems to have only minor effects on plants, over-
expression of TIPs have been shown to result in increased cell surface and volume
(Reisen et al., 2003), as well as increased cell division and cell elongation rates (Oku-
bo-Kurihara et al., 2009). Several studies have shown that overexpression results in
plants with increased growth of both vegetative and reproductive tissues (Lin et al.,
2007, Peng et al., 2007, Sade et al., 2009). Peng and coworkers also showed that
transgenic Arabidopsis plants, overexpressing a ginseng TIP, had an increased toler-
ance to drought and salt stress. During salt stress, the shoots of the overexpressing
plants actually accumulated more sodium than the controls, maybe indicating an
increased possibility of intracellular storage due to larger vacuoles and cells (Peng et
al., 2007). An increased tolerance to drought and salt stress was also seen in the study
by Sade and coworkers, who could not only see an increased biomass, but also an
increased yield, of tomato overexpressing SlTIP2;2. They also found that transpira-
tion rates of the TIP-overexpressing tomato plants was increased and that it did not
decrease as much during stress as in the wild type, suggesting that maybe TIP over-
expression could be a key to convert isohydric plants (drought avoidant, with a tight
stomata control) to anisohydric plants (drought tolerant, with less strict stomata
control) (Sade et al., 2009). Recently, TIP overexpression was also shown to result in
an increased resistance to boron, with overexpressors showing accelerated develop-
ment as well as increased silique production compared to wild type under high boron
stress (Pang et al., 2010). Increased boron content of overexpressors indicates that it
might be the same mechanism as is responsible for the increased sodium tolerance.
33

4.2.3 NIPs
The first plant MIP discovered was NOD26 or Nodulin-26, named after the tissue in
which it was discovered, root nodules of
soybean. As it later on became apparent
that there were many plant MIPs related
to NOD26 although not localized to nod-
ules, these MIPs were named NOD-26 like
intrinsic proteins (NIPs) (Heymann and
Engel, 1999). NIPs have mainly been local-
ized to the plasma membrane (Choi and
Roberts, 2007, Ma et al., 2006, Takano et
al., 2006, Tanaka et al., 2008, Mitani et al.,
2009), but one study has also reported ER
localization (Mizutani et al., 2006).
As seen in Figure 13, NIPs form a very di-
verse subfamily of MIPs, so diverse in fact
that the more stringent subgrouping criteria
used for all Arabidopsis MIPs is only useful
when comparing this subfamily in closely
related species. Therefore when comparing
plantspeciesthelessrigoroussubgroupcrite-
ria used in the description of maize and rice
NIPs is preferred, resulting in NIPs belong-
ing to three major subgroups (NIP1-3). NIP1s show moderate water permeability, but are
also able to transport for example formamide and glycerol, whereas NIP3s showed no water
transport but were still permeable to formamide and glycerol as well as larger uncharged
solutes such as urea (Wallace and Roberts, 2005, Wallace et al., 2002). The third major
subgroup, NIP2, is missing in Arabidopsis and was first recognized as a subgroup when the
MIPs in rice and maize were phylogenetically characterized. In 2006 a NIP2 was identified
as being responsible for low silicon sensitivity of a rice mutant isolated in a genetic screen
(Ma et al., 2006). Not only did this identify a possible physiological role for NIP2s, but it
also established metalloids as a whole new group of substrates for MIP transport in general,
and specifically NIP2 transport. The identified NIP2 (OsNIP2;1 also called Lsi1) is local-
ized in the endo- and exodermal cells, at the location of the suberized casparian strips, in the
rice root. Interestingly the subcellular localization is polar, with OsNIP2;1 only expressed in
the PM facing away from the root center, and it turns out it works in series with an active
transporter (Lsi2) localized in the PM facing the other side, this is illustrated in Figure 14.
A similar arrangement is found in the xylem parenchyma cells of the shoots, where
OsNIP2;2 is located on the side facing the vessel (Yamaji et al., 2008). NIP2s in both barley
and maize has been found to have a similar location pattern and probably share the same
34
NIP3
NIP2
NIP5
NIP1ARABIDOPSIS
TOMATO
RICE
MOSS
(P. PATENS)
Figure 13. Subtree of the NIPs showing the three
major groups of higher plants and the NIP5
group of P. patens. Line styles indicate species of
origin for MIP sequences.

function in planta (Mitani et al., 2009, Chiba et al., 2009). NIP2 overexpressing rice have
been shown to have increased silicon content and increased yields, but unfortunately also
higher arsenite levels, probably resulting from arsenite and silicon having very similar struc-
tures and using the same transport routes in the plant (Ma et al., 2008). Arsenite transport
has also been shown to involve both a NIP1
and a NIP3 in Arabidopsis (Kamiya et al.,
2009, Isayenkov and Maathuis, 2008), in-
dicating that this substrate is not restricted
to the NIP2 subgroup. Two NIP3s in Ara-
bidopsis (AtNIP5;1 and AtNIP6;1) has
been identified to take part in transport of
another metalloid, boron, possibly in a way
similar to that of the silicon transporting
NIP2s in rice (Takano et al., 2006, Tanaka
et al., 2008). AtNIP2;1, a NIP1, was shown
to be upregulated under water logging and
to be permeable to lactic acid, leading to the
suggestion that it might have a role during
lactic acid fermentation in oxygen deprived
roots (Choi and Roberts, 2007).
4.2.4 SIPs
The small and basic intrinsic proteins, SIPs, were originally discovered when the
Arabidopsis genome was screened for MIPs and named after their characteristics as
being smaller than PIPs and NIPs and more basic than TIPs (Johanson et al., 2001,
Johanson and Gustavsson, 2002). Still very little is known about these MIPs, they
are very divergent with unusual NPA boxes and ar/R filters, form two subgroups in
higher plants and tend to cluster together with the mammalian AQP11 and AQP12
in phylogenetic analyses (as seen in Figure 3). They seem to be localized to the rough
ER and some are permeable to water, but so far no phenotype for knock-out plants
have been described (Ishikawa et al., 2005, Maeshima and Ishikawa, 2007, Katsu-
hara et al., 2008).
35
[Si]
[Si]
[Si]
LSI1/OSNIP2;1
LSI2
Figure 14. A schematic picture showing how the
passive transport through OsNIP2;1/Lsi1 works
in series with the active transporter Lsi2 in root
uptake of silicon (red/black dotted line). Both
OsNIP2;1 and Lsi2 are expressed in root cells
localized to the casparian strips but show oppo-
site polar localization. This arrangement makes it
possible for the plant to transport Si against the
concentration gradient (as indicated by the sizes
of [Si] in the picture).

4.2.5 GIPs
In 2005 a plant MIP was identified in mosses and turned out to be different from
all other known plant MIPs in that it belongs to the GLP cluster of MIPs (see Figure
15 and Chapter 3.1). The subfamily was
named after this grouping to the GlpF-
like intrinsic protein (GIP) subfamily.
GIPs have so far only been identified
in two species of moss, are predicted to
be localized to the plasma membrane
and seem to be transporting glycerol
(Gustavsson et al., 2005)
4.2.6 HIPs
Another group of plant MIPs that was found upon identifying all MIPs in the moss P. pat-
ens (paper I) are the HIPs. These MIPs had characteristics of both PIPs andTIPs and were
therefore named hybrid intrinsic proteins (HIPs). HIPs are not restricted to mosses, but
seem to be limited to non-seed plants as they were also found in the spikemoss Selaginella
moellendorffii but not in any higher plants.To date no HIPs have been experimentally char-
acterized, but they have been predicted to be localized in the tonoplast (paper I).
4.2.7 XIPs
The X intrinsic proteins (XIPs) were also identified when characterizing the MIP family
of P. patens, but contrary to the HIPs, XIPs were also found in several higher plants (pa-
per I). An attempt to determine the expression pattern of XIPs by examining the tissues
of origin of the cDNA libraries where the different XIP ESTs was made, but no conclu-
sions could be drawn since XIP transcripts were isolated from too many different sources
(paper I). Analysis of microarray data for poplar also suggests that XIPs are abundantly
expressed but do not show any tissue-specific expression (Gupta and Sankararamakrish-
nan, 2009). Very recently a more detailed study of XIP expression in cotton revealed that
the XIP isoform studied seem to be highly expressed in mature leafs and to some extent
also in other parts of the shoot but not at any detectable level in the roots (Park et al.,
2010). So far no experimental data of XIP permeability is available, but an evaluation of
the ar/R filter shows a hydrophobic filter suggesting that XIPs are not mainly water chan-
nels but rather transporting other substrates (paper I).
36
BACTERIA
ANIMAL
PROTIST
FUNGI
MOSS
GLP I
GLP II
Figure 15. Phylogeny of the GLP subfamily.
Line styles show species of MIP sequences
and brackets indicate the GLPI and GLPII
groups (with GLPII containing the P. patens
GIP).

4.3 Evolution
The MIP family is an ancient gene family with its origin believed to date back some
2.5 to 3 billion years (Pao et al., 1991, Park and Saier, 1996). The ancient origin
is reflected by the fact that MIPs are present in all three domains of life, Bacteria,
Archaea and Eukarya. Even though MIPs are present in all kinds of organisms, they
are particularly abundant in plants. The abundancy likely reflects that MIPs have
important functions in plants and results in that many MIPs probably have overlap-
ping functions. This might explain why traditional loss of function studies, using
knock-out/knock-down techniques, largely have been unsuccessful in identifying
physiological roles for MIPs. Another approach is to try and deduce the function of
proteins by comparing these in different species, in order to see if the evolution of
the protein isoforms and families can be linked to the emergence or disappearance
of some trait. This can be done either by trying to identify the same specific gene in
some evolutionary related species of interest, or in a more “comparative genomics”
kind of approach, where whole gene families are identified in genomes of sequenced
species and subsequently compared. Using the first approach one is free to chose
whatever species one want, but is limited in that conclusions only can be drawn from
positive results, since a negative result does not necessarily indicate that the species
is missing the gene, only that it cannot be detected. In a genome-sequenced species
on the other hand, it is possible to draw that kind of conclusions, making the “com-
parative genomics” approach more powerful. However, the number of species with a
sequenced genome, although rapidly increasing, is still very small.
In order to understand the evolution of the plant MIPs it is necessary to relate it to
the evolution of plants. Therefore follows a brief review of plant evolution, which
together with Figure 16 hopefully will be useful for understanding the rest of this
subchapter. Plants (archaeplastida or “plantae sensu lato”) are descendants from a
unicellular eukaryote that acquired a photosynthetic cyanobacterium as an endo-
symbiont (the origin of the chloroplast) and they are one of the five or so kingdoms
of eukaryotes. Plants contain three groups, a small group of microscopic freshwater
algae called glaucophytes, rhodophytes (red algae) that are primarily marine algae
including for example the reef-building coralline algae, and viridiplantae (green
plants) comprising the chlorophytes and streptophytes. The chlorophyte group of
green plants contains the vast majority of all freshwater and marine algae, and con-
tains both uni- and multicellular species. The streptophytes contain the uni- and
multicellular charophycean freshwater algae (charophytes), and the multicellular
land plants (embryophytes). Land plants are further divided into non-vascular plants
(bryophytes) and vascular plants (tracheophytes), and vascular plants into non-seed
plants (e.g. ferns and lycophytes) and seed plants. Seed plants are divided into non-
flowering (gymnosperms) and flowering (angiosperms), where the flowering plants
contain the two groups of monocots and eudicots (having one and two cotyledons
37

respectively). Many of the main events in plant evolution are reflected in the naming
of the groups, but some others are indicated in Figure 16 and include for example
multicellularity and apical growth (occurred already in the charophyte-“land plant”
ancestor), alternation of generations involving a diploid phase (first land plants) and
three dimensional patterning (Bowman et al., 2007).
38
EUDICOTS (1)
MONOCOTS (2)
BASAL ANGIOSPERMS (3)
GYMNOSPERMS (4)
FERNS (5)
LYCOPHYTES (6)
HORNWORTS
MOSSES (7)
LIVERWORTS (8)
CHAROPHYCEAE (9)
COLEOCHAETOPHYCEAE (10)
ZYGNEMOPHYCEAE
KLEBSORMIDIOPHYCEAE
CHLOROKYBOPHYCEAE
TREBOUXIOPHYCEAE
CHLOROPHYCEAE (11)
ULVOPHYCEAE (12)
“PRASINOPHYTE GRADE”
RHODOPHYTES (13)
GLAUCOPHYTES (14)
1 TOMATO
3 MAGNOLIA
5 FERN
7 PHYSCOMITRELLA
9 NITELLA
11 VOLVOX
13 IRISH MOSS
2 WHEAT
4 SPRUCE
6 SELAGINELLA
8 MARCHANTIA
10 COLEOCHAETE
12 SEA LETTUCE
14 GLAUCOCYSTIS
FLOWERS
SEEDS
VASCULATURE
SPOROPHYTE DOMINANT
EMBRYO
3D PATTERNING
GAMETOPHYTE DOMINANT
APICAL GROWTH
1500
1400
1200
420-
480
360
300
VASCULARPLANTSBRYOPHYTESCHAROPHYTESCHLOROPHYTES
Figure 16. An overview of the evolution of the plant kingdom, adapted from Bowman et al.,
2007. The numbers in the nodes of the tree are rough estimates of divergence dates in million
years ago. Four different subsets of green plants are indicated by different shades of green and
pictures show some examples from different groups. Arrows indicate evolutionary events of
special interest. All photos are part of the public domain.
The first two plant genomes that were sequenced were those of Arabidopsis (an eu-
dicot) and of rice (a monocot). In both these genomes, the whole MIP superfamily
was identified (Johanson et al., 2001, Sakurai et al., 2005), and when comparing
them they seem very similar. But one thing that stands out is that rice seems to have
a subgroup of NIPs absent in Arabidopsis (NIP2s). It turns out that this subgroup
is also present in other monocots, and later it was identified as being involved in
silicon transport in rice (Ma et al., 2006). Silicon is generally not believed to have

any important function in plant cells, and therefore to be non-essential for plants.
However, silicon is beneficial for some plants and for example in monocots silicon
is deposited extracellularly to give an increased rigidity to the plant and prevent
lodging. Most eudicots on the other hand, do not use silicon in this way, which is in
agreement with them not having a NIP2 subgroup. Interestingly some eudicots in
the cucumber/squash family, for which it is well known that silicon is beneficial, turn
out to also have the NIP2 subgroup, even though they are eudicots. This is indicat-
ing that the NIP2 subgroup is likely to have a more ancient origin than monocots/
dicots and possibly that silicon played an important role in rigidity of early vascular
land plants.
In 2007 the genome of the moss Physcomitrella patens was released, and upon iden-
tification and comparison of the moss MIP superfamily to those of higher plants,
they turned out to be surprisingly similar (paper I). All the subfamilies and even
some of the subgroups found in higher plants were present already in moss, and
although the total number of isoforms was lower (23 compared to 35 for P. patens
and A. thaliana respectively) the diversity was greater, with seven subfamilies present
in the moss. This was unexpected, as mosses are relatively “simple” plants, lacking
many of the features believed to make transport in higher plants more complex, such
as root system, vascular tissues, leaf-stem morphology, lignified cell walls and so on.
Still it was clear that some features were much more conserved than others, such as
the extremely conserved PIP subfamily, in which both the PIP1 and PIP2 subgroups
were present in P. patens. In the TIP subfamily on the other hand, no subgroup was
conserved in mosses and higher plants, even though all subgroups were conserved
between monocots and eudicots. This suggests that most of the TIP subgroups of
higher plants have higher-plant-specific functions, which is compatible with results
from some reports, saying that TIPs are for example involved in seed specific pro-
cesses (Gattolin et al., 2010a). Likewise, some NIP subgroups seem to be specific for
higher plants (NIP2s), while some are found in mosses and higher plants (NIP3s),
and yet others are moss specific. While some of the expansion/diversification seen
in the subfamilies of higher plants is probably reflecting higher-plant-specific func-
tions, it could also be that some of it is the result of a compensatory effect where
subgroups in MIP subfamilies of higher plants have replaced the functions of moss
specific subfamilies.
The complete MIP family has been identified in nine algal species, all belonging to
the chlorophyte group (paper IV). Even though each species only contained a lim-
ited number of MIP isoforms (between one and five, on average 2.4) the algal MIPs
were very diverse, with a total of seven different subfamilies (see Figure 17). Of these,
only two (PIPs and GIPs) were recognized in land plants and one subfamily (MIPC)
shows similarities to the “superaquaporin” group consisting of AQP11, AQP12 and
SIPs (Morishita et al., 2004). Summarized, the presence of several different subfamilies
39

might not be very surprising, as several of these algae are very distantly related and
the subfamilies might therefore just reflect the phylogeny of the species. But the
presence of MIP subfamilies shared with land plants is surprising. Especially since
these subfamilies only existed in one to two species of algae and none of these species
contained all of these subfamilies (PIP, GIP and MIPC). What makes the results dif-
ficult to interpret is that none of the algal species is more closely related to land plants
than any other (since they are all belonging to the same sister group of land plants).
To explain the shared subfamilies by a shared evolutionary origin would therefore re-
quire a complex pattern of gene loss in the algae, and a more likely explanation might
therefore be horizontal gene transfer (HGT). A possibility is that this HGT was virus
mediated, since an algal virus has been found to contain a GIP type of MIP, closely
related to the GIPs found in both algae and P. patens (Gazzarrini et al., 2006).
40
CHLOROPHYTES
BRYOPHYTES
ANGIOSPERMS
C. RHEINHARDTII
V. CARTERI
C. C-169
C. NC64A
O. RCC809
O. LUCIMARINUS
O. TAURI
M. PUSILLA
M. RCC299
P. PATENS
A. THALIANA
O. SATIVA
MIPA
MIPB
MIPC
MIPD
MIPE
GIP
PIP
TIP
NIP
SIP
XIP
HIP
-
-
1
-
-
-
-
-
-
-
-
-
-
-
-
-
1
1
-
-
-
-
-
-
-
-
-
-
-
-
1
1
1
-
-
-
2
3
2
1
-
-
-
-
-
-
-
-
-
-
-
3
1
-
-
-
-
-
-
-
-
-
1
1
-
-
-
-
-
1
-
-
-
-
2
-
-
-
-
-
-
8
13
11
-
-
-
-
-
-
-
-
-
4
10
10
-
-
-
-
-
-
-
-
-
5
9
10
-
-
-
-
-
-
-
-
-
2
3
2
-
-
-
-
-
-
-
-
-
2
-
-
-
-
-
-
-
-
-
-
-
1
-
-
TOT.MIPS
2
3
6
4
2
1
1
1
1
23
35
33
TOT.SUBFAM.
1
1
4
3
2
1
1
1
1
7
4
4
Figure 17. A table showing the size and diversity of the identified MIP families in nine chlo-
rophyte algae, the moss P. patens and the higher plants Arabidopsis and Rice. The schematic
tree shows how the species are related to each other (but not in scale) and numbers in table are
number of isoforms except for last column which shows number of subfamilies.
Horizontal gene transfer might be an explanation for the diversity of plant MIPs and
it was suggested already in 2002 that the NIP subfamily originated from a HGT of
bacterial AqpZ (Zardoya et al., 2002). However, a later study found no support for
this (Zardoya, 2005), but a HGT origin of NIPs cannot be rejected. In paper III
we looked closer at the phylogeny of NIPs, and successfully identified several bacte-
rial NIP homologs. These bacteria do not correspond to any taxonomic subgroup

and the sequences do not appear to form a stable monophyletic group. Still there is
a relatively high support for this, among bacteria widely distributed, MIP subfam-
ily sharing an ancestry with the NIP family (see Figure 18 and paper III). Another
study has found XIP homologs in fungi (Gupta and Sankararamakrishnan, 2009),
possibly indicating an origin involving HGT also for this plant MIP subfamily (see
Figure 18).
41
DICOTS
SM
PP
PLANT XIPS
FUNGI XIPS
MIPD
AQPM
100
95
100 50
55
100
0.1
A
PLANT
NIPS
BACTERIAL
NIPS
1
2
5
3
AQPZ
AQPM
100
100
72
72
100
96
36
64
91
64
91
93
0.1
B
Figure 18. Trees showing possible shared ancestry of plant and non-plant MIPs. Tree A is a NJ
tree of plant XIPs and fungi XIPs. MIPD was found to be the subfamily closest to plant XIPs
and is included as an outgroup. Sm and Pp is indicating the group of XIPs from Selaginella
moellendorffii and Physcomitrella patens. The tree is rooted with the Achaean AQPM. Tree B
is a NJ tree of plant NIPs and bacterial NIPs. AqpZ was found to be the subfamily closest to
NIPs in previous studies and was therefore included as an outgroup. The tree was rooted with
AQPM. Bootstrap support from 1000 replicates are showed at the nodes of major groups.

4.4 Physiological Role
Despite a relatively intensive research on plant MIPs in the recent past, the physi-
ological roles of the different plant MIP subfamilies and isoforms are to a large extent
still unknown. The physiological functions that have been ascribed some plant MIPs
are largely associated with non-water specific permeabilities of these MIP isoforms,
such as the involvement of different NIPs in transport of the metalloids silicon and
arsenite. However recent research have also identified other potential roles for MIPs.
For example Secchi and Zwieniecki identified PIP1s as being important for xylem
embolism repair in poplar (Secchi and Zwieniecki, 2010).
There have also been many reports on the general involvement of PIPs in plant wa-
ter relations, but unfortunately these tend to be unspecific, looking at for instance
root hydraulic conductivity or transpiration and to what degree it is blockable by
mercury (which blocks some, but not all, MIPs, and also is likely to affect many
other proteins). Recently a study was presented that showed a reduction in hydraulic
conductivity in AtPIP1;2 knock-out plants, showing that this specific isoform most
likely is a key component in whole plant hydraulics (Postaire et al., 2010). However,
the success of using knock-out plants in this research might be specifically linked
to AtPIP1;2 since it is one of the most highly expressed MIPs in Arabidopsis (Alex-
andersson et al., 2005). For other lowly expressed MIPs knock-out or knock-down
experiments have been unable to reveal any phenotypes suggesting a physiological
role.
Another alternative is to use naturally occurring genetic variants of a species (acces-
sions) to see if differences in traits can be linked to MIP family differences. In paper
II we compared the expressional regulation of MIPs during drought in five different
accessions with known differences in water use efficiency but could only identify
some minor differences. Very recently a study was published that, using a similar ap-
proach, looked at MIP responses to drought stress in two poplar accessions known
to be drought avoidant (isohydric) and drought tolerant (anisohydric) (Almeida-
Rodriguez et al., 2010), identifying some differences possibly linked to hydraulic
and stomatal responses.
42

5 Conclusions and
Future Perspectives
Around the turn of the millennium, life science experienced something close to
a paradigm shift, as techniques producing large amounts of data (e.g. microarray
based RNA detection, automated capillary electrophoresis based DNA sequencing,
mass spectrometry based protein detection) made it possible to conduct more large
scale high-throughput studies (often referred to as “omics” based, e.g. transcriptom-
ics, proteomics, metabolomics and so on). Although people in general were thrilled
about the promising new technologies there were also concerns raised over that they
would result in for example a more descriptive kind of research, perhaps failing in
asking the relevant biological questions. Now, a decade or so later, well on our way in
the “omics”-era the use of these buzzwords, as well as the concerns raised over these
techniques, have steadily declined. The techniques have partially shifted the focus of
biological research from acquisition to analysis of data and from the reductionism of
the traditional hypothesis-driven research to the holism of model building systems
biology. Overall the fears have turned out to be unjustified and the massive amounts
of publicly available data has turned out to be a valuable resource for the research
community at large. In fact, this thesis would never have been possible without the
“omics” revolution, as the research in this thesis to a large part is based on analysis of
such freely available data.
The numerousness and diversity of the plant MIP family has been known for quite
some time and is likely reflecting important roles in several physiological functions
in the plant. However, the exact functions of different subfamilies and specific iso-
forms of MIPs have been difficult to assess and are therefore to a large extent still
unknown. We looked at gene expression of water transporting PIPs during drought
stress in five different genetic isolates (accessions) of A. thaliana. Even though the ac-
cessions differed in water use efficiency (WUE, biomass production per water used)
we found the overall pattern of PIP expression to be similar in all of them, indicating
that differences in PIP expression is not likely to be the cause for differences in WUE.
By analyzing publicly available microarray data we found out that the PIPs most
down-regulated during drought tend to covary also in other conditions, indicating
that they might be part of the same transcriptional network.
By identifying the MIP family in the moss P. patens as well as in nine species of chlo-
rophyte algae we have formed a framework for looking at the evolution of this family
43

in the green branch of life. With the data available we speculate that MIP subfamilies
were numerous already in the first land plants and that some of these originate from
horizontal gene transfers from other species, possibly chlorophyte algae, fungi and
prokaryotes. In prokaryotes, around 80% of all genes show signs of such horizontal
gene transfer indicating that this has been extremely common (Dagan et al., 2008).
As a result of this it has been suggested that the common use of a tree to illustrate
evolution might be better replaced by a web, the web of life. The involvement of
horizontal gene transfers in the evolution of the plant MIP family is still only specu-
lations, but it will be interesting to see how the MIP families of other plant species
(especially charales) will fit into this as their sequences will become available.
And yes, the phrase in the last paragraph should be “as they become available” and
not “if …” because as of today it is a question of “when” and not “if” the genome
of a species will be sequenced. The number of nucleotide sequences deposited in
GenBank has grown exponentially (see Figure 19) as the costs for sequencing a hu-
man genome has plummeted over the last couple of years, from about $10 million
in 2007, $1million in 2008, $48 000 in 2009 to only $4 400 in 2010 (Drmanac et
al., 2010). The price drop is due to introductions of new technologies for massive
parallel sequencing (mostly based on pyrosequencing techniques and referred to as
second-generation sequencing). Improvements in these techniques together with re-
cently introduced third generation sequencing techniques (based on single-molecule
detection, enabling faster and longer reads) make it likely that the “one thousand
dollar genome” goal will not only be reached but surpassed in the near future (Mun-
roe and Harris, 2010). Even though these techniques might not rock the life science
community as much as the introduction
of the “omics” techniques once did, they
are likely to cause a lot of changes to the
way life science research is done. One
way would be that people are no lon-
ger restricted to the genomic sequences
available from sequencing consortia,
but can simply sequence their organ-
ism of interest. Another would be that
researchers will no longer be restricted to
working with model-species when work-
ing with gene expression, but rather can
focus directly on the most biologically
interesting species.
44
Figure 19. Graph showing the number of nu-
cleotides deposited in GenBank over time. At
the moment the database contains more than
1.17×1011
bases in sequences belonging to more
than 250 000 organisms, with the amount of se-
quences doubling about every 18months.

Populärvetenskaplig sammanfattning
Livets minsta beståndsdel, cellen, omges av ett membran som avgränsar den från
omvärlden. För att hålla den inre miljön kontrollerad måste cellen kunna styra trans-
porten av olika ämnen över membranet, vilket görs med hjälp av olika proteiner.
Dessa proteiner kan antingen vara energikrävande pumpar och transportörer som
aktivt förflyttar substanser över membranet, eller de kan vara passiva transportörer
och kanaler som påverkar diffusionen av ämnen över membranet.
Ett exempel på det senare är de kanalbildande proteinerna MIPar (Major Intrin-
sic Proteins). Dessa proteiner identifierades tidigt på grund av att de var så rikligt
förekommande i vissa membran (därav namnet) men först senare lyckades man visa
deras funktion. En av de första MIParna vars funktion påvisades var en vattentrans-
porterande kanal, varför detta protein och ibland hela gruppen av proteiner även
kallas akvaporiner (AQPs). Detta var en mycket stor upptäckt, då det länge varit
omdiskuterat huruvida vatten bara var beroende av passiv diffusion genom cellmem-
branet, eller om det behövdes proteiner för transporten. Upptäckten skulle leda till
ett ökat intresse kring MIPar och resulterade även i att Peter Agre, som ledde den
grupp av forskare som gjorde de första upptäckterna kring vattentransport i MIPar,
tilldelades nobelpriset i kemi 2003.
Sedan proteinfamiljen upptäcktes i slutet av 1980-talet har intensiva studier bedrivits
i syfte att kartlägga och karaktärisera olika varianter (isoformer) av MIPar och det
har visat sig att MIPar finns i de allra flesta levande organismer och att de inte enbart
transporterar vatten, utan även en stor mängd andra ämnen, såsom ammoniak, glyc-
erol, arsenik och urea. Dock kvarstår många frågetecken kring de olika isoformernas
lokalisering, specifika funktion, struktur och reglering.
I växter bildar MIPar en stor och omfångsrik familj, med nästan tre gånger så många iso-
former som det finns i däggdjur (35 respektive 13). Anledningen till denna omfångsrike-
dom är i mångt och mycket fortfarande oklar, men det har spekulerats i att den har en
grund i växters speciella behov av reglering av transport i allmänhet och vattentransport i
synnerhet. Detta behov har sin grund i att växter, till skillnad från djur, inte kan förflytta
sig utan måste anpassa sig till förändringar i den miljö de lever i. Det ligger därför nära till
hands att tänka sig att en ökad möjlighet att reglera t.ex. vattenflöden under perioder av
torka, skulle vara viktigt för växter.Tyvärr har traditionella metoder där man slår ut enskilda
MIP-gener (och därmed proteinerna) till stor del misslyckats med att identifiera funktioner
hos enskilda isoformer, troligtvis på grund av överlappande funktion hos isoformerna.
45

Ett annat sätt att få ökad förståelse och kunskap kring MIPar i växter är genom att
studera MIP-familjen i olika växter och jämföra dessa. Till exempel har vi i Paper II
använt oss av den biologiska variationen inom samma art (s.k. ekotypvariation) för
att studera huruvida variationen i effektiviteten av vattenanvändande kan förklaras
av skillnader i reglering av MIP-gener. I Paper I och IV har vi istället valt att titta på
MIP-familjen i mossa respektive alger för att se hur familjen har utvecklats i växter.
Vi har visat att redan mossa, som är en av de mest primitiva landväxterna, har en
MIP-familj som är mycket varierad och relativt stor (om än något mindre än den i
högre växter). Vidare har de nio arter av alger vi undersökt förvisso endast ett fåtal
MIPar per art, men variationen är även här oväntat stor. Intressant nog hittade vi
även grupper av MIPar i alger som också återfinns i mossa. Dessa återfanns endast
i ett fåtal algarter och detta, tillsammans med det faktum att alla de undersökta al-
gerna är lika avlägset släkt med mossa, tyder på att dessa ”delade” MIP-varianter kan
ha överförts med så kallad horisontell genöverföring och alltså inte är nedärvda från
en gemensam ursprunglig organism. Att även andra grupper av växt MIPar skulle ha
ett ursprung från horisontell genöverföring stöds även av att MIP-gener som visar
släktskap med vissa växt-MIPar har identifierats från såväl bakterier (Paper III) och
svampar.
Det finns redan nu resultat som tyder på att ett modifierat uttryck av MIPar kan
påverka såväl växters tålighet och produktion under torkstress, som växters resistens
mot arsenik och bor. En ökad kunskap om växt-MIPar kommer på sikt att leda till
förståelse kring hur dessa proteiner påverkar transportprocesser i växter. Kunskap
som till exempel kan vara avgörande för att på ett effektivt och hållbart sätt kunna
möta det behov av en ökad matproduktion som världen ställs inför till följd av en
stigande befolkningsmängd.
46

Acknowledgements
I would like to express my sincere gratitude to all the people who in various ways
contributed to the making of this work. Especially I would like to thank:
My supervisor Urban Johanson for the support and advice provided throughout my
PhD. Your genuine interest in and enthusiasm for science and plant biology have
been inspiring and I am sure I will miss those talks we used to have.
My co-supervisor Per Kjellbom for sharing his great knowledge on Aquaporins and
biochemistry in general and for making us feel like a group.
The rest of the Aquaporin group: Kristina Nordén, Maria Fellert, Andreas Kirscht,
Sabeen Survery, Hanna Anderberg, Anna Assarsson, Anders Barfod, Lavanya Mo-
parthi, Michael Rützler and Jan Kvassman.
Erik Alexandersson, former group member who helped me during my first strug-
gling time.
Folke Tjerneld and Lisa Thelin for all those music-nerd-talks, sharing records and
giving concert tips.
Sinan for being my friend, not only at work.
The innebandy and badminton crew for all the fun.
All the other people at CMPS, making it a nice place to work.
The old friends from Göteborg. I miss you all, and the city, but not the weather.
My brother and sister. I know I can always count on you even though I guess you will
always consider me your kid brother.
Mom, for always being supportive and believing in me.
Dad, how I wish you were here.
Finally I would like to thank Helena, for everything.
47

References
ALEXANDERSSON, E., FRAYSSE, L., SJOVALL-LARSEN, S., GUSTAVSSON, S., FELLERT, M.,
KARLSSON, M., JOHANSON, U. & KJELLBOM, P. (2005) Whole gene family expres-
sion and drought stress regulation of aquaporins. Plant Mol Biol, 59, 469-84.
ALLEVA, K., MARQUEZ, M., VILLARREAL, N., MUT, P., BUSTAMANTE, C., BELLATI, J.,
MARTINEZ, G., CIVELLO, M. & AMODEO, G. (2010) Cloning, functional character-
ization, and co-expression studies of a novel aquaporin (FaPIP2;1) of strawberry fruit. J Exp
Bot. 61, 3935-45.
ALMEIDA-RODRIGUEZ, A. M., COOKE, J. E., YEH, F. & ZWIAZEK, J. J. (2010) Functional
characterization of drought-responsive aquaporins in Populus balsamifera and Populus si-
moniixbalsamifera clones with different drought resistance strategies. Physiol Plant. no.
doi: 10.1111/j.1399-3054.2010.01405.x
The Arabidopsis Genome Initiative (2000) Analysis of the genome se-
quence of the flowering plant Arabidopsis thaliana. Nature, 408, 796-815.
BAKER, M. E. & SAIER, M. H., JR. (1990) A common ancestor for bovine lens fiber major intrinsic
protein, soybean nodulin-26 protein, and E. coli glycerol facilitator. Cell, 60, 185-6.
BEEBO, A., THOMAS, D., DER, C., SANCHEZ, L., LEBORGNE-CASTEL, N., MARTY, F.,
SCHOEFS, B. & BOUHIDEL, K. (2009) Life with and without AtTIP1;1, an Arabidopsis
aquaporin preferentially localized in the apposing tonoplasts of adjacent vacuoles. Plant Mol
Biol, 70, 193-209.
BEITZ, E., WU, B., HOLM, L. M., SCHULTZ, J. E. & ZEUTHEN, T. (2006) Point mutations in
the aromatic/arginine region in aquaporin 1 allow passage of urea, glycerol, ammonia, and
protons. Proc Natl Acad Sci U S A, 103, 269-74.
BELLATI, J., ALLEVA, K., SOTO, G., VITALI, V., JOZEFKOWICZ, C. & AMODEO, G. (2010)
Intracellular pH sensing is altered by plasma membrane PIP aquaporin co-expression. Plant
Mol Biol, 74, 105-18.
BIENERT, G. P., MOLLER, A. L., KRISTIANSEN, K. A., SCHULZ, A., MOLLER, I. M., SCHJO-
ERRING, J. K. & JAHN, T. P. (2007) Specific aquaporins facilitate the diffusion of hydro-
gen peroxide across membranes. J Biol Chem, 282, 1183-92.
48

BOURSIAC, Y., BOUDET, J., POSTAIRE, O., LUU, D. T., TOURNAIRE-ROUX, C. & MAUREL,
C. (2008a) Stimulus-induced downregulation of root water transport involves reactive oxy-
gen species-activated cell signalling and plasma membrane intrinsic protein internalization.
Plant J, 56, 207-18.
BOURSIAC, Y., PRAK, S., BOUDET, J., POSTAIRE, O., LUU, D. T., TOURNAIRE-ROUX, C.,
SANTONI, V. & MAUREL, C. (2008b) The response of Arabidopsis root water transport to
a challenging environment implicates reactive oxygen species- and phosphorylation-depen-
dent internalization of aquaporins. Plant Signal Behav, 3, 1096-8.
BOWMAN, J. L., FLOYD, S. K. & SAKAKIBARA, K. (2007) Green genes-comparative genomics of
the green branch of life. Cell, 129, 229-34.
CARBREY, J. M. & AGRE, P. (2009) Discovery of the aquaporins and development of the field.
Handb Exp Pharmacol, 3-28.
CHAUMONT, F., BARRIEU, F., WOJCIK, E., CHRISPEELS, M. J. & JUNG, R. (2001) Aqua-
porins constitute a large and highly divergent protein family in maize. Plant Physiol, 125,
1206-15.
CHIBA, Y., MITANI, N., YAMAJI, N. & MA, J. F. (2009) HvLsi1 is a silicon influx transporter in
barley. Plant J, 57, 810-8.
CHOI, W. G. & ROBERTS, D. M. (2007) Arabidopsis NIP2;1: A major intrinsic protein transporter
of lactic acid induced by anoxic stress. J Biol Chem, 282, 24209-18.
CUKIERMAN, S. (2000) Proton mobilities in water and in different stereoisomers of covalently linked
gramicidin A channels. Biophys J, 78, 1825-34.
DAGAN, T., ARTZY-RANDRUP, Y. & MARTIN, W. (2008) Modular networks and cumulative im-
pact of lateral transfer in prokaryote genome evolution. Proc Natl Acad Sci U S A, 105,
10039-44.
DENKER, B. M., SMITH, B. L., KUHAJDA, F. P. & AGRE, P. (1988) Identification, purification,
and partial characterization of a novel Mr 28,000 integral membrane protein from erythro-
cytes and renal tubules. J Biol Chem, 263, 15634-42.
49

DRMANAC, R., SPARKS, A. B., CALLOW, M. J., HALPERN, A. L., BURNS, N. L., KERMANI,
B. G., CARNEVALI, P., NAZARENKO, I., NILSEN, G. B., YEUNG, G., DAHL, F.,
FERNANDEZ, A., STAKER, B., PANT, K. P., BACCASH, J., BORCHERDING, A. P.,
BROWNLEY, A., CEDENO, R., CHEN, L., CHERNIKOFF, D., CHEUNG, A., CHIR-
ITA, R., CURSON, B., EBERT, J. C., HACKER, C. R., HARTLAGE, R., HAUSER, B.,
HUANG, S., JIANG, Y., KARPINCHYK, V., KOENIG, M., KONG, C., LANDERS, T.,
LE, C., LIU, J., MCBRIDE, C. E., MORENZONI, M., MOREY, R. E., MUTCH, K.,
PERAZICH, H., PERRY, K., PETERS, B. A., PETERSON, J., PETHIYAGODA, C. L.,
POTHURAJU, K., RICHTER, C., ROSENBAUM, A. M., ROY, S., SHAFTO, J., SHA-
RANHOVICH, U., SHANNON, K. W., SHEPPY, C. G., SUN, M., THAKURIA, J. V.,
TRAN, A., VU, D., ZARANEK, A. W., WU, X., DRMANAC, S., OLIPHANT, A. R.,
BANYAI, W. C., MARTIN, B., BALLINGER, D. G., CHURCH, G. M. & REID, C. A.
(2010) Human genome sequencing using unchained base reads on self-assembling DNA
nanoarrays. Science, 327, 78-81.
DYNOWSKI, M., MAYER, M., MORAN, O. & LUDEWIG, U. (2008a) Molecular determinants of
ammonia and urea conductance in plant aquaporin homologs. FEBS Lett, 582, 2458-62.
DYNOWSKI, M., SCHAAF, G., LOQUE, D., MORAN, O. & LUDEWIG, U. (2008b) Plant plasma
membrane water channels conduct the signalling molecule H2
O2
. Biochem J, 414, 53-61.
FETTER, K., VAN WILDER, V., MOSHELION, M. & CHAUMONT, F. (2004) Interactions be-
tween plasma membrane aquaporins modulate their water channel activity. Plant Cell, 16,
215-28.
FISCHER, G., KOSINSKA-ERIKSSON, U., APONTE-SANTAMARIA, C., PALMGREN, M.,
GEIJER, C., HEDFALK, K., HOHMANN, S., DE GROOT, B. L., NEUTZE, R. &
LINDKVIST-PETERSSON, K. (2009) Crystal structure of a yeast aquaporin at 1.15 ång-
ström reveals a novel gating mechanism. PLoS Biol, 7, e1000130.
FORTIN, M. G., MORRISON, N. A. & VERMA, D. P. (1987) Nodulin-26, a peribacteroid mem-
brane nodulin is expressed independently of the development of the peribacteroid compart-
ment. Nucleic Acids Res, 15, 813-24.
FRIGERIO, L., HINZ, G. & ROBINSON, D. G. (2008) Multiple vacuoles in plant cells: rule or
exception? Traffic, 9, 1564-70.
FROGER, A., TALLUR, B., THOMAS, D. & DELAMARCHE, C. (1998) Prediction of functional
residues in water channels and related proteins. Protein Sci, 7, 1458-68.
FU, D., LIBSON, A., MIERCKE, L. J., WEITZMAN, C., NOLLERT, P., KRUCINSKI, J. &
STROUD, R. M. (2000) Structure of a glycerol-conducting channel and the basis for its
selectivity. Science, 290, 481-6.
50

GATTOLIN, S., SORIEUL, M. & FRIGERIO, L. (2010a) Mapping of Tonoplast Intrinsic Proteins in
Maturing and Germinating Arabidopsis Seeds Reveals Dual Localization of Embryonic TIPs
to the Tonoplast and Plasma Membrane. Mol Plant. doi: 10.1093/mp/ssq051
GATTOLIN, S., SORIEUL, M. & FRIGERIO, L. (2010b) Tonoplast intrinsic proteins and vacuolar
identity. Biochem Soc Trans, 38, 769-73.
GATTOLIN, S., SORIEUL, M., HUNTER, P. R., KHONSARI, R. H. & FRIGERIO, L. (2009) In
vivo imaging of the tonoplast intrinsic protein family in Arabidopsis roots. BMC Plant Biol,
9, 133.
GAZZARRINI, S., KANG, M., EPIMASHKO, S., VAN ETTEN, J. L., DAINTY, J., THIEL, G. &
MORONI, A. (2006) Chlorella virus MT325 encodes water and potassium channels that
interact synergistically. Proc Natl Acad Sci U S A, 103, 5355-60.
GORIN, M. B., YANCEY, S. B., CLINE, J., REVEL, J. P. & HORWITZ, J. (1984) The major intrin-
sic protein (MIP) of the bovine lens fiber membrane: characterization and structure based on
cDNA cloning. Cell, 39, 49-59.
GONEN, T., SLIZ, P., KISTLER, J., CHENG, Y. & WALZ, T. (2004) Aquaporin-0 membrane junc-
tions reveal the structure of a closed water pore. Nature, 429, 193-7.
GUPTA, A. B. & SANKARARAMAKRISHNAN, R. (2009) Genome-wide analysis of major intrinsic
proteins in the tree plant Populus trichocarpa: characterization of XIP subfamily of aqua-
porins from evolutionary perspective. BMC Plant Biol, 9, 134.
GUSTAVSSON, S., LEBRUN, A. S., NORDEN, K., CHAUMONT, F. & JOHANSON, U. (2005)
A novel plant major intrinsic protein in Physcomitrella patens most similar to bacterial glyc-
erol channels. Plant Physiol, 139, 287-95.
HACHEZ, C., MOSHELION, M., ZELAZNY, E., CAVEZ, D. & CHAUMONT, F. (2006) Local-
ization and quantification of plasma membrane aquaporin expression in maize primary root:
a clue to understanding their role as cellular plumbers. Plant Mol Biol, 62, 305-23.
HANBA, Y. T., SHIBASAKA, M., HAYASHI, Y., HAYAKAWA, T., KASAMO, K., TERASHIMA, I.
& KATSUHARA, M. (2004) Overexpression of the barley aquaporin HvPIP2;1 increases
internal CO2
conductance and CO2
assimilation in the leaves of transgenic rice plants. Plant
Cell Physiol, 45, 521-9.
HARRIES, W. E., AKHAVAN, D., MIERCKE, L. J., KHADEMI, S. & STROUD, R. M. (2004) The
channel architecture of aquaporin 0 at a 2.2-A resolution. Proc Natl Acad Sci U S A, 101,
14045-50.
51

HENZLER, T., YE, Q. & STEUDLE, E. (2004) Oxidative gating of water channels (aquaporins) in
Chara by hydroxyl radicals. Plant Cell and Environment, 27, 1184-1195.
HEYMANN, J. B. & ENGEL, A. (1999) Aquaporins: Phylogeny, Structure, and Physiology of Water
Channels. News Physiol Sci, 14, 187-193.
HIGUCHI, T., SUGA, S., TSUCHIYA, T., HISADA, H., MORISHIMA, S., OKADA, Y. & MAE-
SHIMA, M. (1998) Molecular cloning, water channel activity and tissue specific expression
of two isoforms of radish vacuolar aquaporin. Plant Cell Physiol, 39, 905-13.
HUB, J. S. & DE GROOT, B. L. (2008) Mechanism of selectivity in aquaporins and aquaglycer-
oporins. Proc Natl Acad Sci U S A, 105, 1198-203.
HUNTER, P. R., CRADDOCK, C. P., DI BENEDETTO, S., ROBERTS, L. M. & FRIGERIO, L.
(2007) Fluorescent reporter proteins for the tonoplast and the vacuolar lumen identify a
single vacuolar compartment in Arabidopsis cells. Plant Physiol, 145, 1371-82.
ISAYENKOV, S. V. & MAATHUIS, F. J. (2008) The Arabidopsis thaliana aquaglyceroporin AtNIP7;1
is a pathway for arsenite uptake. FEBS Lett, 582, 1625-8.
ISHIKAWA, F., SUGA, S., UEMURA, T., SATO, M. H. & MAESHIMA, M. (2005) Novel type
aquaporin SIPs are mainly localized to the ER membrane and show cell-specific expression
in Arabidopsis thaliana. FEBS Lett, 579, 5814-20.
JAUH, G. Y., PHILLIPS, T. E. & ROGERS, J. C. (1999) Tonoplast intrinsic protein isoforms as mark-
ers for vacuolar functions. Plant Cell, 11, 1867-82.
JOHANSON, U. & GUSTAVSSON, S. (2002) A new subfamily of major intrinsic proteins in plants.
Mol Biol Evol, 19, 456-61.
JOHANSON, U., KARLSSON, M., JOHANSSON, I., GUSTAVSSON, S., SJOVALL, S., FRAY-
SSE, L., WEIG, A. R. & KJELLBOM, P. (2001) The complete set of genes encoding major
intrinsic proteins in Arabidopsis provides a framework for a new nomenclature for major
intrinsic proteins in plants. Plant Physiol, 126, 1358-69.
JOHANSSON, I., LARSSON, C., EK, B. & KJELLBOM, P. (1996) The major integral proteins of
spinach leaf plasma membranes are putative aquaporins and are phosphorylated in response
to Ca2+
and apoplastic water potential. Plant Cell, 8, 1181-91.
JOHNSON, K. D., HERMAN, E. M. & CHRISPEELS, M. J. (1989) An Abundant, Highly Con-
served Tonoplast Protein in Seeds. Plant Physiol, 91, 1006-1013.
KAMIYA, T., TANAKA, M., MITANI, N., MA, J. F., MAESHIMA, M. & FUJIWARA, T. (2009)
NIP1;1, an aquaporin homolog, determines the arsenite sensitivity of Arabidopsis thaliana.
J Biol Chem, 284, 2114-20.
52

KAMMERLOHER, W., FISCHER, U., PIECHOTTKA, G. P. & SCHAFFNER, A. R. (1994) Water
channels in the plant plasma membrane cloned by immunoselection from a mammalian
expression system. Plant J, 6, 187-99.
KATSUHARA, M., HANBA, Y. T., SHIRATAKE, K. & MAESHIMA, M. (2008) Review: Expanding
roles of plant aquaporins in plasma membranes and cell organelles. Functional Plant Biology,
35, 1-14.
KHANDELIA, H., JENSEN, M. O. & MOURITSEN, O. G. (2009) To gate or not to gate: using
molecular dynamics simulations to morph gated plant aquaporins into constitutively open
conformations. J Phys Chem B, 113, 5239-44.
KJELLBOM, P. & LARSSON, C. (1984) Preparation and Polypeptide Composition of Chlorophyll-
Free Plasma-Membranes from Leaves of Light-Grown Spinach and Barley. Physiologia Plan-
tarum, 62, 501-509.
KOZONO, D., DING, X., IWASAKI, I., MENG, X., KAMAGATA, Y., AGRE, P. & KITAGAWA,
Y. (2003) Functional expression and characterization of an archaeal aquaporin. AqpM from
methanothermobacter marburgensis. J Biol Chem, 278, 10649-56.
LEE, H. K., CHO, S. K., SON, O., XU, Z., HWANG, I. & KIM, W. T. (2009) Drought stress-
induced Rma1H1, a RING membrane-anchor E3 ubiquitin ligase homolog, regulates aqua-
porin levels via ubiquitination in transgenic Arabidopsis plants. Plant Cell, 21, 622-41.
LIN, W., PENG, Y., LI, G., ARORA, R., TANG, Z., SU, W. & CAI, W. (2007) Isolation and func-
tional characterization of PgTIP1, a hormone-autotrophic cells-specific tonoplast aquaporin
in ginseng. J Exp Bot, 58, 947-56.
LIN, Y. C., HSU, J. Y., SHU, J. H., CHI, Y., CHIANG, S. C. & LEE, S. T. (2008) Two distinct arsen-
ite-resistant variants of Leishmania amazonensis take different routes to achieve resistance as
revealed by comparative transcriptomics. Mol Biochem Parasitol, 162, 16-31.
LUDEVID, D., HOFTE, H., HIMELBLAU, E. & CHRISPEELS, M. J. (1992) The Expression Pat-
tern of the Tonoplast Intrinsic Protein gamma-TIP in Arabidopsis thaliana Is Correlated with
Cell Enlargement. Plant Physiol, 100, 1633-9.
MA, J. F., TAMAI, K., YAMAJI, N., MITANI, N., KONISHI, S., KATSUHARA, M., ISHIGURO,
M., MURATA, Y. & YANO, M. (2006) A silicon transporter in rice. Nature, 440, 688-91.
MA, J. F., YAMAJI, N., MITANI, N., XU, X. Y., SU, Y. H., MCGRATH, S. P. & ZHAO, F. J. (2008)
Transporters of arsenite in rice and their role in arsenic accumulation in rice grain. Proc Natl
Acad Sci U S A, 105, 9931-5.
MA, S., QUIST, T. M., ULANOV, A., JOLY, R. & BOHNERT, H. J. (2004) Loss of TIP1;1 aqua-
porin in Arabidopsis leads to cell and plant death. Plant J, 40, 845-59.
53

MAESHIMA, M. & ISHIKAWA, F. (2007) ER membrane aquaporins in plants. Pflugers Arch, 456,
709-16.
MAUREL, C., REIZER, J., SCHROEDER, J. I. & CHRISPEELS, M. J. (1993) The vacuolar mem-
brane protein gamma-TIP creates water specific channels in Xenopus oocytes. EMBO J, 12,
2241-7.
MAUREL, C., TACNET, F., GUCLU, J., GUERN, J. & RIPOCHE, P. (1997) Purified vesicles of to-
bacco cell vacuolar and plasma membranes exhibit dramatically different water permeability
and water channel activity. Proc Natl Acad Sci U S A, 94, 7103-8.
MAUREL, C., VERDOUCQ, L., LUU, D. T. & SANTONI, V. (2008) Plant aquaporins: membrane
channels with multiple integrated functions. Annu Rev Plant Biol, 59, 595-624.
MITANI, N., YAMAJI, N. & MA, J. F. (2009) Identification of maize silicon influx transporters. Plant
Cell Physiol, 50, 5-12.
MIZUTANI, M., WATANABE, S., NAKAGAWA, T. & MAESHIMA, M. (2006) Aquaporin NIP2;1
is Mainly Localized to the ER Membrane and Shows Root-Specific Accumulation in Arabi-
dopsis thaliana. Plant Cell Physiol, 47, 1420-6.
MORISHITA, Y., SAKUBE, Y., SASAKI, S. & ISHIBASHI, K. (2004) Molecular mechanisms and
drug development in aquaporin water channel diseases: aquaporin superfamily (superaqua-
porins): expansion of aquaporins restricted to multicellular organisms. J Pharmacol Sci, 96,
276-9.
MOSHELION, M., BECKER, D., BIELA, A., UEHLEIN, N., HEDRICH, R., OTTO, B., LEVI,
H., MORAN, N. & KALDENHOFF, R. (2002) Plasma membrane aquaporins in the motor
cells of Samanea saman: diurnal and circadian regulation. Plant Cell, 14, 727-39.
MUNROE, D. J. & HARRIS, T. J. (2010) Third-generation sequencing fireworks at Marco Island.
Nat Biotechnol, 28, 426-8.
MURAMATSU, S. & MIZUNO, T. (1989) Nucleotide sequence of the region encompassing the
glpKF operon and its upstream region containing a bent DNA sequence of Escherichia coli.
Nucleic Acids Res, 17, 4378.
MURATA, K., MITSUOKA, K., HIRAI, T., WALZ, T., AGRE, P., HEYMANN, J. B., ENGEL, A. &
FUJIYOSHI, Y. (2000) Structural determinants of water permeation through aquaporin-1.
Nature, 407, 599-605.
NEDVETSKY, P. I., TAMMA, G., BEULSHAUSEN, S., VALENTI, G., ROSENTHAL, W. &
KLUSSMANN, E. (2009) Regulation of aquaporin-2 trafficking. Handb Exp Pharmacol,
133-57.
54

OKUBO-KURIHARA, E., SANO, T., HIGAKI, T., KUTSUNA, N. & HASEZAWA, S. (2009) Ac-
celeration of vacuolar regeneration and cell growth by overexpression of an aquaporin Nt-
TIP1;1 in tobacco BY-2 cells. Plant Cell Physiol, 50, 151-60.
OLBRICH, A., HILLMER, S., HINZ, G., OLIVIUSSON, P. & ROBINSON, D. G. (2007) Newly
formed vacuoles in root meristems of barley and pea seedlings have characteristics of both
protein storage and lytic vacuoles. Plant Physiol, 145, 1383-94.
OTTO, B., UEHLEIN, N., SDORRA, S., FISCHER, M., AYAZ, M., BELASTEGUI-MACADAM,
X., HECKWOLF, M., LACHNIT, M., PEDE, N., PRIEM, N., REINHARD, A., SIEG-
FART, S., URBAN, M. & KALDENHOFF, R. (2010) Aquaporin tetramer composition
modifies the function of tobacco aquaporins. J Biol Chem. 285, 31253-60
PANG, Y., LI, L., REN, F., LU, P., WEI, P., CAI, J., XIN, L., ZHANG, J., CHEN, J. & WANG, X.
(2010) Overexpression of the tonoplast aquaporin AtTIP5;1 conferred tolerance to boron
toxicity in Arabidopsis. J Genet Genomics, 37, 389-97, 1-2.
PAO, G. M., WU, L. F., JOHNSON, K. D., HOFTE, H., CHRISPEELS, M. J., SWEET, G., SAN-
DAL, N. N. & SAIER, M. H., JR. (1991) Evolution of the MIP family of integral mem-
brane transport proteins. Mol Microbiol, 5, 33-7.
PARK, J. H. & SAIER, M. H., JR. (1996) Phylogenetic characterization of the MIP family of trans-
membrane channel proteins. J Membr Biol, 153, 171-80.
PARK, W., SCHEFFLER, B. E., BAUER, P. J. & CAMPBELL, B. T. (2010) Identification of the
family of aquaporin genes and their expression in upland cotton (Gossypium hirsutum L.).
BMC Plant Biol, 10, 142.
PENG, Y., LIN, W., CAI, W. & ARORA, R. (2007) Overexpression of a Panax ginseng tonoplast
aquaporin alters salt tolerance, drought tolerance and cold acclimation ability in transgenic
Arabidopsis plants. Planta, 226, 729-40.
PHONGPHANPHANEE, S., YOSHIDA, N. & HIRATA, F. (2010) Molecular selectivity in aqua-
porin channels studied by the 3D-RISM theory. J Phys Chem B, 114, 7967-73.
POSTAIRE, O., TOURNAIRE-ROUX, C., GRONDIN, A., BOURSIAC, Y., MORILLON, R.,
SCHAFFNER, A. R. & MAUREL, C. (2010) A PIP1 aquaporin contributes to hydrostatic
pressure-induced water transport in both the root and rosette of Arabidopsis. Plant Physiol,
152, 1418-30.
PRAK, S., HEM, S., BOUDET, J., VIENNOIS, G., SOMMERER, N., ROSSIGNOL, M., MAU-
REL, C. & SANTONI, V. (2008) Multiple phosphorylations in the C-terminal tail of plant
plasma membrane aquaporins: role in subcellular trafficking of AtPIP2;1 in response to salt
stress. Mol Cell Proteomics, 7, 1019-30.
55

PRESTON, G. M. & AGRE, P. (1991) Isolation of the cDNA for erythrocyte integral membrane
protein of 28 kilodaltons: member of an ancient channel family. Proc Natl Acad Sci U S A,
88, 11110-4.
PRESTON, G. M., CARROLL, T. P., GUGGINO, W. B. & AGRE, P. (1992) Appearance of water
channels in Xenopus oocytes expressing red cell CHIP28 protein. Science, 256, 385-7.
REISEN, D., LEBORGNE-CASTEL, N., OZALP, C., CHAUMONT, F. & MARTY, F. (2003) Ex-
pression of a cauliflower tonoplast aquaporin tagged with GFP in tobacco suspension cells
correlates with an increase in cell size. Plant Mol Biol, 52, 387-400.
REIZER, J., REIZER, A. & SAIER, M. H., JR. (1993) The MIP family of integral membrane channel
proteins: sequence comparisons, evolutionary relationships, reconstructed pathway of evolu-
tion, and proposed functional differentiation of the two repeated halves of the proteins. Crit
Rev Biochem Mol Biol, 28, 235-57.
RICHEY, D. P. & LIN, E. C. (1972) Importance of facilitated diffusion for effective utilization of
glycerol by Escherichia coli. J Bacteriol, 112, 784-90.
ROBINSON, D. G., SIEBER, H., KAMMERLOHER, W. & SCHAFFNER, A. R. (1996) PIP1
Aquaporins Are Concentrated in Plasmalemmasomes of Arabidopsis thaliana Mesophyll.
Plant Physiol, 111, 645-649.
SADE, N., GEBRETSADIK, M., SELIGMANN, R., SCHWARTZ, A., WALLACH, R. & MOSHE-
LION, M. (2010) The role of tobacco Aquaporin1 in improving water use efficiency, hydrau-
lic conductivity, and yield production under salt stress. Plant Physiol, 152, 245-54.
SADE, N., VINOCUR, B. J., DIBER, A., SHATIL, A., RONEN, G., NISSAN, H., WALLACH, R.,
KARCHI, H. & MOSHELION, M. (2009) Improving plant stress tolerance and yield pro-
duction: is the tonoplast aquaporin SlTIP2;2 a key to isohydric to anisohydric conversion?
New Phytol, 181, 651-61.
SAKURAI, J., ISHIKAWA, F., YAMAGUCHI, T., UEMURA, M. & MAESHIMA, M. (2005) Iden-
tification of 33 rice aquaporin genes and analysis of their expression and function. Plant Cell
Physiol, 46, 1568-77.
SANDAL, N. N. & MARCKER, K. A. (1988) Soybean nodulin 26 is homologous to the major intrin-
sic protein of the bovine lens fiber membrane. Nucleic Acids Res, 16, 9347.
SANTONI, V., VERDOUCQ, L., SOMMERER, N., VINH, J., PFLIEGER, D. & MAUREL, C.
(2006) Methylation of aquaporins in plant plasma membrane. Biochem J, 400, 189-97.
SAVAGE, D. F., O’CONNELL, J. D., 3RD, MIERCKE, L. J., FINER-MOORE, J. & STROUD, R.
M. (2010) Structural context shapes the aquaporin selectivity filter. Proc Natl Acad Sci U S
A. 107, 17164-9.
56

SCHUSSLER, M. D., ALEXANDERSSON, E., BIENERT, G. P., KICHEY, T., LAURSEN, K. H.,
JOHANSON, U., KJELLBOM, P., SCHJOERRING, J. K. & JAHN, T. P. (2008) The
effects of the loss of TIP1;1 and TIP1;2 aquaporins in Arabidopsis thaliana. Plant J, 56,
756-67.
SECCHI, F. & ZWIENIECKI, M. A. (2010) Patterns of PIP gene expression in Populus trichocarpa
during recovery from xylem embolism suggest a major role for the PIP1 aquaporin subfamily
as moderators of refilling process. Plant Cell Environ, 33, 1285-97.
SHIELS, A., KENT, N. A., MCHALE, M. & BANGHAM, J. A. (1988) Homology of MIP26 to
Nod26. Nucleic Acids Res, 16, 9348.
SMITH, B. L. & AGRE, P. (1991) Erythrocyte Mr 28,000 transmembrane protein exists as a multisub-
unit oligomer similar to channel proteins. J Biol Chem, 266, 6407-15.
STEIN, W. D. & DANIELLI, J. F. (1956) Structure and function in red cell permeability. Discussions
of the Faraday Society, 21, 238-251.
SUI, H., HAN, B. G., LEE, J. K., WALIAN, P. & JAP, B. K. (2001) Structural basis of water-specific
transport through the AQP1 water channel. Nature, 414, 872-8.
TAIZ, L. & ZEIGER, E. (2006) Plant physiology, 4th
edition, Sunderland, Mass., Sinauer Associates.
TAKANO, J., WADA, M., LUDEWIG, U., SCHAAF, G., VON WIREN, N. & FUJIWARA, T.
(2006) The Arabidopsis major intrinsic protein NIP5;1 is essential for efficient boron uptake
and plant development under boron limitation. Plant Cell, 18, 1498-509.
TANAKA, M., WALLACE, I. S., TAKANO, J., ROBERTS, D. M. & FUJIWARA, T. (2008) NIP6;1
is a boric acid channel for preferential transport of boron to growing shoot tissues in Arabi-
dopsis. Plant Cell, 20, 2860-75.
TORNROTH-HORSEFIELD, S., WANG, Y., HEDFALK, K., JOHANSON, U., KARLSSON, M.,
TAJKHORSHID, E., NEUTZE, R. & KJELLBOM, P. (2006) Structural mechanism of
plant aquaporin gating. Nature, 439, 688-94.
UEHLEIN, N., LOVISOLO, C., SIEFRITZ, F. & KALDENHOFF, R. (2003) The tobacco aquaporin
NtAQP1 is a membrane CO2
pore with physiological functions. Nature, 425, 734-7.
UEHLEIN, N., OTTO, B., HANSON, D. T., FISCHER, M., MCDOWELL, N. & KALDEN-
HOFF, R. (2008) Function of Nicotiana tabacum aquaporins as chloroplast gas pores chal-
lenges the concept of membrane CO2 permeability. Plant Cell, 20, 648-57.
WALLACE, I. S. & ROBERTS, D. M. (2005) Distinct transport selectivity of two structural subclasses
of the nodulin-like intrinsic protein family of plant aquaglyceroporin channels. Biochemistry,
44, 16826-34.
57

WALLACE, I. S., WILLS, D. M., GUENTHER, J. F. & ROBERTS, D. M. (2002) Functional selec-
tivity for glycerol of the nodulin 26 subfamily of plant membrane intrinsic proteins. FEBS
Lett, 523, 109-12.
WUDICK, M. M., LUU, D. T. & MAUREL, C. (2009) A look inside: localization patterns and func-
tions of intracellular plant aquaporins. New Phytol, 184, 289-302.
YAMAJI, N., MITATNI, N. & MA, J. F. (2008) A transporter regulating silicon distribution in rice
shoots. Plant Cell, 20, 1381-9.
YE, Q., WIERA, B. & STEUDLE, E. (2004) A cohesion/tension mechanism explains the gating of
water channels (aquaporins) in Chara internodes by high concentration. J Exp Bot, 55, 449-
61.
ZARDOYA, R. (2005) Phylogeny and evolution of the major intrinsic protein family. Biol Cell, 97,
397-414.
ZARDOYA, R., DING, X., KITAGAWA, Y. & CHRISPEELS, M. J. (2002) Origin of plant glycerol
transporters by horizontal gene transfer and functional recruitment. Proc Natl Acad Sci U S
A, 99, 14893-6.
ZARDOYA, R. & VILLALBA, S. (2001) A phylogenetic framework for the aquaporin family in eu-
karyotes. J Mol Evol, 52, 391-404.
ZELAZNY, E., BORST, J. W., MUYLAERT, M., BATOKO, H., HEMMINGA, M. A. & CHAU-
MONT, F. (2007) FRET imaging in living maize cells reveals that plasma membrane aqua-
porins interact to regulate their subcellular localization. Proc Natl Acad Sci U S A, 104,
12359-64.
ZELAZNY, E., MIECIELICA, U., BORST, J. W., HEMMINGA, M. A. & CHAUMONT, F. (2009)
An N-terminal diacidic motif is required for the trafficking of maize aquaporins ZmPIP2;4
and ZmPIP2;5 to the plasma membrane. Plant J, 57, 346-55.
58

PhD thesis_Opt

More Related Content

What's hot (20)

Viewers also liked (18)

Similar to PhD thesis_Opt (20)

PhD thesis_Opt