Download Feeling green: mechanosensing in plants

Review
Feeling green: mechanosensing in
plants
Gabriele B. Monshausen and Simon Gilroy
Department of Botany, University of Wisconsin, Birge Hall, 430 Lincoln Drive, Madison, WI 53706, USA
Owing to the sessile nature of their lifestyle, plants have
to respond to a wide range of signals, such as the force of
the wind or the impedance of the soil, to entrain their
development to prevailing environmental conditions.
Indeed, mechanically responsive growth has been documented in plants for many years but new work on lateral
root formation strongly supports the idea that biophysical forces can elicit complete de novo developmental programs. In addition, only recently have molecular
candidates for plant mechanosensors emerged. Such
advances in understanding plant mechanoresponsive
development have relied heavily on comparison with
mechanosensors characterized in organisms such as
Saccharomyces cerevisiae and Escherichia coli, but
key questions remain about the cellular basis of the
plant mechanosensory system.
Mechanical forces and morphogenesis
Mechanical forces have a crucial role in plant morphogenesis, whether it be the sculpting of a tree by the wind, the
twining of a tendril as a vine grows up a support, or the
development of the root system as it navigates past rocks in
the soil. In all these cases, mechanical sensing and response
have dramatic effects on the final form a plant will adopt.
Although remarkably detailed descriptions of these mechanical responses have been available for >100 years (e.g. in
Darwin’s classic treatise on The Power of Movement in
Plants from 1880 [1]), the molecular mechanisms behind
these processes have begun to be unraveled only recently.
Plants have evolved two broad classes of mechanical
response. The first is characterized by extremely rapid
movements of organs in response to touch stimulation,
as exemplified by the closing of the trap of the Venus
flytrap (Dionaea muscipula) or the rapid folding of the
leaflets of the sensitive fern (Mimosa pudica; Box 1). In
both of these cases, rapid organ movements are linked to a
highly specialized mechanosensory apparatus [2,3]. By
contrast, a second broad class of mechanical response
occurs in all plants over developmental time. Thus, plants
subjected to mechanical stimulation tend to be shorter and
more robust, develop more support tissues, and entrain
their overall architecture to the prevailing mechanical
forces in their vicinity.
Mechanical forces as morphogenetic factors
Although mechanical forces can clearly shape plant form
through alterations in the growth habit of existing organs,
Corresponding author: Gilroy, S. ([email protected]).
228
they have also long been proposed to act as plant morphogenetic factors [4,5]. However, evidence for mechanical
forces reprogramming development has been largely indirect. For example, microactivation of an inducible transgene
was used to drive local production of the expansin proteins
that support cell enlargement in the apical meristem of the
plant. This treatment was shown to induce a developmental program that leads to leaf formation [6]. Such observations imply that alterations in the biomechanics of the
apex might induce new organ formation. However, recent
work on lateral root production provides evidence for a
direct role of mechanical stimuli in such developmental
reprogramming.
Lateral root formation in plants represents postembryonic organ formation whereby cells in the pericycle
of the root are directed to a lateral root founder cell fate,
undergo divisions and form a primordium that will develop
into an emergent lateral root [7] (Figure 1a). Although
periodic maxima in the levels of the hormone auxin could
well pre-pattern the distribution of these founder cells, and
although auxin is known to have a crucial role in subsequent lateral root development [8,9], the positioning of
founder events is also highly entrained to the environment
[10–12]. As early as 1900, Noll recognized that a major
determinant of the positioning of the lateral roots was the
physical architecture of the main root, with laterals emerging on the convex side of curves in the root system [13].
One explanation of this bend-related phenomenon [14] is
that lateral root induction is being triggered by the redistribution of auxin that occurs as roots undergo directional
(tropic) growth responses as they make the curves [15].
However, recent evidence indicates that biomechanics are
actually responsible for reprogramming a pericycle cell to a
founder cell fate. Thus, manually bending the root can
elicit lateral root formation [16,17] (G.L. Richter, G.A.B., A.
Krol and S.G., unpublished). However, it is important to
note that manual bending represents an extreme stimulus
that has the potential to elicit responses unrelated to
normal plant development. Crucially, therefore, this same
phenomenon can be generated by means of the endogenous
forces of the root growing through soil. Thus, when a root
encounters a barrier to growth such as a rock or a hardpan
layer of soil, it adopts an avoidance response to circumnavigate the obstacle [18] (Figure 1b). The initial bending of
the root to grow round the object is dominated by mechanical responses, and the curve that is elicited by the
biophysical forces at play leads to lateral root formation
to the convex side (i.e. the endogenous forces of growth can
elicit bend-related organ formation) (Figure 1c). In this
0962-8924/$ – see front matter ß 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.tcb.2009.02.005 Available online 1 April 2009
Review
Trends in Cell Biology
Vol.19 No.5
Box 1. Rapid leaflet movement in M. pudica
Mechanoreceptors in the leaf of M. pudica elicit an electrical,
chemical or, possibly, hydraulic signal that moves to the leaflet
base. Specialized motor organs (pulvini) release ions from their
cells, causing a loss of internal hydrostatic pressure (turgor; for a
review, see Ref. [2]). Within 1 s, water equivalent to up to 25% of the
cell volume is lost via osmosis. Such rapid water movements are
probably sustained by aquaporins or other water transporters in the
plasma membrane of the pulvinar cells. The loss of turgor in these
cells at the base of each leaflet causes the leaflet to very rapidly fold
inwards (Figure I). The larger the mechanical stimulus, the further
the signal travels, leading to a progressive wave of leaflet movement. With sufficiently intense stimuli, the entire leaf can fold
downwards. These movements are reversible, with recovery
occurring as ions are pumped back into the pulvinar cells, with an
accompanying uptake of water and reinstatement of turgor
pressure. Pulvini have abundant H+ATPase activity, probably to
sustain these ion fluxes. These rapid movements are thought to
either startle potential herbivores or to make the plant seem less
appetizing (for reviews, see Refs [2,3]).
Figure I. Leaflet folding after touch stimulation in M. pudica.
way, the main root axis grows in one direction and the
emerging lateral root axis in the other, optimizing soil
exploration and anchorage. Although current data indicate
that the mechanical signaling system lies upstream of or
operates in parallel to the auxin-dependent system to elicit
lateral root formation [16], the mechanosensor(s) leading
to this reprogramming of development, or indeed any of the
mechanosensors that lead to the many other features of
mechanically entrained plant development, remain to be
defined. Current ideas about these elusive receptors for
mechanical signals in the plant rely on comparisons with
mechanotransduction sensors in other organisms and include involvement of ion channels, osmotic sensors and
cell-wall associated kinases.
Mechanical sensing and ion fluxes
There is much evidence from patch clamp analyses that the
plasma membranes of plant cells contain a wide diversity
of mechanosensitive ion channels [3]. However, none of
these mechanosensitive conductances characterized
electrophysiologically have to date been identified to the
molecular level. There is also a wealth of evidence linking
changes in ion fluxes (principally Ca2+ fluxes) to mechanoresponses at the whole plant level, with mechanical
stimulation ranging from touch to wind disturbance (for
Figure 1. Lateral root formation in Arabidopsis. (a) Lateral roots initiate by the
induction of asymmetrical cell divisions in two adjacent cells of the pericycle, a cell
layer surrounding the vasculature (i). (ii) Continued divisions of these founder cells
yields a primordium that eventually becomes organized into a lateral root (iii) that
forces its way through overlying tissues to emerge from the surface of the primary
root [7]. (b) Upon encountering a barrier to downward growth, the root undergoes an
avoidance response that is characterized by an initial sliding sideways. This is
dominated by mechanical responses. A second phase of directional tropic growth
enables the root tip to traverse the barrier. This avoidance response can be imposed
experimentally by growing the root into a coverslip placed perpendicularly to the
axis of growth [18]. (c) In response to barrier avoidance, a lateral root forms to the
convex side of a bend in the Arabidopsis primary root. This curve was formed during
the initial mechanically related phase of the avoidance response (photo courtesy of
G.L. Richter, G.B.M., A. Krol and S.G.). Scale bar, 500 mm.
a review, see Ref. [3]; Figure 2a,b), and possibly even to
responses to gravity [19,20], leading to a transient increase
in cytosolic Ca2+ levels.
Recent work has also highlighted a role for proton fluxes
and reactive oxygen species (ROS) production in mechanosensation, both of which might contribute to linking Ca2+
signals to downstream developmental responses. Thus,
suspension cells of Taxus (yew), parsley and soybean
respond to mechanical stimulation with an oxidative burst
and an alkalinization of the culture medium within 2–
10 min [21–23], although the temporal resolution of these
measurements could not precisely determine the onset of
these ROS and pH changes at a cellular level. Intriguingly,
similar changes have been observed in tip-growing cells
such as pollen tubes and root hairs [24–26]. During tip
growth, highly localized wall loosening processes are carefully balanced with restriction of turgor-driven expansion
to sustain cell elongation. During peak expansion, relaxation of the cell wall will be accompanied by stretching of the
tightly appressed plasma membrane; this is similar to the
stretching that occurs during external mechanical deformation of the cell. Thus, it is not surprising that a burst in
229
Review
Trends in Cell Biology Vol.19 No.5
duction, modulating the activity of membrane transporters
and regulating gene transcription, the accompanying
alterations in extracellular ROS and alkalinization are
likely to directly impact cell wall rigidity and, therefore,
growth by promoting the formation of intermolecular crosslinks (e.g. see Refs [29,30]; Figure 2c).
The mechanoreceptor in plants
Despite such evidence for a central role of Ca2+-dependent
mechanical signaling in plants, there are at present no
molecularly identified mechanoresponsive receptors or
channels supporting these changes. Indeed, a paucity of
obvious molecular candidates for a mechanically sensitive
Ca2+ channel in plant genome sequences has meant that
the search for the plant mechanoresponsive receptor has
largely revolved around models of mechanoresponsive
elements defined in other kingdoms.
Figure 2. Mechanical stimulation induces ion fluxes in plants. (a) Mechanical
stimulation, provided by directing a stream of air onto the aerial parts of a plant
expressing the Ca2+-dependent luminescent protein aequorin, revealed that
mechanical signals lead to rapid and transient increases in cytosolic Ca2+. Cold
shock (0 8C) was used as a control stimulus also known to elicit large scale
increases in cytosolic Ca2+ in plants (adapted from Ref. [65]). (b) Repeated
mechanical stimulation leads to attenuation of subsequent Ca2+ responses
(adapted from Ref. [66]), suggesting either downregulation of the
mechanosensor upon repeated stimulation, or exhaustion of the store of Ca2+
being released to increase cytosolic ion levels. The recovery after a period of no
stimulation could represent refilling of the discharged Ca2+ stores or
resensitization of the mechanoreceptor system. (c) Model of possible plasma
membrane signaling events upon mechanical stimulation drawn from events seen
during turgor-driven root-hair and pollen-tube elongation (e.g. see Refs [24–
28,67]). Membrane tension leads to opening of a Ca2+-permeable channel. A
cytosolic Ca2+ increase then activates a proton transporter leading to cytosolic
acidification and cell-wall alkalinization. Simultaneously, at least in root hairs, the
NADPH oxidase ATRBOHC is activated through its Ca2+-binding EF hand motifs
leading to ROS production to the cell wall [68]. Wall ROS might then leak back to
the cytoplasm. Cytosolic ROS, pH decrease and Ca2+ increase are all known to elicit
signaling events. Similarly, elevated wall pH and ROS are known to rigidify the
wall matrix (e.g. see Refs [29,30]). Thus, this Ca2+-dependent system can directly
modulate growth, through altering wall properties, and can elicit the cytoplasmic
signaling cascades known to be linked to mechanical response.
tip growth is rapidly succeeded by an elevation of cytosolic
Ca2+ [25,27,28], followed by extracellular ROS (superoxide)
production [26] and influx of protons [25,26]. The timing of
these events indicates that Ca2+ triggers the subsequent
H+ and ROS responses. Although these changes in cytosolic
Ca2+, ROS and pH could all contribute to signal trans230
Models for plant mechanosensation I: the role of
channels
Although a variety of mechanosensitive channels have
been identified across a broad range of organisms, only
the mechanosensitive channels of bacteria have so far
proven to be a useful model for potential plant mechanoperception. For example, the transient receptor potential
(TRP) channels of animal cells (and the yeast TRPY homolog) [31,32], the DEG/ENaC voltage-independent Na+
channel family [31,33], and the TREK K+ channel family
[32] are all strong candidates for mechano- and osmosensitive channels in a range of cell types. However, there
seem to be no clear homologs of these mechanosensitive
channels in any of the currently sequenced plant genomes.
Fortunately, the bacterial mechanoresponsive channel of
small conductance (MscS) has proven an important starting point for identifying at least one family of plant
mechanosensor candidates.
The MSL gene family
The mechanosensitive channels of small (MscS) and large
(MscL) conductance represent osmotic safety valves for
bacteria [34]. When the bacterium encounters a sudden
drop in the osmotic strength of its environment, the channels open, enabling the efflux of solutes to prevent cellular
bursting. The channels are formed of multimers with an
iris-like pore, which is opened by the increasing tension in
the membrane as the cell begins to swell (Figure 3a,b).
There are six homologs of the MscS channels in the rice
genome and ten MscS-like (MSL) genes in Arabidopsis [35]
but, to date, no clear mechanoresponsive whole plant
phenotype has been found in MSL knockout mutants
generated in Arabidopsis. However, MSL2 and MSL3
are localized to the plastids where they have a redundant
role in plastid division, as does their ortholog MSC in
Chlamydomonas [35]. These observations are consistent
with a conserved role for MscS in the division of the
endosymbiotic progenitors of the chloroplast.
Of the other eight Arabidopsis MSL genes [35], MSL9
and MSL10 have received most attention to date. The
proteins are found mainly in the plasma membrane of root
cells. They are expressed in an overlapping but not identical pattern, with both appearing in cortical cells [36]. An
Review
Trends in Cell Biology
Vol.19 No.5
they are required to facilitate mechanosensitive gating of a
probably Cl -permeable channel in the plasma membrane
[36]. The physiological function of these proteins also
remains enigmatic. Apart from the intriguing role in plastid division of msl2 and msl3 mutants, knockouts in the
other MSL genes have no overt phenotype. Even the
quintuple mutant of all root-expressed MSL genes
(msl4/msl5/msl6/msl9/msl10) shows no obvious disruption of development and seems to respond as much as the
wild type to osmotic, salt, mechanical, dehydration and
rehydration stresses [36].
Figure 3. MscS structure and MSL channel activity. (a,b) X-ray crystal structure of
the Escerichia coli MscS channel in open configuration determined to 3.45 Å
resolution [69]. Structure was rendered using Cn3D and the molecular modeling
database at NCBI [70]. (c) Normalized mechanosensitive channel conductances
measured in root cortical protoplasts derived from the knockout mutants msl9-1,
msl10-1 and wild-type plants [36]. Because, under the assay conditions used, MSL9
and MSL10 represent the major mechanically sensitive channel in the root cortex,
the 20 pA conductance in the msl9-1 knockout implies MSL10 supports this 20 pA
channel activity. Conversely, the 8 pA conductance in the msl10-1 background
indicates that it is generated by MSL9. When present together in the wild-type
background, a mechanically sensitive conductance of 10 pA is revealed, implying
MSL9 and MSL10 interact to contribute to form a 10 pA heteromeric channel
complex in vivo.
electrophysiological analysis of protoplasts derived from
the root cortex showed that the major mechanoresponsive
channel activity in wild-type plants was most probably a
Cl conductance [36]. This activity was largely lost in the
msl9/msl10 double mutant background. Residual mechanosensitive channel activity was attributed to MSL4–6,
which are also expressed in roots and show 40% amino
acid identity to MSL9 and MSL10 (which themselves are
75% identical). A quintuple knockout of all these MSL
genes abolished all mechanosensitive channel activity
[36]. MSL9 and MSL10 are thought to form the core of a
multimeric channel, as inferred from the effects of knockout mutants in each gene on mechanosensitive channel
conductance [36] (Figure 3c), consistent with the known
subunit structure of MscS [34].
The MSL gene family provides some very strong candidates for mechanoresponse elements in plants. However,
elucidating whether or not these proteins represent
mechanosensitive channels or accessory proteins that
are required for mechanoresponse will require an analysis
of heterologously expressed protein. At present, we can say
MCA1 – a MID1 homolog from plants?
Another candidate for a component of a plant mechanosensory channel has come from functional complementation of the Saccharomyces cerevisiae MID1 mutant with
plant cDNAs. MID1 mutants lack a component of a yeast
stretch-activated, Ca2+-permeable channel complex. Nakagawa and colleagues obtained an Arabidopsis clone named
MCA1 that partially complemented the mid1 phenotype
[37]. The predicted protein shares only 10% identity and
41% similarity to MID1. It also has no obvious homology to
known channel components and seems unlikely to be acting in the same fashion as the yeast protein. However,
MCA1 was shown to increase Ca2+ uptake in yeast, thereby
indicating that it does have a link to Ca2+ homeostasis and
potentially explaining its partial complementation of the
mid1 mutant.
A knockout mutant of MCA1 showed a reduced ability of
its roots to penetrate a layer of hard agar, suggesting some
defect in either growth or mechanical responsiveness.
However, constitutive MCA1 overexpressing lines exhibited more obvious defects in development, with short
stems, small rosettes, no petals and shrunken seed pods.
These plants also showed an increased basal Ca2+ uptake
and an elevated cytosolic Ca2+ level in response to osmotic
shock that was not evident in wild-type plants. Similarly,
when MCA1 was heterologously expressed in Chinese
hamster ovary cells, a novel Ca2+ increase could be elicited
upon stretching the cells [37].
Thus, MCA1 seems to provide a possible link between
Ca2+ fluxes and mechanical response in Arabidopsis,
although its precise role remains enigmatic. It might be
a regulatory component of a mechanosensitive channel
complex conserved between yeast and plants. The evidence
for a link to mechanoresponse is tantalizing. For example,
the gene TCH3 (CML11) encodes a Ca2+-dependent protein
that has been closely linked to touch response in Arabidopsis [38]. This gene is upregulated in MCA1-overexpressing plants, indicating that touch sensing is constitutively
activated in these lines. However, as a note of caution,
TCH3 expression is also responsive to environmental
stimuli such as darkness and to developmental regulation
and so its altered levels could simply reflect the disruption
of growth that accompanies MCA1 overexpression.
Models for plant mechanosensation II: the role
of the wall
Although the plasma membrane is the primary interface
between the living protoplast and the external environment and transduces many environmental cues into phys231
Review
iological responses, an external mechanical perturbation
will first act on the plant cell wall encasing the protoplast
and cause cell-wall deformation. Because the large hydrostatic pressure of 2–50 atmospheres exerted by the plant
protoplast (turgor) presses the plasma membrane against
the wall, any such deformation will immediately be conveyed to the plasma membrane. The relaxation in the wall
that occurs as plant cells expand will lead to similar wall
stresses and probably activate similar monitoring systems
as external mechanical stimulation. Such cell-wall surveillance systems might monitor wall polymer status directly
or act through the accompanying secondary deformation of
the plasma membrane and associated cytoskeleton. No
clear candidate sensor for monitoring cell-wall deformation
has yet been identified in plants, but mechanical signaling
via the extracellular matrix (ECM) in other kingdoms
again provides insight into some of the general mechanisms likely to underlie such processes.
In the ECM of animal cells, large multimodular proteins
such as the ECM adhesion glycoprotein fibronectin are
thought to unravel stepwise upon exposure to mechanical
forces, revealing previously hidden catalytic domains or
recognition sites. The number and type of newly exposed
sites could thus encode information about the magnitude
and localization of mechanical load acting on the protein
[39,40]. No proteins of similar architecture have been
identified in proteomic analyses of the plant cell wall,
but the intercalated polysaccharide and glycoprotein networks, which form the structural composite of the cell wall,
could potentially serve a similar function by presenting
formerly occupied binding sites to cell surface receptors
upon mechanical disruption.
In animals, integrins have a vital role as surface
adhesion receptors by transmitting mechanical deformation of the ECM to the cytoskeleton and/or cell interior
[41]. The idea that integrin-like proteins also mediate
mechanosensation in plants arose from the observation
that treatment with the integrin-binding tripeptide RGD
altered growth of soybean root suspension cells and inhibited mechanical signaling in internodal cells of the green
alga Chara [42,43]. Seemingly substantiating evidence
was provided by reports that a polyclonal antibody against
the avian b(1) integrin subunit recognized a protein in
Arabidopsis and Chara membranes [44,45], but neither the
Arabidopsis nor rice genomes seem to contain a true
integrin homolog. However, intriguingly, further studies
continued to accumulate support for the involvement of an
RGD-like recognition system in physically connecting the
plasma membrane to the cell wall and in sensing mechanical stress. Thus, during plasmolysis of Arabidopsis
suspension or onion epidermal cells, formation of cytoplasmic Hechtian strands linking the plasma membrane
of the shrinking protoplast to wall attachment sites was
abolished by RGD treatment [46,47]. In addition, regenerating protoplasts normally show oriented division in
response to unidirectional compressive forces [48] but
division became randomized after pre-treatment with
RGD [49]. RGD also interfered with induction of defense
responses elicited by fungal penetration [47] and with
mechanical signaling of Taxus suspension cells exposed
to shear stress [23]. Although the plant RGD-recognition
232
Trends in Cell Biology Vol.19 No.5
system(s) remains unknown, an exciting recent investigation identified eight receptor-like kinases (RLKs) as
potential RGD-binding proteins [50]. One of the RLKs
contained an extracellular lectin-like domain that bound
RGD-containing peptides in vitro [50]. Although the function of this lectin RLK remains to be elucidated, the
cytosolic kinase domain of the protein was predicted to
be active, making the protein an attractive candidate for
signaling changes in cell wall status to the cell interior.
The receptor-like wall-associated kinases (WAKs) and
WAK-like kinases (WAKLs) are similarly positioned to
communicate cell wall perturbations to the cytoplasm
[51]. WAKs consist of a highly conserved cytoplasmic
kinase domain, which is linked to the more variable extracellular domain via a single transmembrane region. The
external domain interacts with glycine-rich proteins of the
cell wall and binds tightly to pectins [52,53]. Intriguingly,
the affinity of the recombinant extracellular subdomain
WAK167–254 for pectins (polygalacturonic acid) was greatly
enhanced by desterification of pectins and the presence of
Ca2+, both of which promote the formation of intermolecular Ca2+ bridges to cross-link pectins in the cell wall [54].
If, as Decreux and Messiaen suggest [54], subtle changes in
pectin architecture result in differential WAK binding, this
would provide a potential mechanism for monitoring cell
wall perturbances. WAKs are upregulated upon exposure
to various biotic and abiotic stresses such as pathogens,
wounding and Al3+ toxicity [53], but have thus far not been
implicated in mechanical stress responses. However,
WAKs are expressed in cells undergoing expansion, and
a reduction in WAK levels leads to a reduction in leaf size
and to shorter roots [51,55]. Because cell expansion
requires a constant remodeling of the cell wall [56], it will
be interesting to elucidate the relationship between
growth-associated (and mechanically triggered) changes
in pectin organization and WAK-dependent signaling.
In S. cerevisiae, the two sensors WISC1 and MID2
monitor cell wall status during vegetative cellular expansion or the formation of a mating projection, respectively.
Structurally, MID2 and WISC1 share many features: their
C-terminal cytoplasmic domains both interact with the
ROM2 guanosine-nucleotide-exchange factor to elicit
Rho1 and protein-kinase-C-dependent signal transduction
events, both possess a single transmembrane domain and
both extend O-glycosylated rod-like extracellular domains
into the periplasmic space. This extracellular domain presumably interacts with the cell wall via an N-terminal
glycan (MID2) or cysteine-rich region (WISC1) [57,58].
Although no homologs of the yeast WISC and MID2
sensors have been found in the Arabidopsis genome, THESEUS1 (THE1) has recently emerged as an exciting candidate for an analogous cell wall integrity sensor in plants
[59,60]. THE1, a RLK expressed in elongating cells and the
vasculature, was identified in a suppressor screen of the
cellulose synthase mutant cesA6prc1–1, which shows short
hypocotyls and accumulates ectopic lignin when grown in
the dark [60]. These phenotypes were partially rescued in
the cesA6prc1–1/the1 double mutant, although, importantly,
the cellulose biosynthesis deficiency of cesA6prc1–1 was not
restored. These results indicate that the reduced cell
expansion of some cellulose-deficient mutants is not simply
Review
caused by an inability to elongate, but is the result of active
inhibition in response to damaged wall structure, triggered
by potential sensors such as THE1 [59]. Interestingly,
several members belonging to the same RLK subfamily
(CrRLK) as THE1 are expressed in cells exhibiting polar
growth such as trichomes and conical cells of Antirrhinum
petal epidermis as well as tip growing cells of Arabidopsis
[59]. Because root hairs and pollen tubes show highly
dynamic expansion, balanced on the brink of cell rupture
(see earlier), the presence of such RLKs is particularly
intriguing.
Continuing with this theme of kinases as possible
mechanosensors, AtHK1, an Arabidopsis plasma-membrane-localized histidine kinase, can complement yeast
lines containing a deletion in their own putative osmosensing histidine kinase SLN1 [61]. In the plant, lesions in
AtHK1 cause an increased sensitivity to osmotic stress
Trends in Cell Biology
Vol.19 No.5
[62], leading to the speculation that this protein might
directly respond to changes in cellular pressure, perhaps
transmitted through tension in the plasma membrane, as
proposed for SLN1 [63]. However, whether AtHK1 could in
fact be responding secondarily to alterations induced by a
mechanosensitive channel, such as an influx of Ca2+,
remains to be defined.
Two major themes arise from the spectrum of putative
plant mechanosensors discussed here. The MSL family
and possibly AtHK1 seem likely to sense mechanical forces
through monitoring tension in the plasma and, possibly, in
other membranes (Figure 4). By contrast, by potentially
linking the ECM to the plasma membrane and cell interior,
RLKs are ideally placed to monitor mechanical strain
while it affects cell wall integrity or the interaction between the cell wall and the plasma membrane. There are
>600 RLKs in the Arabidopsis genome that are thought to
Figure 4. Action of possible mechanoresponsive elements. Plant mechanosensors probably fall into two broad classes: those activated by tension in the membrane, as
exemplified by the MSL family of Cl permeable channels, and those monitoring wall status and/or sheer between the wall and the plasma membrane. SAC represents the
stretch-activated Ca2+-permeable conductance identified through electrophysiology but yet to be cloned [3]. AtHK1 operates in an osmosensing pathway that potentially
functions through a phosphorelay cascade that uses the antagonistic response regulators ARR3/4 and ARR 8/9 [60]. The RLKs such as WAKs [51–53], THE1 [59,60] and the
lectin domain containing RLK [50] provide models for how a wall sensor might operate. They are likely to elicit protein-kinase-dependent signals that could relay mechanical
information directly to mechanoresponsive gene expression for example, or interact with the Ca2+-dependent signaling cascade that, to date, remains the best characterized
mechanically induced signal transduction event in plants.
233
Review
be involved in processes as varied as hormone response,
pathogen recognition and cell differentiation [64]. A subset
of this superfamily could well represent a suite of mechanoresponsive elements. The potential functional redundancy within this large group of proteins might help to
explain why, at present, none of the mutants in the candidate mechanosensing WAKS, lectin-like RLK or THE1
have been directly linked to mechanical sensing. Defining
the direct substrates and the wider signaling networks
they elicit should greatly aid in understanding the degree
to which these RLKs share overlapping function.
Concluding remarks and future perspectives
Despite the key role mechanical forces have in plant
growth and development, molecular candidates such as
the MSL gene family have only recently been identified as
potential mechano-receptors. However, a major question
still remains as to the role of the MSL genes outside of
plastid division. The lack of a root growth phenotype of the
quintuple msl4/msl5/msl6/msl9/msl10 knockout clearly
shows that there are other mechanosensors active in the
plant and their identification remains a pressing goal for
the field. In addition, whether or not the Ca2+ increase seen
upon mechanostimulation represents direct gating of a
mechanosensitive, Ca2+-permeable channel or a downstream event triggered by an alternate primary mechanosensor remains an unanswered question. Although
several important studies have identified RLKs with the
potential of signaling cell wall status to the cytoplasm,
future work will need to determine whether or not these or
related proteins have roles as true mechanosensors.
Because none of these RLKs are predicted to possess ion
channel function, it will also be crucial to elucidate their
relationship to the Ca2+ signaling cascade, which at present is the most firmly established rapid response pathway
triggered by mechanical stimulation.
Acknowledgements
We thank Sarah Swanson for discussion and critical reading of the
manuscript. This work was supported by NSF MCB 0641288.
References
1 Darwin, C. and Darwin, F. (1880) The Power of Movement in Plants.
William Clowes and Sons
2 Braam, J. (2005) In touch: plant responses to mechanical stimuli. New
Phytol. 165, 373–389
3 Monshausen, G.B. et al. (2008) Touch Sensing and Thigmotropism.
Blackwell
4 Green, P. (1997) Expansin and morphology: a role for biophysics.
Trends Plant Sci. 2, 365–366
5 Selker, J.M. et al. (1992) Biophysical mechanisms for morphogenetic
progressions at the shoot apex. Dev. Biol. 153, 29–43
6 Pien, S. et al. (2001) Local expression of expansin induces the entire
process of leaf development and modifies leaf shape. Proc. Natl. Acad.
Sci. U. S. A. 98, 11812–11817
7 Casimiro, I. et al. (2003) Dissecting Arabidopsis lateral root
development. Trends Plant Sci. 8, 165–171
8 De Smet, I. et al. (2007) Auxin-dependent regulation of lateral root
positioning in the basal meristem of Arabidopsis. Development 134,
681–690
9 De Smet, I. et al. (2006) Lateral root initiation or the birth of a new
meristem. Plant Mol. Biol. 60, 871–887
10 Malamy, J.E. (2005) Intrinsic and environmental response pathways
that regulate root system architecture. Plant Cell Environ. 28, 67–77
234
Trends in Cell Biology Vol.19 No.5
11 Malamy, J.E. and Ryan, K.S. (2001) Environmental regulation of
lateral root initiation in Arabidopsis. Plant Physiol. 127, 899–909
12 O’Brien, E.E. et al. (2007) Roots in space: a spatially explicit model for
below-ground competition in plants. Proc. Biol. Sci. 274, 929–934
13 Noll, F. (1900) Uber den bestimmenden Einfluss von
Wurzelkrummengen auf Entstehung und Anordnung der
Seitcnwurzeln. Landtwirtschaftliche Jahrbuchecr 29, 361–426
14 Lucas, M. et al. (2008) Auxin fluxes in the root apex co-regulate
gravitropism and lateral root initiation. J. Exp. Bot. 59, 55–66
15 Muday, G.K. and Rahman, A. (2008) Auxin transport and the
integration of gravitropic growth. In Plant Tropisms (Gilroy, S.
and Masson, P.H., eds), pp. 47–78, Blackwell Publishing
16 Ditengou, F.A. et al. (2008) Mechanical induction of lateral root
initiation in Arabidopsis thaliana. Proc. Natl. Acad. Sci. U. S. A.
105, 18818–18823
17 Laskowski, M. et al. (2008) Root system architecture from coupling cell
shape to auxin transport. PLoS Biol. 6, e307
18 Massa, G.D. and Gilroy, S. (2003) Touch modulates gravity sensing to
regulate the growth of primary roots of Arabidopsis thaliana. Plant J.
33, 435–445
19 Plieth, C. and Trewavas, A.J. (2002) Reorientation of seedlings in the
earth’s gravitational field induces cytosolic calcium transients. Plant
Physiol. 129, 786–796
20 Toyota, M. et al. (2008) Cytoplasmic calcium increases in response to
changes in the gravity vector in hypocotyls and petioles of Arabidopsis
seedlings. Plant Physiol. 146, 505–514
21 Gus-Mayer, S. et al. (1998) Local mechanical stimulation induces
components of the pathogen defense response in parsley. Proc. Natl.
Acad. Sci. U. S. A. 95, 8398–8403
22 Yahraus, T. et al. (1995) Evidence for a mechanically induced oxidative
burst. Plant Physiol. 109, 1259–1266
23 Gao, H. et al. (2007) RGD-dependent mechanotransduction of
suspension cultured Taxus cell in response to shear stress.
Biotechnol. Prog. 23, 673–679
24 Feijo, J.A. et al. (1999) Growing pollen tubes possess a constitutive
alkaline band in the clear zone and a growth-dependent acidic tip. J.
Cell Biol. 144, 483–496
25 Messerli, M.A. et al. (1999) Pulsatile influxes of H+, K+ and Ca2+ lag
growth pulses of Lilium longiflorum pollen tubes. J. Cell Sci. 112,
1497–1509
26 Monshausen, G.B. et al. (2007) Oscillations in extracellular pH and
reactive oxygen species modulate tip growth of Arabidopsis root hairs.
Proc. Natl. Acad. Sci. U. S. A. 104, 20996–21001
27 Messerli, M.A. et al. (2000) Periodic increases in elongation rate
precede increases in cytosolic Ca2+ during pollen tube growth. Dev.
Biol. 222, 84–98
28 Monshausen, G.B. et al. (2008) Imaging of the Yellow Cameleon 3.6
indicator reveals that elevations in cytosolic Ca2+ follow oscillating
increases in growth in root hairs of Arabidopsis. Plant Physiol. 147,
1690–1698
29 Brady, J.D. and Fry, S.C. (1997) Formation of Di-isodityrosine and loss
of isodityrosine in the cell walls of tomato cell-suspension cultures
treated with fungal elicitors or H2O2. Plant Physiol. 115, 87–92
30 Cannon, M.C. et al. (2008) Self-assembly of the plant cell wall requires
an extensin scaffold. Proc. Natl. Acad. Sci. U. S. A. 105, 2226–2231
31 Christensen, A.P. and Corey, D.P. (2007) TRP channels in
mechanosensation: direct or indirect activation? Nat. Rev. Neurosci.
8, 510–521
32 Folgering, J.H. et al. (2008) Molecular basis of the mammalian
pressure-sensitive
ion
channels:
focus
on
vascular
mechanotransduction. Prog. Biophys. Mol. Biol. 97, 180–195
33 Bianchi, L. (2007) Mechanotransduction: touch and feel at the
molecular level as modeled in Caenorhabditis elegans. Mol.
Neurobiol. 36, 254–271
34 Corry, B. and Martinac, B. (2008) Bacterial mechanosensitive
channels: experiment and theory. Biochim. Biophys. Acta 1778,
1859–1870
35 Haswell, E.S. and Meyerowitz, E.M. (2006) MscS-like proteins control
plastid size and shape in Arabidopsis thaliana. Curr. Biol. 16, 1–11
36 Haswell, E.S. et al. (2008) Two MscS homologs provide
mechanosensitive channel activities in the Arabidopsis root. Curr.
Biol. 18, 730–734
Review
37 Nakagawa, Y. et al. (2007) Arabidopsis plasma membrane protein
crucial for Ca2+ influx and touch sensing in roots. Proc. Natl. Acad.
Sci. U. S. A. 104, 3639–3644
38 Braam, J. and Davis, R.W. (1990) Rain-, wind-, and touch-induced
expression of calmodulin and calmodulin-related genes in Arabidopsis.
Cell 60, 357–364
39 Smith, M.L. et al. (2007) Force-induced unfolding of fibronectin in the
extracellular matrix of living cells. PLoS Biol. 5, e268
40 Vogel, V. (2006) Mechanotransduction involving multimodular
proteins: converting force into biochemical signals. Annu. Rev.
Biophys. Biomol. Struct. 35, 459–488
41 Ingber, D.E. (2006) Cellular mechanotransduction: putting all the
pieces together again. FASEB J. 20, 811–827
42 Schindler, M. et al. (1989) RGD-dependent linkage between plant cell
wall and plasma membrane: consequences for growth. J. Cell Biol. 108,
1955–1965
43 Staves, M. and Wayne, R. (1993) The touch-induced action-potential in
Chara – inquiry into the ionic basis and the mechanoreceptor. Aust. J.
Plant Physiol. 20, 471–488
44 Katembe, W.J. et al. (1997) Immunolocalization of integrin-like
proteins in Arabidopsis and Chara. Physiol. Plant. 99, 7–14
45 Swatzell, L.J. et al. (1999) Integrin-like proteins are localized to plasma
membrane fractions, not plastids, in Arabidopsis. Plant Cell Physiol.
40, 173–183
46 Canut, H. et al. (1998) High affinity RGD-binding sites at the plasma
membrane of Arabidopsis thaliana links the cell wall. Plant J. 16, 63–
71
47 Mellersh, D.G. and Heath, M.C. (2001) Plasma membrane-cell wall
adhesion is required for expression of plant defense responses during
fungal penetration. Plant Cell 13, 413–424
48 Lynch, T.M. and Lintilhac, P.M. (1997) Mechanical signals in plant
development: a new method for single cell studies. Dev. Biol. 181, 246–
256
49 Zhou, J. et al. (2007) Responses of Chrysanthemum cells to mechanical
stimulation require intact microtubules and plasma membrane-cell
wall adhesion. J. Plant Growth Regul. 26, 55–68
50 Gouget, A. et al. (2006) Lectin receptor kinases participate in proteinprotein interactions to mediate plasma membrane-cell wall adhesions
in Arabidopsis. Plant Physiol. 140, 81–90
51 Anderson, C.M. et al. (2001) WAKs: cell wall-associated kinases linking
the cytoplasm to the extracellular matrix. Plant Mol. Biol. 47, 197–206
52 Kohorn, B.D. et al. (2006) Wall-associated kinase 1 (WAK1) is
crosslinked in endomembranes, and transport to the cell surface
requires correct cell-wall synthesis. J. Cell Sci. 119, 2282–2290
Trends in Cell Biology
Vol.19 No.5
53 Humphrey, T.V. et al. (2007) Sentinels at the wall: cell wall receptors
and sensors. New Phytol. 176, 7–21
54 Decreux, A. and Messiaen, J. (2005) Wall-associated kinase WAK1
interacts with cell wall pectins in a calcium-induced conformation.
Plant Cell Physiol. 46, 268–278
55 Kohorn, B.D. et al. (2006) An Arabidopsis cell wall-associated kinase
required for inveratse activity and cell growth. Plant J. 46, 307–316
56 Cosgrove, D.J. (2005) Growth of the plant cell wall. Nat. Rev. Mol. Cell
Biol. 6, 850–861
57 Hutzler, F. et al. (2008) Protein N-glycosylation determines
functionality of the Saccharomyces cerevisiae cell wall integrity
sensor Mid2p. Mol. Microbiol. 68, 1438–1449
58 Piao, H.L. et al. (2007) NPFXD-mediated endocytosis is required for
polarity and function of a yeast cell wall stress sensor. Mol. Biol. Cell
18, 57–65
59 Hématy, K. and Höfte, H. (2008) Novel receptor kinases involved in
growth regulation. Curr. Opin. Plant Biol. 11, 321–328
60 Hématy, K. et al. (2007) A receptor-like kinase mediates the response of
Arabidopsis cells to the inhibition of cellulose synthesis. Curr. Biol. 17,
922–931
61 Urao, T. et al. (1999) A transmembrane hybrid-type histidine kinase in
Arabidopsis functions as an osmosensor. Plant Cell 11, 1743–1754
62 Wohlbach, D.J. et al. (2008) Analysis of the Arabidopsis histidine
kinase ATHK1 reveals a connection between vegetative osmotic
stress sensing and seed maturation. Plant Cell 20, 1101–1117
63 Reiser, V. et al. (2003) Yeast osmosensor Sln1 and plant cytokinin
receptor Cre1 respond to changes in turgor pressure. J. Cell Biol. 161,
1035–1040
64 Morris, E.R. and Walker, J.C. (2003) Receptor-like protein kinases: the
keys to response. Curr. Opin. Plant Biol. 6, 339–342
65 Knight, M.R. et al. (1991) Transgenic plant aequorin reports the effects
of touch and cold-shock and elicitors on cytoplasmic calcium. Nature
352, 524–526
66 Knight, M.R. et al. (1992) Wind-induced plant motion immediately
increases cytosolic calcium. Proc. Natl. Acad. Sci. U. S. A. 89, 4967–
4971
67 Pierson, E.S. et al. (1996) Tip-localized calcium entry fluctuates during
pollen tube growth. Dev. Biol. 174, 160–173
68 Takeda, S. et al. (2008) Local positive feedback regulation determines
cell shape in root hair cells. Science 319, 1241–1244
69 Wang, W. et al. (2008) The structure of an open form of an E. coli
mechanosensitive channel at 3.45 A resolution. Science 321, 1179–1183
70 Wang, Y. et al. (2007) MMDB: annotating protein sequences with
Entrez’s 3D-structure database. Nucleic Acids Res. 35, D298–D300
235