Content uploaded by Robert Seagull
Author content
All content in this area was uploaded by Robert Seagull on Aug 24, 2020
Content may be subject to copyright.
Provided for non-commercial research and educational use only.
Not for reproduction, distribution or commercial use.
This article was originally published in the book Reference Module in Food
Sciences, published by Elsevier, and the attached copy is provided by Elsevier for
the author's benefit and for the benefit of the author's institution, for non-commercial
research and educational use including without limitation use in instruction at your
institution, sending it to specific colleagues who you know, and providing a copy to your
institution's administrator.
All other uses, reproduction and distribution, including without limitation
commercial reprints, selling or licensing copies or access, or posting on open
internet sites, your personal or institution’s website or repository, are
prohibited. For exceptions, permission may be sought for such use through
Elsevier’s permissions site at:
http://www.elsevier.com/locate/permissionusematerial
From Seagull, R.W., 2016. The Plant Cytoskeleton. Reference Module in Food Sciences.
Elsevier, pp. 1–22. doi: http://dx.doi.org/10.1016/B978-0-08-100596-5.02873-0
© 2016 Elsevier Inc. All rights reserved.
Academic Press
Author's personal copy
The Plant Cytoskeleton
Robert W Seagull, Hofstra University, Hempstead, NY, USA
Ó2016 Elsevier Inc. All rights reserved.
Introduction 1
Microtubules (MTs) 3
Tubulin Proteins and Genes 3
MAPS 4
Plant Specific MAPS 7
MT Arrays in Plant Cells 7
MT Nucleation from the Nuclear Envelope 7
Formation of Interphase Arrays of MTs 7
Dynamics of CMTs 8
Actin Microfilaments (MFs) 10
Actin Proteins and Genes 11
Actin Binding Proteins (ABPs) 11
ABPs That Control the G-actin Pool (Profilin, ADF/Cofilin, CAP) 11
Regulation of MF Ends (CP Proteins, Villin/Gelsolin) 13
Promotors of Actin Polymerization and Bundling In Vivo (Arp2/3 Complex, Formin, Fimbrin) 13
MF Motor Proteins (Myosin) 13
Actin-Membrane Binding Proteins 13
Microfilament Dynamics 14
Interaction between MTs and MFs 14
The Cytoskeleton during the Cell Cycle 14
Establishment of the Division Plane 14
Mitosis –Spindle Structure and Function 15
Cytokinesis 16
Cytoskeletal Functions 17
Cellulose Microfibril Organization 17
Cell Growth and Shape 17
Diffuse Cell Expansion 17
Tip Growing Cells 18
Trichomes and Pavement Cells 19
Tracheary Elements 20
Conclusion and Future Perspectives 21
References 22
Introduction
Cytoskeletal elements are involved in many essential processes in plant biology, such as cell division, morphogenesis, intracellular
transport, perception and transmission of environmental signals, and plant responses to pathogens. Being sessile organisms, plants
have evolved complex and efficient mechanisms for responding to biotic and abiotic signals from within the plant and from the
plant’s environment. The mechanisms by which the cytoskeleton accomplishes these functions remains of major importance in
our understanding of plant growth, development and interaction with the environment.
The cytoskeleton, which consists mainly of microtubules (MTs) and actin microfilaments (MFs), is involved in many cellular
processes, such as (but not limited to) the determination of cell shape, cell movement, nuclear division, and cytokinesis. In animal
cells, MFs play prominent roles in cell shape determination and cytokinesis, whereas in plants these roles have been taken over by
MTs. Plant cells are surrounded by a relatively rigid cell wall whose physical and chemical characteristics defines their shape. The
manipulation of cell wall characteristics involves MTs that control the direction of cellulose microfibril deposited, and the location
of new walls during cell division. Well-orchestrated and predictable changes in MTs are observed as plant cells progress through the
cell cycle. Cortical MT (CMT) arrays produced during interphase are preplaced by the pre-prophase band of MTs (PPB) during late
G2 phase to prophase. MT arrays in spindles are apparent from prometaphase to anaphase, ultimately giving rise to the MT arrays
associated with the phragmoplast at telophase through cytokinesis. In animal cells, intracellular vesicular transport is a function of
MTs, whereas in plants this transport requires actin-like MFs. More recently, MFs have been shown to be involved in numerous other
processes, originally believed to involve only MTs in plants.
The cytoskeleton in plants has been under intense investigation since the early 1960s. In the “early days”(first 20 years or
so) research was limited to describing the patterns of MTs and MFs in various cell types undergoing specificprocesses
Reference Module in Food Sciences http://dx.doi.org/10.1016/B978-0-08-100596-5.02873-0 1
Reference Module in Food Science, First Edition, 2016, 1–22
Author's personal copy
(mitosis, cytokinesis, elongation, differentiation). These structural studies initially employed electron microscopy due to the
relatively small size of cytoskeletal elements. While providing high resolution images of MTs and MFs and their possible
interactions with other cellular components, these studies were often limited to the examination of only a few cells at
a time (due to the time consuming nature of sample and image preparation). With the development of immunocytochem-
ical techniques, light microscopy became a major tool for observing cytoskeletal components. Using antibodies that specif-
ically bind to the various protein components of MTs and MFs greatly decreased the time required to collect data and
increased the numbers of cells which could be observed. Such observations led to the development of proposed chronolog-
ical sequences for the changes in cytoskeletal components in processes such as the cell cycle, changes in cell expansion, or the
development of the specific cell architecture observed during the morphogenesis of cells, such as guard cells, root hairs,
xylem cells, etc.
Observational studies, using either light or electron microscopy, were often supplemented with pharmacological studies using
various chemical inhibitors of cytoskeletal components. A variety of chemical agents were identified to alter the function or distri-
bution of MTs (such as colchicine, oryzalin, triflualin (MT disruption agents), or taxol (MT stabilizing agent)), or MFs (various cyto-
chalasins or latrunculin-B (MF depolymerizing agents), phalloidin or jasplakinolide (MF stabilizers)). The combination of
observing the effects of such chemical agents on cytoskeletal components and the biological processes under investigation, resulted
in numerous proposals to elucidate the role of MTs, MFs, or both in specific developmental processes.
In the past 3 decades, with the advances in genome sequencing in plants and the exploitation of plant systems such as
Arabidopsis thaliana,significant advances have been made in our understanding of the involvements of MTs and MFs in
various developmental processes and the mechanisms by which the dynamics and function of cytoskeletal elements are regu-
lated. The ability to transform plants by inserting genes which code for various fluorescent proteins, such as a GFP (green
fluorescent protein) has greatly increased our understanding of cytoskeletal dynamics and function. Generating fusion
proteins with GFP attached to various cytoskeletal components (either the proteins with constitute the filaments or proteins
which specifically bind to filament polymers) allows for the visualization of cytoskeletal arrays in either preserved or living
specimens (Figure 1).
Because of the highly conserved nature of cytoskeletal components across eukaryotic systems, great strides have been made in
dissecting the specific associations between the filamentous polymers of the cytoskeleton (MTs and MFs) and the various
microtubule-associate proteins (MAPs) and actin binding proteins (ABPs). Using gene sequence homologies between cytoskeletal
components identified in other systems (mammals, yeasts, etc.) and plant gene sequences, a wide array of associated proteins
have been identified as playing regulatory roles in plant cytoskeletal function. Through this type of analysis it has become evident
that the diversity of cytoskeletal functions observed in plants is in part due to differences in associated proteins (either MAPs
or ABPs)
Since the previous publication of this review in 1992, most of the increases in our understanding of cytoskeletal function in
plants has been made through molecular approaches, supplemented by techniques in microscopy. This review will focus primarily
on the more recent findings regarding the mechanisms controlling the dynamics of the plant cytoskeleton and its role in plant
morphogenesis and as a transducer of information from within, between and from the outside of plant cells.
Figure 1 An example of the use of confocal microscopy and transgenic Arabidopsis plants expressing various fluorescent markers attached to cyto-
skeletal proteins. YFP-TUA5 plants (expressing a yellow fluorescent marker on a-tubulin to visualize MTs) are crossed with RFR-CSI1 plants
(expressing a red fluorescent marker on proteins that interact with cellulose synthase). The resultant plants express both yellow and red fluorescent
markers revealing the location of MTs (YFP-TUA5) and cellulose synthase enzymes (RFP-CSI1), revealing the colocalization of cellulose synthase
complexes along MTs (Merge image). Image analysis reveals the proportion of CSI1 proteins (white dots) which precisely align along MTs. Reprinted
from Li, S., Lei, L., Sommerville, C.R., Gu, Y., 2012. Cellulose synthase interactive proteins 1 (CSI1) links microtubules and cellulose synthase
complexes. PNAS 109, 185–190.
2The Plant Cytoskeleton
Reference Module in Food Science, First Edition, 2016, 1–22
Author's personal copy
Microtubules (MTs)
Tubulin Proteins and Genes
The MT arrays in all eukaryotes bear striking similarities in their general protein composition and the dynamics of their assembly.
The MT polymer is make of heterodimers of a- and b-tubulin. MT polymers consist of linear protofilaments (usually 13) which
associate laterally to form a 24 nm wide hollow cylinder. As a result of the specific orientation of the heterodimers in the polymer,
MTs exhibit an inherent polarity, i.e. MTs exhibit different polymerization/depolymerization rates at the two ends. The end of the
MT that exhibits higher rates of heterodimer addition (and slower rates of heterodimer removal) is designated the “fast-growing”or
“þ”end of the MT, whereas the end that exhibits slower rates of addition (and higher rates of heterodimer removal) is designated as
the “slow-growing”or “”end of the MT.
MTs can undergo rapid cycles of assembly and disassembly (Figure 2). Both a- and b-tubulin bind GTP, which functions to regu-
late polymerization. In particular, the GTP bound to b-tubulin is hydrolyzed to GDP during or shortly after polymerization. As
a result the binding affinity of tubulin for adjacent molecules is weakened, thus favoring depolymerization. MTs undergo treadmil-
ling, a process by which tubulin molecules bound to GDP are continually lost from the minus end and replaced by the addition of
tubulin molecules bound to GTP to the plus end of the same MT. GTP hydrolysis also results in the behavior known as dynamic
instability, in which individual MTs alternate between cycles of growth and shrinkage. Dynamic instability facilitates the formation
of new MT arrays throughout the cell cycle and allows for MT reorganization during different phases of cell expansion (enlarge-
ment) and in response to hormone signals and changes in the environment. Overall MT growth is determined by the relationship
between the addition of GTP bound tubulins and the hydrolysis of GTP to GDP. As long as new GTP-bound tubulin molecules are
added more rapidly than GTP is hydrolyzed, the MT retains a GTP cap at its plus end and MT growth continues. However, if the rate
of tubulin addition (polymerization) slows, the GTP bound to tubulin at the plus end of the MT will be hydrolyzed to GDP. If this
occurs, the GDP-bound tubulin will dissociate, resulting in rapid depolymerization of the MT.
Most organisms, including plants, contain multiple, non-identical tubulin genes. Plant a- and b-tubulins are encoded by two
multigene families, resulting in a discrete number of different tubulin isotypes that is often higher for b-tubulins than for a-tubulins.
The overall number of tubulin genes for both families (a,b) is higher in plants families (which can be as high as nine for b-tubulin
and six for a-tubulin) than that observed in animals, probably related to the sessile nature of plants, resulting in their need for more
complex and dynamic arrays of MTs to adapt to their environment. These genes are differentially expressed during development and
according to tissue type, thus contributing to the observed variations in MT arrays in different tissue types or at different develop-
mental stages. For example, distinct a-tubulin genes are expressed in cells that exit from mitosis and enter G1 where MT arrays
occupy the cortical cytoplasm and function to regulate cell shape. There is a certain amount of redundancy in the tubulin genes,
as indicated by the observation in Arabidopsis that the six different a-tubulin genes code for only 4 different a-tubulin proteins
and the nine b-tubulin genes code for only eight b-tubulin isotypes. Most of the tubulin protein isotypes exhibit a high degree
of amino acid homology (resulting in many of the shared traits among MTs).
Figure 2 Microtubule dynamic instability: Dynamic instability is characterized by the coexistence of polymerizing and depolymerizing MTs. GTP-
tubulin is incorporated at polymerizing (fast-growing or “þ”) MT ends, the bound GTP is hydrolyzed during or soon after polymerization. Thus the
MT lattice is predominantly composed of GDP-tubulin. Polymerizing MTs infrequently transit to the depolymerization phase (catastrophe). Depolymer-
ization is characterized by the very rapid loss of GDP-tubulin subunits and oligomers from the slow growing or “”MT end. Depolymerizing MTs
can also infrequently transit back to the polymerization phase (rescue). From Inoué, S., Salmon, E.D., 1995. Force generation by microtubule
assembly/disassembly in mitosis and related movements. Mol. Biol. Cell 6, 1619–1640.
The Plant Cytoskeleton 3
Reference Module in Food Science, First Edition, 2016, 1–22
Author's personal copy
Genetic, immunocytochemical and biochemical data indicate that cell and tissue development coincides with the expression of
distinct tubulin genes and isotypes. The utilization of specific tubulin isotypes, coupled with post-translational modification of
tubulin may be used to control the binding of microtubule-associated proteins (MAPs) and thus the specific characteristics/
functions of MTs. Although MTs contained in the various arrays found in plant cells are structurally indistinguishable, these MTs
may differ in the types and abundance of aand btubulin isoforms which may impart different MT stabilities or functions.
The complexity of the MT system is further enhanced by reversible post-translational modifications, some of which affect MT
stability. Specific modifications appear to relate to specific binding characteristics between MAPs and MTs. For example, glutamy-
lation and glycylation of tubulin appears to influence MT severing by MT-severing proteins, detyrosination/tyrosination of tubulin
affects the binding of plus-end tracking proteins and motor proteins; acetylation appears to affect MT stability and may be involved
in regulating kinesin-based motility. Phosphorylation of tyrosine or serine/threonine residues in tubulins appears to be important
in MT stabilization and orientation because inhibition of phosphorylation results in disruption and disorganization of MT arrays.
Inhibition of phosphorylation affects MT stability, formation of PPB, progression through mitosis, changes in root morphology. It
has been proposed that phosphorylation can also modulate tubulin–tubulin, tubulin-MAP, MT-MF, or MT-membrane interactions.
Thus, MTs arrays can differ in their protein composition (distinct a- and b-tubulin isotypes, specific post-translational modifi-
cations, different MAPs), resulting in differing properties and functions. Although the various tubulin isotypes are interchangeable,
the differences in isotypes could modify some features of the MT array, such as responses to hormones, environmental changes, MT
dynamics, etc. Differential regulation of tubulin genes during development appears to influence the properties of MTs (dynamics,
interactions with other components), and be an important feature of plant tubulin gene families.
MAPS
Although many proteins can interact with MTs, only a few of them can be termed as microtubule-associated proteins (MAPs). Tradi-
tionally, a protein was considered to be a MAP if it co-purifies in vitro with MTs. The problem with plants is that because they lack
tissues with very high concentrations of MTs (unlike mammalian brain tissue that contains large amounts of MTs and MAPs), it is
difficult to perform the “classic”assembly/disassembly procedures originally used to identify MAPs. A more useful definition of
aMAP could be “a protein that co-localizes with MTs and whose alteration/absence (usually seen through mutations) affects
MT function”. Proteins that interact with and control MT function can be divided into two broad categories: motor proteins
(that probably travel up or down MTs, carrying a variety of cargo) and regulatory proteins. Regulatory proteins can be further cate-
gorized as functioning in MT assembly/disassembly, or MT stability and organization (Table 1). The interaction between MAPs and
MTs is critical at almost every stage of cell development and function.
Movements along MTs in plants involves kinesin, eukaryotic-conserved motor protein that moves along MTs with ATP hydro-
lysis. Each kinesin has a motor domain (which converts the energy of ATP hydrolysis into mechanical work) and another domain
(which defines the specific function of the kinesin, such as cargo binding or regulation). In addition to functioning in organelle trans-
port, kinesin also functions in MT bundling, MT depolymerization, and cross-linking to non-MT structures. Dynein protein, the most
common MT-motor protein in other eukaryotes, appears to be absent from higher plants. This loss of dynein proteins from higher
plants has be proposed to partially explain the very large number of kinesin proteins in plants. Of the 61 kinesins in Arabidopsis, fewer
than 20 have been studied to any degree. Although 14 families of kinesin have been identified in eukaryotes, plants have 10
conserved families (also found in other eukaryotes) and one plant-specific family. Of the conserved families, kinesin 5 and
13 predominate. Kinesin-5 cross-links interdigitating MTs in the spindle and phragmoplast, which is important for spindle
pole separation and phragmoplast organization. The kinesin-5 subfamily contains members that might dimerize allowing the
sliding of anti-parallel MTs. Kinesin-13 depolymerizes MT, an activity that has been widely conserved in eukaryotes; however, kine-
sin-13 localizes to Golgi bodies in plants, a characteristic not observed in other eukaryotes. The kinesin-4 subfamily contains
members localized in the CMT array and are putatively involved in regulating the organization of cellulose microfibrils. The
plant-specific clades, kinesin-14,-7, and -12 are prominent in plant cells. The kinesin-14 family is especially diverse in plants and
six subgroups are recognized as being conserved from moss to flowering plants. Four of these six subgroups have identified specific
functions: KCH (kinesin-14 motor proteins) which is involved in actin–MT interaction; KCA which is involved in chloroplast posi-
tioning and mitosis; KCBP which possesses a calmodulin-binding domain; ATK which is a minus-end directed MT motor protein.
Kinesin-14 is unusual in that it is a minus-end directed motor protein (most kinesins are plus-end directed motors). It has been
proposed that this minus-end directed motor evolved to compensate for the loss of cytoplasmic dynein proteins. In addition,
some kinesin-14 proteins contain a MF-binding region. These kinesins are the only known plant proteins containing both MF-
and MT-binding domains. The Kinesin-7 family in plants has three or four subgroups, such as the NACK subgroup, which participates
in phragmoplast organization. Some kinesin-7 proteins are essential for cytokinesis, being required for MT depolymerization during
phragmoplast expansion. Kinesin-12 has two subgroups: the POK subgroup (important for PPB function) and the PAKRP1 subgroup
(involved in MT interdigitation at the phragmoplast midline).
MAPs that nucleate, stabilize/destabilize, crosslink, or anchor MTs are all required to organize plant MT arrays. A large number of
proteins involved in building, remodeling, and interconnecting MT networks in plants have now been identified. Because of the
highly conserved nature of tubulins, it is reasonable that MAPs in plants must also have shared similarities with MAPs from other
systems. The search for plant MAPs has incorporated immunological methods (antibodies raised against animal MAPs), biochem-
ical approaches (identify plant proteins that bind to MTs from animal systems), and genetic analyses (mutants with defects in MT
organization and/or function). This review focuses mainly on recent findings relating to traditionally defined MAPs, which influence
4The Plant Cytoskeleton
Reference Module in Food Science, First Edition, 2016, 1–22
Author's personal copy
MT growth dynamics and organization. The proteins described in this review most likely represent only a fraction of the proteins
potentially involved in regulating MT dynamics and function. A number of plant MAPs exhibit striking similarities to MAPs found in
other eukaryotes; however, other MAPs appear to be plant specific, perhaps contributing to plant-specific distributions and func-
tions of MTs. There are a number of plant specificMAPs (EDE1,MAP190,RIP/MIDD, etc.) which have been identified as plant
MAPs through their in vivo association with MTs, but for whom, no specific function has been identified.
A major function of MAPs is to orchestrate the production of various MT arrays. In most eukaryotic systems, MTs are nucleated
and organized by an assortment of MAPs that form a structure called microtubule-organizing center (MTOC). A major protein
component found in MTOCs is g-tubulin, another type of tubulin protein which is distinct from the a- and b-subunits of the
MTs. g-tubulin is essential for in vivo MT polymerization. The g-tubulin combines with several other associated proteins to form
the “g-tubulin ring complex”(g-TuRC), containing g-TuSC (gamma tubulin small complex þseveral other GCP (Gamma-tubulin
Complex Proteins) proteins. GCPs exhibit 2 conserved binding motifs, GRIP1 (gamma ring protein motif 1) and GRIP2.g-TuRCs
form a nucleating cap at the –end of the MT and attract dimers to form new MTs. g-TuRCs nucleate MTs at a 25X higher rate that
g-TuSC.
gTuSC is a highly conserved, essential core component in MT assembly. The gTuSC has two copies of g-tubulin and one each of
GCP2 and GCP3. In many eukaryotes, multiple gTuSCs assemble with GCP4,GCP5 and GCP6 into the gTuRC.GCP4,GCP5 and
GCP6 function to bind g-tubulin together in the cap-like scaffold for arranging multiple gTuSCs into a distinctive ring shape of
the gTuRC (Figure 3). GCPs are not directly involved in MT assembly, but rather function to determine where in the cell than
assembly occurs. These proteins define recruitment locations for gTuRCs.
In vivo,g-TuRC localization in the cytoplasm is regulated by gTuRC-specificGCPs. Such proteins function to localize the gTuRCs
to MTOCs, sites on the plasmalemma or along pre-existing MTs, etc. to initiate new MT formation. This complex acts as a template
for a/b-tubulin dimers to begin polymerization; it is located at the slow-growing () end with MT polymerization (addition of
dimers at the þend), resulting in MT extending away from the g-TuRC.
A number of “recruitment”proteins have been identified. Augmin-like proteins have been identified to function as non-
centrosomal gTuRC localization proteins. NEDD1 (Neural Precursor Cell Expressed Developmentally Down-regulated protein 1)
appears to be involved in spindle assembly. TPX2 (Targeting Proteins of Xenopus kinesin-like protein 2) is located on the nuclear
envelop and is involved in nuclear envelope breakdown and “pro-spindle”MT assembly. PP2A (Protein Phosphatase 2A) is
involved in promoting “branch nucleation”of MTs. GIP1 (GCP3-interacting Protein 1) binds the GCP3 proteins of the gTuRC to
existing MTs.
Table 1 MAPs (microtubule-associate proteins) identified in plants
General MAP function MAP name Specific function
Motor proteins Kinesin Organelle transport, sliding of anti-parallel MTs
MT nucleation g-tubulin Part of “g-tubulin ring complex”(g-TuRC) to nucleate MTs
GCPs Part of “g-tubulin ring complex”(g-TuRC) to nucleate MTs
Augmin Localize g-TuRCs to specific cellular sites
NEDD1 Spindle assembly
TPX2 Nuclear envelope disassembly and pro-spindle assembly
PPA2 Promotes branch nucleation
GIP2 Binds g-TuRC to existing MTs
MT dynamics MOR1 (MAP215 family) Promote MT growth in all MT arrays
EB1 (TIPs þfamily) Inhibits MT nucleation
CLASP Increase MT stability
Augmin Promotes MT reorganization/reorientation
Katanin Promotes MT severing
SPR2 Promotes turnover by enhancing both polymerization and
catastrophic depolymerization
Inhibits katanin induced MT severing
TPX2 Slows MT depolymerization
Kinesin Kinesin 13 depolymerizes MTs, kinesin 7 depolymerizes
MTs in phragmoplast
MT organization P161/P90 Promotes separation of newly formed MT from g-TuRC,
stabilizes MTs
WLD3 MT bundling in CMTs
MAP 65 Possible role in MT bundling
Kinesin Kinesin 5 and 12 cross-link interdigitating MTs in spindle
and phragmoplast
Kinesin 14 (KCH protein) cross-links MTs and MFs
TON Regulate nucleation and organization in CMTs and PPB
Unknown function MAP70 Found in all organs
AIR9 Associates with CMTs, spindles, phragmoplasts, CDS
The Plant Cytoskeleton 5
Reference Module in Food Science, First Edition, 2016, 1–22
Author's personal copy
MOR1 is an Arabidopsis thaliana protein, which belongs to the MAP215 family. Animal and plant MAP215 family members
increase MT assembly in vivo and in vitro based on an apparently unique ability to enhance MT dynamics by promoting MT
growth and shrinkage rates as well as catastrophe frequency. In plants, MOR1 localizes to all MT structures, suggesting that
MOR1 is involved in general MT organization. MOR1 controls MT length and regulates MT dynamics throughout the cell cycle.
Mutations in MOR1 also affect distribution of EB1 (MT end-binding protein 1), suggesting an interaction between these
two MAPs.
TON (tonneau) proteins are highly conserved in land plants. TON1 accumulate along CMTs during interphase and along MTs of
the PPB and appears to be involved in regulating MT organization in these arrays. TON2/FASS appears to promote MT nucleation in
CMT arrays. TRM (TON1 Recruiting Motif) proteins regulate the binding of TON proteins to MTs.
þTIPs proteins (MT plus end tracking proteins) are a heterogeneous group of MT binding proteins that specifically accumulate at
MT þends. A variety of proteins from this family have been identified in plants, the largest being EB 1 and CLASP (CLIP (cyto-
plasmic linker associated protein) associated proteins) proteins.
EB (end binding)1 proteins alter MT dynamics by inhibiting MT nucleation. Several different EB1 proteins localizes to MTs in
the phragmoplast, mitotic spindle, and interphase CMT array. In the spindle, EB proteins locate to the spindle poles (the location
of MT minus ends). During interphase, EB proteins locate to the MT plus ends, as well as sites where MTs nucleate. Such sites
often appear mobile in the cortical cytoplasm. The movement of EB proteins represents mobile nucleation sites that are still
attached to slowly depolymerizing MT minus ends. EB1 may act to stabilize MTs, because it is localized preferentially to stabile
MT regions.
CLASP (cytoplasmic linker protein-associating protein) is found in many eukaryotes, including plants. CLASP is involved in cell
division and expansion. CLASP proteins enhance MT stability by promoting the attachment of microtubule plus ends to the cell
cortex and chromosomal kinetochores. CLASP localizes to MTs at both the plus ends and along the length of MTs and is essential
for spindle and phragmoplast organization. In addition, CLASP accumulates at newly formed cell edges (created by the fusion of the
cell plate to the parent cell wall), where it acts to stabilize incoming MT plus ends and suppresses MT catastrophe events, thus allow-
ing MTs to grow and follow the contours of the cell edges.
Augmin specifically binds to the growing þends of MTs and promotes reorientation by regulating MT dynamics. Augmin func-
tions to recruit g–TuRCs giving rise to localized amplification of MT numbers. Also localized to the branch site at the side walls of
existing MTs before the formation of nascent MT branches, augmin might also contribute to MT reorientation. Augmin does not
initiate new MT orientations directly, but rather by the recruitment of g–TuRC it promotes new MT assembly.
TPX2 (Targeting Protein) is a multifunctional MAP the regulated mitotic spindle organization. TPX2 appears to function in
nuclear envelope breakdown and alters MT dynamics by slowing MT depolymerization. TPX2 localizes to the nuclear envelope
(NE) and then relocates to the spindle after NE breakdown.
Formins are a well-characterized family of proteins that promote actin filament assembly. However, formin also has the ability to
bind MTs directly in both animals and plants. Formin seems to participate in MT organization and MT-actin filament interactions in
both animals and plants. For example P161/P90 (members of the formin family) are a pair of MAPs identified as having the ability
to bind MTs to the plasmalemma and promote MT polymerization but not bundling. P90 protein also exhibits phospholipase D
activity, suggesting possible involvement in signal transduction pathways.
Figure 3 Schematic model of the assembly and targeting of g-TuRC. Two g-tubulins, GCP2 and GCP3, combine to form g-TuSC. Whereas the GRIP1
(gamma ring protein) motifs of GCP2 and GCP3 are part of their interaction surfaces, the GRIP2 motifs are involved in g-tubulin binding. GCP4,GCP5,
and GCP6 all possess these GRIP motifs, and are postulated to form g-TuSC-like complexes. Targeting proteins recruit g-TuRC to a particular subcel-
lular location, and activate the complex to initiate MT nucleation. From Hashimoto, T., 2013. A ring for all: g-tubulin-containing complexes in acen-
trosomal plant microtubule arrays. Curr. Opin. Plant Biol. 16, 698–701.
6The Plant Cytoskeleton
Reference Module in Food Science, First Edition, 2016, 1–22
Author's personal copy
Plant Specific MAPS
MAP65 represents a divergent family of proteins involved in MT bundling. There are nine MAP-65 family members in Arabidopsis.
The expression of various MAP-65 proteins is controlled by the cell cycle and hormones. MAP65 proteins appear to associate with all
MT arrays in plant cells. However, MAP65 does not promote MT assembly, but rather, stimulates the formation of MT bundles. The
MAP65 protein family performs a critical role in all MT arrays because distinct members of MAP65 are localized to specific MT arrays
during the cell cycle. During interphase MAP65-1 localizes to CMTs. MAP65-1 moves into the PPB, but is not found in the meta-
phase spindle. MAP65-3 localizes to the spindle midzone while MAP65-4 is preferentially localized at the spindle poles. MAP65-6
and -8 localize to cortical MTs as dots. Many MAP65 proteins have the ability to form antiparallel MT bundles; however, MAP65-1
and -4specifically have the potential to form parallel MTs. At prophase, MAP65-1,-2,-5, and -6localize to PPB while MAP65-4 and
-5 accumulate around the nucleus. At metaphase, MAP65 members diminish in general, but MAP65-4 specifically accumulates at
kinetochore MTs. At anaphase, MAP65-1,-2, and -5 become concentrated in the spindle midzone. In addition, MAP65-4, which
bound kinetochore fibers, becomes concentrated at the sides of the spindle pole. At telophase, MAP65-1,-2, and -5 accumulate
at phragmoplast midzone, while MAP65-4 rapidly disappears. MAP65-3 accumulates in a narrow midline in the phramoplast where
it forms antiparallel bundles at the midline and is essential for phragmoplast midline localizations of kinesins.
MAP70 decorates all MT structures in cells and binds to MTs directly in vitro. The MAP70 family has two subgroups: the MAP70-5
clade found only in angiosperms, and the MAP70-1 clade, found in all land plants thus far examined. MAP70-1 is expressed in all
organs, whereas MAP70-5 is expressed only in xylem differentiation. Molecular functions of MAP70 members remains unclear.
AIR9 (AUXIN INDUCED IN ROOTS 9) protein function remains unclear; however, this protein is found associated with MTs in
cortex, PPB, and phragmoplast. Interestingly, AIR9 remains in the cell wall after the disappearance of phragmoplast MTs. AIR9 has
two separate domains that are responsible for its localization at the Cortical Division Site (CDS) during different stages of cell divi-
sion. The first domain co-localizes with PPBs whereas the second localizes to the CDS and the cell plate during cell plate insertion.
AIR9 does not localize at the CDS during the spindle stage or during phragmoplast expansion, but only later, as the cell plate inserts
into the mother cell wall.
WDL proteins have been identified through their association with MTs after cycles of polymerization/depolymerization. In vitro,
WDL3 has direct MT-binding and bundling activity. In vivo,WLD 1 and 3are localized almost exclusively with CMTs. WLD3 levels
correlate with cell elongation, with increased levels inducing decreased elongation and decreased protein levels increasing cell
elongation.
SPR2 (Spiral two) proteins localize to all MT arrays and bind MTs directly in vitro. Similar to the MAP215 family, SPR2 enhances
MT dynamics by promoting both MT growth rate and catastrophe frequency. SPR2 localizes to both growing and shrinking MT plus
ends and has unique bidirectional movements on the sides of MTs. SPR2 also accumulates at MT-crossing sites especially in pave-
ment cells and appears to inhibit katanin’s MT-severing activity.
CSI (Cellulose Synthase Interactive) proteins have been identified as the protein link responsible for the principal role of cortical
MTs (CMTs) in the guidance of cellulose microfibril deposition. The MAP connects CMTs and cellulose synthase A (CesA)
complexes. CSI1 directly binds to MTs in vitro and localizes in plants to CMTs as a punctate staining pattern (Figure 1). Analyses
of CSI1 mutants suggested that CSI1 influences the velocity of CesA complex movement and maintaining a linear movement of
CesA complexes. CSI proteins appear to be a contributor for linking CMTs to the plasmalemma.
MT Arrays in Plant Cells
Unlike animal cells, where MTs are nucleated in discrete regions of the cell, defined by the MTOC, plant cells produce a wide array of
MT patterns without discrete, easily recognizable MTOCs. MT arrays appear to be regulated by a wide array of MAPs, throughout the
cell cycle (Figure 4).
MT Nucleation from the Nuclear Envelope
In actively cycling cells (prominent in plant meristems), the nuclear envelope can initiate MT assembly. Within these cells, MT poly-
merization associated with the nuclear envelope is transient, being found in: (1) cells entering mitosis (formation of the pre-
prophase band) and the pre-spindle; and (2) post-cytokinetic cells (re-establishment of the interphase MT array).
In post-cytokinetic cells, the nuclear envelope generates endoplasmic MTs (EMTs) that extend through the cytoplasm in all direc-
tions, forming a radial MT array. These radial EMTs enter the cell cortex, to develop into the cortical MT (CMT) array. Formation of
these initial MT arrays helps position the nucleus in the cell. At this point CMT exhibit both a mixed and longitudinally orientation
within the cell.
Formation of Interphase Arrays of MTs
MTs, originally initiated from the reforming nuclear envelope of daughter nuclei (see above), position themselves in the cortical
cytoplasm. In addition to MT initiation by the NE, new MTs are also generated in the cell cortex. As cells expand, CMTs are contin-
uously being assembled and reorganized. Formation of CMT is dependent of the interplay between MT nucleation and MT
dynamics, as a result of localization of MAPs, such as CLASP. These proteins may be delivered to the cell cortex during cytokinesis
and be inserted into the newly formed cell edges created by the fusion of the cell plate to the parent cell wall. If CLASP is present at
these cell edges, MTs emanating from the NE will be stabilized and CMTs will remain in a longitudinal orientation. However, if
The Plant Cytoskeleton 7
Reference Module in Food Science, First Edition, 2016, 1–22
Author's personal copy
CLASP is absent, the MTs will undergo catastrophic depolymerization which favors the establishment of a transverse CMT array.
After nucleation, the majority of the MTs are released from their site of initiation and move in the cell cortex by a “treadmilling”
mechanism (net addition of tubulin subunits to the plus-end and the simultaneous removal of subunits at the minus-end). As
MTs enter the cell cortex, they anchor to the plasmalemma. At this point (depending of the cell) MTs elongate and/or become incor-
porated into bundles. The organization of CMT arrays occurs after the separation of MTs from the gTuRCs on the nuclear envelope.
The dynamic nature of CMTs results in an ability to self-organize into specific arrays.
Dynamics of CMTs
Once established in the cortical cytoplasm, MT assembly continues from multiple cortical sites. gTuRCs are localized along the
length of existing CMTs, indicating that MT nucleation is probably dispersed throughout the cell cortex, rather than being localized
to specific sites. New MTs nucleate from the sites of existing CMTs. Localization of gTuRCs along MTs or throughout the cytoplasm is
dependent on a gTuRC-associate protein GCP-WD (aka NEDD1). The orientation of newly formed MTs, relative to existing MTs
may be dependent on the specific protein composition of the gTuRCs. New MTs emanate from existing MTs at various angles
(see below) and that orientation appears to be influenced by the specificGCP proteins in the gTuRCs.
New CMTs are formed by several types of nucleation, based on the angles at which the new MTs emanate from existing CMTs
(Figure 5). Branch-form nucleation occurs when a new CMT is initiated at an acute angle along a pre-existing CMT. Measurements of
the angle between the newly formed CMT and the preexisting CMT show a fairly wide distribution (20–60) around an angle of
40. Branch-form nucleation occurs equally on all sides of the CMTs and exhibits a preference for branching and MT elongation
towards the plus end of the “mother”CMT, suggesting that binding of the gTuRC with an existing CMT has a directional bias,
possibly arising from the inherent polarity of the CMT. Branch-form nucleation promotes formation of a uniformly dense CMT
array by creating new CMTs at divergent angles.
A second type of MT nucleation, parallel-form nucleation, occurs half as frequently as branch-form nucleation. As in branch-form
nucleation, parallel-form nucleation also has a preference for growth in the same direction as the pre-existing CMT. Parallel-form
nucleation effectively reinforces bundle formation within the CMT array, affecting its formation and reorganization. Parallel-form
nucleation contributes to alignment of CMTs, generating parallel arrays of CMTs.
The least frequent (1–2% of nucleation events) type of CMT nucleation free or de novo nucleation. New CMTs initiate without
existing CMTs. It is unknown whether a distinctive mechanism for activating the g–tubulin nucleation complex is required for
free CMT nucleation. Given the rarity of free nucleation, perhaps free nucleation represents the chance activation of cortical g–TuRCs
rather than a specific activating factor.
Nucleation activity not only functions to replace CMTs that are lost due to depolymerization, but also serves to regulate the
degree of CMT alignment and overall array orientation. The nucleation of MTs from g–TuRCs controls the location, timing and
geometry of CMT nucleation, thus affecting the generation of particular CMT arrays and the subsequent remodeling of these arrays.
Variation in the proteins involved in nucleation (g–TuRSs þGCPs) may contribute to the rate, efficiency and geometry of nucle-
ation, whereas, the location of the g–TuRCs may influence the overall array orientation.
The orientation of CMTs plays a major role regulating subsequent cell expansion elongation. MTs must not only be organized on
each face of the cell (sometimes different faces exhibit different MT arrangements), but these arrays must remain continuous in the
cell with MTs bending into and out of cell corners. A mechanism of self-organization in CMT arrays has been proposed, wherein
cell-wide organization emerges from local interactions between individual CMTs. The MT–MT interactions at each cell surface that
Figure 4 The localization of MAPs to distinct MT arrays throughout the cell cycle. MAPs show differential localization to MT arrays throughout the
cell cycle, suggesting their diverse functions in cell division. Some MAPs (Katanin,EB1,CLASP, and MOR1) are found in all MT arrays, while MAPs
(MAP65,TON1) are found only in some arrays. From Struck, S., Dhonukshe, P., 2014. MAPs: cellular navigators for microtubule array orientations in
Arabidopsis. Plant Cell Rep. 33, 1–21.
8The Plant Cytoskeleton
Reference Module in Food Science, First Edition, 2016, 1–22
Author's personal copy
influence CMT self-organization include angle-dependent bundle formation, as well as nucleation of new MTs from pre-existing
ones (Figure 5). An apparent outcome of MT self-organization is the emergence of large groups of parallel MTs within the CMT
array that grow with a unidirectional bias (most, if not all þends facing the same direction) and often change position and orien-
tation over time.
Once formed (regardless of the type of formation), new CMTs do not remain associated with gTuRCs (Figure 6). Katanin
proteins interact with gTuRCs, to sever MTs from their nucleating sites. Removal of the MT from the gTuRC increases MT treadmil-
ling and the depolymerization of MTs from the free () end. Stabilization of the growing end of MTs involves binding to specific
proteins, such as MOR1, and/or by a possible association with the plasma membrane through plus-end proteins like CLASP,P161 or
P90 (MT stabilizing proteins). Treadmilling of CMTs brings new MTs in contact with other CMTs. Due to “branch form”MT nucle-
ation (see above), new MTs form cross-over points with existing MTs. These crossover points generate severing of new MTs (via kata-
nin) creating new polymerizing ends and thus more MTs in the new orientation. Severing by katanin may be regulated through
post-translational modifications (glutamylation and glycylation) of tubulins. As CMTs move using treadmilling, they can interact
with other MTs via MAP65 proteins that facilitate MT bundling. The combination of new CMTs oriented at an angle to “older”CMTs
and the depolymerization of the older CMTs results in the generation of parallel arrays of MTs with a new (transverse to the old
MTs) orientation.
Plus-end binding proteins (þTIPs), such as CLASP,EB1 or Augmin, might be involved in the regulation of MT reorientation and
association with membranes, both the endomembrane system and the plasmalemma. CLASP, associated with the plasmalemma,
accumulates at specific cell edges to overcome sharp edge-induced MT depolymerization, thus stabilizing membrane associate
CMTs as they bend into and out of cell corners. CLASP may also contribute to MT ordering by modulating the location and degree
of MT catastrophe (depolymerization). MT polymerization is increased by EB1.Augmin stabilizes CMTs by binding to the plus ends of
growing MTs, however Augmin does not affect MT growth rates. Thus multiple plus-end binding proteins might be required for MT
reorientation, with partially overlapping functions in the regulation of the dynamics of MT plus ends. Furthermore, SPR1, another MT
plus end binding protein, is expressed in rapidly elongating tissues, and leads to MT reorientation and enhanced MT stability. The
regulatory mechanisms of plus-end binding proteins in the control of MT organization and orientation are complex, with different
MT plus-end binding proteins functioning differently but coordinately in the regulation of MT reorientation.
Association of þTIPs with MTs is potentially regulated by the detyrosination/tyrosination of tubulin. Although detyrosinated
MTs are less dynamic than those containing tyrosinated tubulin, the detyrosinated form of tubulin interacts specifically with kine-
sin-1, while tyrosinated MTs interact preferentially with the plus-end protein CLIP170. The tyrosination/detyrosination of plus-end
tubulin can regulate the binding to MAPs and, consequently, MT dynamics.
Association of MTs with motor MAPs is affected by other post-translational modifications. For example, acetylated tubulins
increase the interaction between MTs and the motor protein kinesin. Post-translational modifications of tubulin may be involved
Figure 5 Types of CMT nucleation. (a) In branch-form nucleation, a new CMT initiates from the surface of a mother CMT at an acute angle (/). In
parallel-form nucleation, a new CMT initiates from the surface of a mother CMT in a co-aligned manner. In free nucleation, a new CMT initiates at the
cell cortex independently of pre-existing CMTs. Black circles represent c–tubulin-containing nucleation complexes. In the case of free nucleation, it is
not known whether the nucleation complex is bound to the plasma membrane. (b) Hypothetical mechanisms for branch-form and parallel-form
nucleation. In these diagrams, the g–tubulin-containing nucleation complex is shown as a purple cone, the mother CMT is colored blue, and the
newly initiated CMT is colored green. During branch-form nucleation, it is envisioned that the g–tubulin-containing nucleation complex is bound to
the surface of the mother CMT relatively weakly. A weakly tethered nucleation complex may pivot within an angular range that centers around 40.
During parallel-form nucleation, the g–tubulin-containing nucleation complex is envisioned to be more tightly bound to the mother CMT surface,
possibly due to recruiting factors that extensively bind to the mother CMT or additional proteins that prevent pivoting of the nucleation complex.
From Fishel, E.A., Dixit, R., 2013. Role of nucleation in cortical microtubule array organization: variations on a theme. Plant J. 75, 270–277.
The Plant Cytoskeleton 9
Reference Module in Food Science, First Edition, 2016, 1–22
Author's personal copy
in regulating the transition from interphase to mitotic MTs. Acetylated tubulin and kinesin-1 are localized at the poles of the plant
mitotic spindle. Acetylation of tubulin could mark specific subsets of MTs involved in the assembly of the mitotic spindle.
Interactions between MTs during the self-organization process is required to regulate the patterning of large-scale MT arrays.
These self-organizational mechanisms include: 1) bundling, when MTs interact at small contact angles, 2) MT catastrophe when
MTs interact at steep angles, 3) MT severing at MT cross over points, 4) MT branch-and-release from pre-existing MTs, and 5)
MT-detachment-based bundle formation.
Most newly divided cells contain sharp cell edges at the junction between new and old cell walls. These edges present a physical
barrier that impedes assembly of MTs growing into them. Such physical barriers impede MT growth and increase MT catastrophe
events. MTs growing towards a particular edge have a probability of edge-induced catastrophe that depends on the edge’s curvature
and MT-edge encounter angle. To facilitate the continuity of CMT arrays on different faces of the cell, during the early G1 phase,
g-tubulin nucleation complexes (GCP2, and GCP3) are temporally localized at the edge along the newly formed end wall. CMTs
are nucleated from the cell edge to form a longitudinal array, which is then replaced with a transverse array (as described above).
Edge-induced CMT depolymerization is also prevented due to the localization of CLASP into the cell edges. This local regulation of
CMT behavior may ensure the global and flexible organization of CMTs in cells with a polygonal geometry.
The intrinsic geometry of cells may provide an external organizing cue that can bias the orientation of CMTs. In multicellular
tissues, the distribution of physical forces between cells represents a potential source of information for the cells “to know”their
relative positon in the tissue. Within tissues, the maximal direction of physical stress imposed by cell wall strain may regulate
the orientation of CMTs. CMT patterns can be influenced by auxins, which in turn results in anisotropic cell growth. Such growth
and subsequent cell wall deposition contributes to the physical stresses on cells which form a feed-back loop to reinforce CMT
patterns and anisotropic growth.
Actin Microfilaments (MFs)
Plant cells exhibit a highly dynamic cytoplasmic architecture with the location and organization of various subcellular components
exhibiting a continuous reorganization. This dynamic nature depends on the continuous remodelling of cytoskeletal elements
Figure 6 Model of MAP localization in relation to CMT assembly. (a) Image of two polymerizing MTs that have grown out of nucleating centers
(purple ovals). The lower MT is depicted as being severed from the nucleating center by a p60/p80 katanin hexameric complex. EB1 binds to both the
nucleating centers and the growing MT plus ends, where it presumably acts as a stabilizing factor and also recruits proteins. g-tubulin acts to seed
MT polymerization in the nucleating center. One or more MAP-65 homologues form a dimer to crosslink MTs, whereas MOR1 stabilizes MTs. SPR1
preferentially localizes to MT plus ends (it is not known whether SPR1 binds MTs directly or through an intermediate). (b) An MT that is depolymeriz-
ing from both ends (faster from the plus end). Note the absence of concentrated EB1 and SPR1 protein at the depolymerizing plus end. Katanin is
shown to localize along the tubule (hypothetical). Also hypothetical is a kinesin interacting with MOR1 to assist in MT depolymerization (a process
shown to occur with MOR1 homologues in vertebrates). From Sedbrook, J.C., 2004. MAPs in plant cells: delineating microtubule growth dynamics
and organization. Curr. Opin. Plant Biol. 7, 632–640
10 The Plant Cytoskeleton
Reference Module in Food Science, First Edition, 2016, 1–22
Author's personal copy
(particularly actin MFs) in response to developmental and environmental signals and requirements. The arrangement and proper-
ties of actin MFs are known to be regulated by associated proteins that bind either to monomeric or polymerized cytoskeletal
proteins. Plant cells use MF arrays to rearrange cytoplasmic components and rely on myosin proteins for most of the cargo traf-
ficking, including that of carrier vesicles and other endomembranes.
Actin Proteins and Genes
Plant actin, like actin from other systems, is a globular, nucleotide-binding protein containing 376–377 amino acid residues.
Plant actin shares most of these residues with actin in other kingdoms (83–88% identity with actins from green algae, most other
protists, fungi, and animals). G-actin (globular-actin) monomers, containing ATP, polymerize into F-actin (filamentous) poly-
mers, forming 7–9nmdiameterfilaments. In vitro, the initial formation of the polymer (assembly of dimer and trimers) starts
slowly, but is followed by a rapid filament elongation phase with the assembly of ATP-monomeric actin onto filament ends
(Figure 7). The orientation of the actin monomers and the slow dephosphorylation of ATP in the polymer results in an inherent
polarity in the filament. New ATP-bound monomers preferentially add at the growing, “þ”end, or barbed end whereas mono-
mers containing ADP preferentially are removed from the slow growing, “-”end, or pointed end of the filament (the terminology
of barbed and pointed end of an actin filament comes from the “arrowhead”binding pattern of myosin heavy-chain molecules to
the actin filament). In vivo, most of the actin is found in the monomeric form, with only as small percentage (5–10%) of the actin
in the filamentous form.
Most plant systems studied possess numerous actin genes and actin protein isotypes. For examples, Arabidopsis contains 10 actin
genes, at least eight of which are strongly expressed at some point during plant development. Unlike actins found in vertebrate
organisms, in Arabidopsis, many of the actin isotypes exhibit a high degree of non-conservative amino acid substitutions. As a result,
many plant species exhibit five to six distinct actin isotypes.
Actin gene expression and protein composition (protein isotypes) can be divided into vegetative and reproductive classes.
Within the vegetative sub-classes, subclass 1 (ACT2 and ACT8) is found in most vegetative tissues. On the other hand, subclass
2(ACT7) is expressed in young, rapidly growing vegetative tissues and is responsive to most phytohormones. Of the three repro-
ductive subclasses, subclass 3 (ACT11) is expressed in gynoecia and pollen, subclass 4 (ACT1 and ACT3) localizes to young ovules,
pollen, and organ primordia, and subclass 5 (ACT4 and ACT12) is expressed during pollen development. Most tissues and organs
simultaneously express two or more actin subclasses. For example, all eight of the actin genes appear to be expressed into 8 isotype
proteins in developing vascular tissues. Multiple actin isotypes are expressed in individual cells. The interaction of various isotypes
in MF polymerization contributes to the dynamic behavior of MFs in cells. In addition, different isotypes contributes to variations in
the binding of numerous actin binding proteins (ABPs).
In plant cells, actin dynamics are controlled by more than 70 classes of ABPs. These proteins have a variety of functions,
including: 1) binding to monomers to regulate the size and activity of the G-actin pool, 2) binding to the side of actin filaments
and alter disassembly properties (promote or inhibit depolymerization), 3) severing filaments and 4) inhibiting assembly or disas-
sembly. Plant ABPs exhibit strikingly similar amino acid sequences to mammalian or yeast ABPs, however the specific activities of
ABPs between plants and mammalian (or yeast) vary significantly.
Actin Binding Proteins (ABPs)
ABPs That Control the G-actin Pool (Profilin, ADF/Cofilin, CAP)
Although most of what is known about profilin comes from the study of non-plant systems, most plants studied show a multigene
family (5–10 members) of profilin that is expressed in a tissue-specific and developmentally regulated manner. Profilin is a small
protein that binds to ATP-bound actin monomers with high affinity, preventing spontaneous filament formation and suppressing
the addition of monomers to the “þ”end of existing filaments. Profilin can also inhibit nucleation and prevent growth of actin fila-
ments at minus ends. Unlike animal profilin, plant profilin does not promote the nucleotide exchange in actin, thus plant profilin
appears to function in maintaining the pool of unpolymerized actin monomer, rather than promoting filament assembly. Isoforms
of profilin have been reported to regulate cellular growth and morphogenesis by organizing MF dynamics in higher plants. Profilin–
actin complexes can act with formin to enhance the rate of MF assembly at plus ends, while increasing the cellular profilin concen-
tration leads to a decrease in filamentous actin levels. Profilin also binds proline-rich proteins, such as formin, to promote MF
assembly and bundle formation.
Proteins in the ADF (Actin Depolymerizing Factor)/cofilin family work with profilin to: 1) increase the dissociation rate of actin
subunits from the “slow growing”end of MFs (end containing actin monomers with ADP attached); 2) increase the rate of conver-
sion of ATP to ADP in actin monomers in the filaments; and 3) create new filament ends by severing MFs. ADF/cofilin proteins pref-
erentially bind to regions of the MF containing ADP, as opposed to ATP, thus promoting the disassembly of older regions of MF
networks. In plants, the concentration of these proteins is much higher than found in animal systems, with molar ratios to actin
reported as 1:1–1:3. Such high levels are predicted to increase severing activity, resulting in the creation of new MF ends. If new
ends remain uncapped then a net increase in polymerization into MFs results. However, if ends are capped by other ABPs, such
as villin, increased MF formation does not occur. Thus ADF can either drive MF turnover or induce total depolymerization.
The ability of ADF/cofilin to depolymerize MFs or to foster MF polymerization depends on the activity of other ABPs. One such
factor is actin interacting protein AIP1, a protein which enhances the activities of ADF/cofilin. In the absence of ADF/cofilin,AIP1
The Plant Cytoskeleton 11
Reference Module in Food Science, First Edition, 2016, 1–22
Author's personal copy
Figure 7 A model for actin MF dynamics. This figure displays the major features and key molecules that regulate the assembly, disassembly and
organization of actin filament arrays in plant cells. The gray box displays the key steps of filament assembly and turnover. Red squiggly arrows
denote filament severing events, which are a major feature dynamic MF turnover. From Henty-Ridilla, J.L., Li, J., Blanchoin, L., Staiger, C.J., 2013.
Actin dynamics in the cortical array of plant cells. Curr. Opin. Plant Biol. 16, 678–687.
12 The Plant Cytoskeleton
Reference Module in Food Science, First Edition, 2016, 1–22
Author's personal copy
shows rather little interaction with MFs; whereas in the presence of ADF/cofilin,AIP1 caps filament “fast-growing”ends and binds
weakly along the sides of filaments. AIP1 can also enhance severing and depolymerization activity of ADF/cofilin.
CAP (cyclase-associated protein) proteins also regulate MFs in plant cells. CAPs are conserved actin-monomer-binding proteins
that sequester actin monomers, thus playing a role in MF dynamics. CAP enhances the exchange of ATP for ADP in actin monomers,
and also promotes the severing of actin filaments in cooperation with ADF/cofilin. Plant CAP probably acts as a major intermediary
in actin filament assembly/turnover.
Regulation of MF Ends (CP Proteins, Villin/Gelsolin)
The availability of MF ends for subunit loss and addition is a fundamental way to regulate actin dynamics. As previous mentioned
AIP proteins bind to the fast-growing ends of MFs in an ADF/profilin dependent manner. CP (capping proteins) in plants bind to the
fast-growing ends of MFs much less efficiently than their mammalian counterparts. However, once attached, CP proteins remain
associated with the MF ends, thereby preventing the addition of monomers. CP proteins thus regulate the abundance of filamentous
actin in plant cells. The activity of CP is regulated though its interaction with phospholipids, such as Phosphatidic Acid (PA). Inac-
tivation of CP by PA results in an increase in filamentous actin, perhaps though the uncapping of filament ends. Inactivation of CP
by PA appears to be prevalent in association with organelles or the plasmalemma and may contribute of intracellular motility.
Gelsolins and villins are similar proteins. Gelsolins contain six conserved gelsolin-homology domains, whereas villins comprise a gel-
solin core and an additional carboxy-terminal actin-binding domain. Gelsolins are actin-filament severing and capping proteins.
Villins bind and crosslink or bundle adjacent MFs, a property that is lacking in gelsolins. Plant villins are Ca
2þ
-calmodulin (CaM)-
regulated, capable of Ca
2þ
-stimulated severing and capping activities. Gelsolin also appears to sever or cap actin filaments in
aCa
2þ
-regulated fashion. All villin/gelsolin family members do not necessarily exhibit the same actin-regulatory activities.
Promotors of Actin Polymerization and Bundling In Vivo (Arp2/3 Complex, Formin, Fimbrin)
In plants, genes for two actin nucleation proteins have been identified: the Arp2/3 complex and formin.InArabidopsis, homologs genes
have been identified for all the subunits of a putative Arp2/3 complex, and a large family of formin genes. Although the proteins have
not yet been localized within cells, mutations in the Arp2/3 genes result in defects in actin organization and the inhibition of cell
expansion.
All of the putative subunits of the Arp2/3 complex, including Arp2,Arp3, and ARPC1 to ARPC5, exist and are ubiquitously
expressed in plants. Arp2 is essential for membrane association and Arp3 is localized to actin nucleation sites in vivo.ARPC4 controls
the assembly and stability of Arp2/3 complex. Loss-of-function mutations of the Arp2/3 complex result in altered MF distribution
(increased MF bundles) and misdirected cell expansion in trichomes, pavement cells, and hypocotyl epidermis.
Formin, an actin-nucleating protein, binds to the þends of actin filaments, promoting de novo actin nucleation. Flowering plants
have two classes of formins, typically having at least 10 genes in each family. Arabidopsis formins comprise more than 20 isoforms that
are separated into two subfamilies depending on the presence of a transmembrane domain. Plant formins contain specific binding
domains for interacting with other ABPs(FH1 binds profilin) and actin monomers (FH2 binds actin). Formin can act as a nucleator,
capable of using plant profilin-actin complexes to generate new MFs. Formin is regulated through the binding of membrane bound
phospholipids. Only formin bound to phospholipids appear to initiate MFs. The location of these phospholipids in the cell helps
regulate where new MFs will form. Once MFs are initiated, formin migrates from the fast growing end to along the length of the MF.
By generating new actin polymers, formin is involved in the bundling of MFs. Mutations in formin genes result in defects in cytoki-
nesis and cell polarity.
In plants, fimbrin cross-links MFs, resulting in tightly packed MF bundles. Fimbrin proteins are highly conserved and generally
contain two actin-binding domains. Plant fimbrin stabilizes MF arrays and prevents profilin-induced depolymerization of MFs.
MF Motor Proteins (Myosin)
Higher plants possess only two classes of myosin proteins. Myosin VIII is concentrated at newly formed short end walls and myosin XI
which is implicated in the movement of organelles. Myosin VIII is involved in connecting neighboring cells through plasmodesmata.
It is possible that myosin VIII may function as a membrane anchor to MF cables in adjacent cells, resulting in a continuity of move-
ment between adjacent cells. Myosin XI appears to be involved in the rapid movement of plant organelles. Members of the myosin XI
sub-family have been implicated in the movement of Golgi and mitochondria. Loss of multiple myosin XI members disrupts the
translocation Golgi stacks and affects the organization of the endoplasmic reticulum. Binding myosin to various membrane-
bound compartments (cargos) is accomplished by MyoB (myosin binding) proteins. Evidence indicates that not all moving
compartments are directly attached to myosin (and thus MF cables). While some transport vesicles are actively transported, other
move passively, being swept along in the flow of the actively transported vesicle.
Actin-Membrane Binding Proteins
A plant specific Networked (NET) superfamily of actin-binding proteins function to localize the actin cytoskeleton to different
membrane compartments. Members of this family function to bind actin filaments to plasmodesmata, the tonoplast membrane,
nuclear envelope or plasmalemma. These proteins appear to be found only in tracheophytes (plants with an evolved vascular
system) and are absent from mosses. In Arabidopsis, 13 potential protein isotypes have been identified. Although exhibiting a variety
of sizes, all have an actin binding region and a variable region, thought to determine specific membrane interactions. These proteins
may function to link, via MFs, various membrane components of the cell, such as plasmalemma to endoplasmic reticulum.
The Plant Cytoskeleton 13
Reference Module in Food Science, First Edition, 2016, 1–22
Author's personal copy
Fine arrays of actin filaments (not the large actin cables involved in cytoplasmic streaming) have been identified in association
with the plasmalemma. Fine filament arrays remain associated with membrane fragments (membrane ghosts). Although MFs are
often visualized in association with MTs, their interaction with the plasmalemma is not dependent on MTs since loss of MTs does
not alter MF patterns associated with the plasmalemma. Most of the known ABPs (profilin,fimbrin,ADF,myosin) have not been thor-
oughly tested for their possible role in MF binding to the plasmalemma. Formin proteins form connections between the plasma-
lemma and actin filaments.
Microfilament Dynamics
Reorganization of MF arrays occurs via two distinct mechanisms. The first mechanism involves MF severing, followed by new poly-
merization at the severed ends. This mechanism results in a reorganization of MFs in the cortical cytoplasm. In vivo, actin-bundling
proteins stabilize MFs and prevent their disassembly. Because MFs do not exhibit treadmilling in vivo, severing may be the primary
mechanism for organizing the MF network. Severing of MFs can occur at multiple locations in the cell, generating numerous smaller
fragments that move around the cell cortex by cytoplasmic streaming or diffusion. In this manner, severing is spread throughout the
cells, resulting in the disassembly of the entire MF network. Fragments, resulting from severing undergo rapid depolymerization to
G-actin, which can then be recycled for the polymerization of new MFs. The second mechanism does not require formation of new
filaments and involves filament bundling, unbundling, and MF sliding (via myosin), resulting in a continual dynamic rearrange-
ment. This sliding is typical of thick actin cables within cytoplasmic strands, as well as in the cortical cytoplasm.
Interaction between MTs and MFs
Interaction between MT and MF arrays has been implied by numerous observations of co-alignment between MTs and MFs and
pharmacological experiments where changes in one cytoskeletal component result in the alteration of the other component. In
elongating cells, single or small bundles of cortical MFs lie parallel to the transversely oriented CMTs. Disruption of these MTs
results in the loss of MF order, producing more randomly organized cortical MFs, whereas stabilization of MTs has no significant
effect on the dynamics of MF arrays. However, depolymerization of MTs induces an increase in both the actin polymerization and
depolymerization rates, indicating that MF dynamics is modulated by MTs.
Integration of MF and MT interactions through protein connections provides an effective tool for enabling local “cross-talk”
between the two cytoskeletal systems. A plant specific region of a membrane-integrated formin mediates an association with the
MT and MF cytoskeleton. Formin accumulate at the endoplasmic reticulum membrane and co-aligns the endoplasmic reticulum
with MTs. Formin can nucleate and anchor actin filaments. Formin 14 and Kinesin-14 proteins are found in the PPB and directly
bind and bundle MFs and MTs. MAP190 is a novel plant MAP which exhibits affinity for both MTs and MFs. It is localized to
the nucleus during interphase, and the mitotic spindle during mitosis. Kinesin-13 binds both MTs and MFs in developing cotton
fibers and may coordinate the function and organization of these two cytoskeletal elements.
The Cytoskeleton during the Cell Cycle
For the first couple of decades of cytoskeletal study certain processes (mitosis and cytokinesis) were considered MT functions,
whereas other processes (cytoplasmic streaming) were considered MF functions. With expansion of technologies used to examine
cytoskeletal function (mutation screening, genetic analysis, plant transformation, etc.) it has become ever more clear that in many
instances, MTs and MFs work together to accomplish various processes. In addition, numerous MAPs (Figure 4) are involved in the
dynamics of the cytoskeleton during the cell cycle. Pharmacological experiments with chemical agents that disrupt the assembly/
disassembly of either MTs or MFs, reinforce the belief that these cytoskeletal elements play pivotal roles in the progression through
the cell cycle. More recent data using plants with mutations in specific cytoskeletal components or transformed plants that are
capable of revealing cytoskeletal components in the living system demonstrate the complexity of interactions both within and
between MTs and MFs.
As cells progress through the cell cycle MTs and MFs exhibit dynamic reorganizations. In interphase, MTs are localized in the
cortical cytoplasm. The organization of these MTs varies depending on cell location. In plant meristems, if cells are programed
for continuous division, producing cell files, interphase MTs are organized into generally parallel arrays oriented perpendicular
to the direction of cell elongation (more details in subsequent sections). As cells enter into G2, the interphase array is replaced
by a very dense array of highly organized MTs, called the pre-prophase band (PPB). This band of MTs is short-lived and disappears
in late prophase, as the nuclear envelope begins to disassemble. The PPB predicts the location where the new cell plate will fuse with
the parent cell wall.
Establishment of the Division Plane
In the vast majority of plant cells, when cytokinesis occurs, the site for the formation and fusion of the new cell wall with the parent
wall is predicted/determined by the formation of the PPB. Originally believed to contain only MTs, we now know that both MTs and
MFs play roles in the formation and function of the PPB. The PPB is a three-dimensional, highly aligned MT bundle, which forms
14 The Plant Cytoskeleton
Reference Module in Food Science, First Edition, 2016, 1–22
Author's personal copy
just inside the plasmalemma during late G2 phase to prophase, prior to the formation of the mitotic spindle and phragmoplast. The
PPB disassembles rapidly during the late prophase and prometaphase. The locus defined by PPB is called the cortical division site
(CDS). Following PPB disassembly, the CDS is marked throughout the remainder of the cell cycle by local depletion of cortical
F-actin and kinesin. In addition, the proteins TAN (an MT binding protein) and RanGAP1 (GTPase regulator protein) mark the
CDS, meaning that both proteins remain at CDS after PPB MTs disassembly. At the end of cytokinesis, the new cell wall fuses to
the parent cell wall at the site of CDS.
The location of the PPB is determined by internal and external forces (stresses) that affect cytoskeletal patterns in the cell. In pre-
mitotic cells, cytoplasmic strands populated by MTs and MFs span the space between the nucleus and the plasmalemma. During
preprophase, these cytoplasmic strands coalesce into a cytoskeletal structure known as the phragmosome that bridges the nucleus
and the cell cortex. The phragmosome functions to recruit and stabilize CMTs into the PPB. These strands are under tension and
are believed to “sense”the geometry of the cell in order to “select”nuclear location and the division plane. A dynamic cytoskeleton
is critical to allow the population of MTs to “explore”(through dynamic instability) cell geometry and determine a minimal divi-
sion plane. Internal stress created by forces such as turgor pressure and cell geometry will affect the orientation of the interphase
CMT array and the PPB, and therefore the location of the CDS. Intact MFs are required for such reorientations, thus actin may
play a role in interpreting these stresses. Kinesin-14 proteins may play a role in the interaction between MTs and MFs in the forma-
tion of the phragmosome.
During PPB formation, MTs of mixed polarity are enriched at the cell equator, and the cell cortex above and below clears of
MTs. MAPs most likely mediate the switch from the interphase CMT array, which covers the entire cell cortex, to the equatorial
enrichment of MTs that is characteristic of the PPB. PPB formation occurs via changes in the rates of MT polymerization and
depolymerization, as well as selective MT stabilization and destabilization. In addition, components of the g–TuRC localize
to the PPB and probably promote de novo MT nucleation. MTs in the PPB are organized into bundles, and these bundles contain
highly stable MTs.
The PPB contains an array of MAPs. Many of the proteins required for PPB formation or that associate with the PPB serve
multiple functions in both mitosis and interphase. Many of these proteins alter MT dynamics and include MAP65,MOR1,
TON1,TON2 and CLASP.MOR1 localizes to PPBs and other MT arrays. CLASP is implicated in PPB formation and narrowing.
The SABRE protein plays important roles in orientation of cell division and planar polarity. Moreover, SABRE stabilizes the orien-
tation of CLASP-labeled MT in the PPB, which is essential for cell division plane orientation. MAP65 is involved in PPB formation by
forming cross-bridges between overlapping MTs, thus forming MT bundles. TON1 proteins co-localize with PPBs and the loss of
TON1 prevents PPB formation. Katanin proteins appear to function in sensing cell geometry for the establishment of cell division
plane. Katanin mutants exhibit disorganized interphase MT arrays, altered phragmoplast function, oblique division planes, and
alterations in PPB organization and duration.
TAN proteins form a broad ring around the cell circumference that co-localizes with the PPB and remains at the CDS after PPB
disassembly. After mitosis, the TAN ring becomes more tightly focused as the phragmoplast approaches the CDS. TAN is not
required for PPB formation or positioning but does facilitate the guidance of the expanding phragmoplast to the CDS. Similarly
to TAN,AIR9 is localized at the CDS during different stages of cell division. TAN and AIR9 are similar in that they both have separate
domains mediating early versus late localization of CDS and require the PPB for late CDS localization. CDS maintenance after PPB
disassembly is an unexpectedly dynamic process. The CDS is not ‘set up’by the PPB and subsequently maintained in a static
manner, but rather is continuously modified during mitosis and cytokinesis by as yet undescribed mechanisms involving TAN
and AIR9 proteins.
In addition to containing MTs, the PPB also contains actin filaments. Intact MTs are critical for formation of the actin filaments in
the PPB. The actin PPB is wider than the MT PPB. Actin in the PPB appears to constrain the MTs of the PPB, because actin depo-
lymerization during preprophase/prophase leads to dramatic widening of the MT PPB. Disruption of MF function, either by depo-
lymerizing agents or mutations results in defects in establishing the division plane. The precise mechanisms by which MFs interact
with MTs in the PPB and the various proteins that may be involved, have yet to be determined. However, formin 14 is found on the
PPB and directly binds and bundles AFs and MTs in vitro.Kinesin localizes to MTs and AFs in the PPB and also may act as an MT–MF
cross-linker.
Mitosis –Spindle Structure and Function
The mitotic spindle consists of a complex, highly organized array of MTs and MFs that replaces the PPB. The spindle has several
populations of MTs, bundles of kinetochore microtubules (kMTs), as well as other MTs that may or may not end at one or the other
spindle pole regions. Unlike animal cell spindles, with sharply focused spindle poles (due to the centrioles at the spindle poles),
plant spindle poles (lacking centrioles) are broader, usually consisting of numerous MT foci. The primary function of the spindle
is the separation and transport of complete sets of chromosomes to the daughter cells. Spindle formation in plants differs from most
other eukaryotes because plant cells lack centrosomes or spindle pole bodies, which act as MTOCs.
The formation and location of the spindle appears to be influenced by the PPB. Spindle assembly starts prior to PPB breakdown
at prometaphase, with the spindle axis perpendicular to the plane of the PPB. MTs that connect the nucleus and PPB are important
for proper spindle assembly and a timely progression through the cell cycle. Cells that either have no PPB or a double PPB initially
form multipolar spindles, although a normal bipolar spindle eventually forms in these cells. Such cells also exhibit a longer progres-
sion through metaphase.
The Plant Cytoskeleton 15
Reference Module in Food Science, First Edition, 2016, 1–22
Author's personal copy
Several MAP proteins (MOR1/TMBP200,CLASP, and MAP65) are implicated in spindle formation. Targeting Protein (TPX2),
a regulator of spindle assembly in animal cells, has been identified in plants. TPX2 localizes to the inside of the nucleus from inter-
phase to G2, to a perinuclear region in early prophase, and to the spindle in mitosis. TPX2 binds importin aand promotes spindle
assembly in vitro and is therefore believed to be involved in the RanGTP-regulated spindle assembly pathway. TPX2 may be required
to initiate plant spindle assembly.
The mitotic spindle in plant cells originates from the nuclear envelope in prophase. Spindle MTs emanate from nuclear
envelope-associated g-tubulin that becomes concentrated at the polar regions. The broad spindle poles consist of numerous subsets
of converging MTs. These perinuclear MTs are initially distributed randomly along the surface of the nuclear envelope, with their
fast-growing ends radiating out into the cytoplasm in all directions. With time, MTs become more focused into the plane of the PPB.
As the nuclear envelope breaks down, g-tubulin moves into the metaphase spindle and is most concentrated at spindle poles in
anaphase. NEDD1, which acts as an anchoring factor of g-tubulin complexes, decorates spindle MTs preferentially toward theirs
minus ends. In anaphase, kMTs (via MT depolymerization) move chromosomes towards the poles as non-kinetochore, interpolar
MTs, slide past one another to elongate the spindle and further separate the chromosome masses. g-tubulins then move from poles
(distal surfaces of telophase nuclei) to proximal surfaces of daughter nuclei where they nucleate the bipolar array of phragmoplast
MTs that direct cell plate deposition during the completion of cytokinesis. Spindle structure is stabilized by the kinetochores on the
chromosomes. Plant kinetochores appear to initiate the polymerization of their own kinetochore MTs in early prometaphase,
contributing to the structure of the spindle.
At least eight different kinesins are involved in various aspects of spindle function. Some facilitate chromosome movement, some
function to stabilize the binding of kMTs to chromosomes, and some modulate spindle dynamics, affecting spindle polarity and
length. Kinesin-14 (the only minus-end directed kinesin) has been implicated in the gathering of MT minus ends into spindle poles
and the compaction of non-kinetochore MTs, resulting in a more compact plant spindle.
A cage of MFs surrounds the spindle and connects it to the cell periphery, thereby maintaining the spindle’s position during
mitosis. Disruption of the MF network results in mis-oriented spindles and oblique cell plates. Kinesins have been implicated in
cross-linking MTs and MFs at various stages of the cell cycle (PPB, spindle, phragmoplast). MAP190 co-sediments with both MFs
and MTs and is localized in the spindle.
Cytokinesis
New cell walls are formed by phragmoplasts, a complex of cytoskeletal arrays (MTs and MFs), and the developing cell plate (which
forms from the fusion of Golgi-derived vesicles and matures into a new cell wall encased in new plasmalemma). The developing cell
plate begins with the accumulation of Golgi derived vesicles in the mid-region, between arrays of MTs extending from the cell
periphery (location of the Golgi bodies) to the mid-region, where the MTs arrays overlap. Golgi-derived vesicles migrate along
MTs via þend directed kinesin proteins. The phragmoplast progressively expands during cell-plate formation via MT depolymeriza-
tion at the phragmoplast center and new polymerization on its outside edges. MTs on opposite sides of the mid-region are of oppo-
site polarity. The polarity of the phragmoplast MTs is not generated by preferential nucleation, but is instead derived from the
pre-existing polarity of the spindle MTs, with a majority polymerizing (þend) towards the cell plate. Depolymerization is regulated
by a mitogen-activated protein kinase cascade that prevents MAP65 cross-bridging activity at the mid-zone. As Golgi vesicles fuse
(via dynamin-like proteins) to form a tubular network, MT arrays begin to migrate towards the cell periphery. As the cell plate tran-
sitions from a tubular network to a fenestrated sheet, MTs become more prevalent at the outer margins of the sheet and less apparent
(absent) in the cell center.
The interaction between the growing phragmoplast and the CDS remains to be fully described. MTs and MFs have been
implicated as bridging structures between the phragmoplast and CDS, but more work is required to clarify their role in
CDS maintenance. Mutants have been identified that lack proper guidance of the phragmoplast to the CDS, but still appear
to form normal PPBs, phragmoplasts and cell plates. Mutations related to altering TAN activity (an MT binding protein that
localizes with the PPB and remains in the CDS during mitosis) result in a failure to properly guide the phragmoplast to
the CDS.
Maturation of the cell plate refers to the biochemical and physical changes which occur as the cell plate fuses with the parent cell
wall. Callose, a polysaccharide polymer, is observed in young cell plates, which later are enriched with cellulose. Several proteins
have been identified that have potential roles in cell plate anchoring and maturation. RSH (an extensin-like protein) may be
involved in generating connections between the cell plate and the mother cell. The TPLATE (an adaptor/coatomer-like protein)
is present at the growing edge of the cell plate, and appears at the CDS shortly before fusion of the cell plate with the parental
wall. The potential cell-plate maturation factor, (AIR9), initially co-localizes with the PPB, and then disappears from the CDS until
late cytokinesis. AIR9 has two distinct domains that localize it to the division site during different phases of the cell cycle. After
fusion of the cell plate and parental wall, AIR9 assumes a filamentous pattern at the CDS and extends into the cell plate before
the interphase MT array is reestablished.
In addition to their involvement in the formation and function of the PPB and/or spindle, MOR1,CLASP,MAP65,AtKRP125c,
NEDD1, and GCPs also contribute to phragmoplast establishment and configuration. Members of the kinesin-12 family localize to
the mid-zone of the phragmoplast. The phragmoplast fails to assemble normally in the absence of kinesin.Kinesin may play a role in
phragmoplast formation by precluding the plus ends of the opposing MT sets from crossing the mid-zone. MT interdigitation in the
phragmoplast depends on kinesin, which may function as a motor for vesicle transport in the phragmoplast.
16 The Plant Cytoskeleton
Reference Module in Food Science, First Edition, 2016, 1–22
Author's personal copy
The role of MFs in the phragmoplast is less clear. Because arrays of MFs and MTs closely co-exist and play important roles in the
phragmoplast, cooperation or interaction between MFs and MTs is assumed. Several proteins have been proposed to mediate the
cooperation or interaction between MFs and MTs in the cytoskeletal structure. Myosin VIII links phragmoplast MTs to the cortical
division site via MFs during phragmoplast expansion. MFs may interact with the MTs, thus connecting the cell cortex and the
phragmoplast.
Cytoskeletal Functions
Cellulose Microfibril Organization
Since their discovery in the early 1960s, interphase arrays of MTs have been postulated to be involved in controlling the orientation
of cellulose microfibrils in the developing cell wall. The patterning of the most recently deposited cellulose microfibrils controls the
expansion characteristics of the wall and thus the cell, resulting in either isotropic growth (growth in all directions) or anisotropic
growth (expansion predominantly in one direction). Evidence for the involvement of MTs has been largely circumstantial and can
be basically broken down as follows; 1) the orientation of the most recently deposited microfibrils follows the orientation of the
CMT; 2) disruption of CMT results in a loss in microfibril patterns; 3) changes in microfibril patterns are predicted by changes in
CMT patterns; 4) CMTs are very close to the plasmalemma, the site of cellulose synthesis; 5) when cells deposit localized microfibril
patterns (wall thickenings in xylem cells), CMTs are found in localized arrays, adjacent to and paralleling the micofibrils in the wall
thickenings.
Numerous models have been proposed for how CMTs in the cytoplasm influence the patterns of cellulose microfibrils synthe-
sized on the outer surface of the plasmalemma. The most recent observations on this topic indicate that the cellulose synthesizing
complexes (CESA) are directly attached to CMT by a CESA interactive protein (CSI). Several different CSI proteins (CSI1, CSI3) have
been identified. Simultaneous, live cell in vivo imaging of CESA,CSI and CMTs shows that CESA proteins move along CMTs, in
association with CSI. Binding between CSI and MTs has been show in vitro and disruption of CMTs results in significant changes
in CSI patterns (Figure 1). Without CSI,CESA complexes do not interact in CMT and cellulose microfibril patterns form indepen-
dent of CMT patterns. The association with CSI simply directs that movement along the CMTs. Although the movement of CESA in
the plane of the plasmalemma was originally believed to be the result of polymerization and crystallization of the cellulose micro-
fibril, more recent data indicates that movement of CESA is too rapid to be the result of polymerization/crystallization alone. The
mechanism by which CESA moves along CMTs remains unclear.
Cell Growth and Shape
The specifics of tissue structure and function in plants is achieved through metabolic specialization and morphogenetic programs
that regulate the cell’s capacity to expand and change shape. The chemical composition and structural organization of the cell wall
plays a pivotal role in plant cell expansion. It is the interaction between turgor pressure and the mechanical properties of cell walls
that defines the direction and extent of wall, and thus cell expansion. For cell growth to occur, the wall has to allow extension by
yielding to turgor pressure and, at the same time, new wall material has to be deposited to reinforce the structural stability of the
wall to resist turgor pressure. Internal pressure and the vicinity of neighboring cells exert mechanical stimuli, which are sensed by the
cell in expansion. Mechano-sensing and mechano-transduction appear to alter gene activity and affect subsequent plant morpho-
genesis. Changes in the mechanical stress within cells is mediated by MTs and promotes changes in morphogenesis. Therefore, the
cell’s ability to detect and respond to mechanical stress (whether from internal turgor pressure or external neighboring cells) is
central to proper morphogenesis.
Plant cells display two main mechanisms of growth: “diffuse”and “tip-growth”. These terms refer to the specific regions of the
cell which are expanding. Diffuse growth indicates that new wall and plasmalemma are added in all regions of the cell, whereas tip
growth refers to the addition of new wall and plasmalemma to a specific location in the cell, which becomes the growing tip.
However, within diffuse growth cells can expand globally (isotropically) or directionally (anisotropically).
Diffuse Cell Expansion
MTs play an important role in modulating plant cell growth. Loss of, or alterations in MT arrays cause a loss of growth direction and
isotropic swelling. Although MTs are not involved in cell growth directly, they do influence the direction of growth (isotropic or
anisotropic). MTs do not directly stiffen the cell, but guide cellulose microfibril deposition (see previous section). Alterations in
either MTs or CSI proteins result in changes in cellulose microfibril organization and thus changes in subsequent cell
morphogenesis.
Many studies describe altered cell expansion in transgenic plants with altered expression patterns of tubulin or other genes
related to MT dynamics/homeostasis. For example, mutations affecting the incorporation of a-tubulins result in organs with
a predominant left-handed twist (lefty mutation). Transgenic plants overexpressing MAP20 also show left-handed twisting. Plant
cell growth is modulated in response to the mechanical stress generated by adjacent cells. Katanin (an MT-severing protein)
promotes the organization of MTs in to parallel arrays and is required to orchestrate growth between neighboring cells by providing
cells with the ability to respond efficiently to mechanical stress. At the protein level, katanin severs MTs and enhances MT interac-
tions and bundling. At the cellular level, katanin activity enhances CMT self-organization into parallel arrays and these parallel CMTs
The Plant Cytoskeleton 17
Reference Module in Food Science, First Edition, 2016, 1–22
Author's personal copy
drive growth anisotropy. At the tissue level, differential growth (anisotropic growth) modifies the pattern of mechanical stress in
neighboring cells, which in turn controls the CMT orientation and thus growth anisotropy. This in turn further amplifies differential
growth patterns in a feedback loop.
Actin also plays a role in diffuse cell growth. Localization of fine and bundled MFs affects subsequent cell expansion. Loss of MF
organization results in small, misshapen cells and mutations in villin genes (which promote actin bundling) results changes in cell
elongation and twisting of organs during development.
Tip Growing Cells
Tip growing cells (root hairs, pollen tubes) exhibit several regions of development. The apex is the region where new membrane and
wall material is added, the apical flank and sub-apex regions accumulate the vesicles and the non-growing shank region contains the
central vacuole and most of the organelles needed to produce the new membrane and wall material for cell expansion. (Figure 8)
Tip growing cells require the accumulation of large quantities of new wall and membrane material into a discrete location (the
growing cell tip). Vesicles, containing wall and new membrane are transported along MF cables in a myosin-dependent manner.
In the tip region, an array of fine, cross-linked MFs functions to allow secretory vesicles (containing new membrane and wall
Figure 8 Working model to explain the control of actin cytoskeleton in different stages of root hair development. (a) Schematic of actin architecture
and its regulation in the root hair initiation stage. ROPs (contribute to actin polymerization at the bulge. (b) Schematic of actin architecture and its
regulation in the root hair elongation stage. Calcium facilitates actin disassembly while ROPs and PI(4,5)P2 prompt actin assembly, which results in
actin turnover rapidly at the tip. (c) Schematic of actin architecture and its regulation in mature root hairs. Calcium, PI(4,5)P2 and ROPs disappear at
the tip and actin bundles are distributed in the whole hair. (ROP-G-protein regulators; VLN, villin; PRF, profilin; ADF, actin depolymerizing factor;
Arp2/3, Arp2/3 nucleating factor; FH, formin; WAVE, MF nucleation factor) The image is not to scale. Bar ¼0.5 mm. From Pei, W., Du, F., Zhang, Y.,
He, T., Ren, H., 2012. Control of the actin cytoskeleton in root hair development. Plant Sci. 187, 10–18.
18 The Plant Cytoskeleton
Reference Module in Food Science, First Edition, 2016, 1–22
Author's personal copy
components) to approach the plasmalemma but exclude large organelles. These two arrays of MFs (sub-apical MF cables and fine,
cross-linked arrays in the cell apex) work together to transport secretory vesicles form sub-apical regions to the growing tip and then
facilitate the binding of those vesicles to the plasmalemma. MTs are evident in sub-apical and shank regions of tip-growing cells
(Figure 9). Although the disruption of MTs does not affect vesicle migration towards the tip or cell elongation, the identification
of kinesin-related proteins in pollen tubes suggests a possible role in vesicle transport. Inhibition of MTs has been shown to affect the
movements of the generative cell and the pulsatory growth observed in pollen tubes. Unlike cells growing by diffuse growth, MTs in
tip-growing cells do not appear to be involved in controlling the organization of cellulose microfibrils in the cell wall.
New actin filaments probably arise from membrane associated protein complexes containing formin. The function of formin in
root hairs mimics its role in pollen tubes, where it nucleates actin assembly and directs the formation of actin cables at the shank.
Actin filaments develop as very short filaments associated with the plasmalemma and then organize into a fine MF array in the apical
region and finally into the MF cables observed in in sub-apical region. Formin and Arp2/3 complex proteins play a role in the arrange-
ment of actin in epidermal cells during the initiation of root hairs and subsequent tip growth (Figure 8) but is not involved in deter-
mining the location where roots hairs will form. The transition of MF arrays from short filaments in the tip to long cables along the
shank of the cell is believed to be regulated by a ROP-WAVE-Arp2/3 pathway.
Villin proteins function to form unipolar MF bundles in the sub-apical and shank regions. The combination of activities of villin,
gelsolin and profilin function to organize actin filaments into the two different arrays (MF cables and fine, cross-linked arrays). Profilin
accumulates near the bulge formation site as well as in the apical region of growing root hairs and likely participates in the poly-
merization of short actin filaments during bulge formation. Profilin:actin may also contribute to formin-mediated actin polymeriza-
tion in root hair development.
The polymerization and organization of MTs in tip-growing cells is not well understood. MTs are observed in the sub-apical and
shank regions of the cell. In the sub-apex, MTs are short and disorganized. In the shank, MTs are organized into bundles that parallel
the length of the cell. The mechanism by which MTs organize and function in tip-growing cells remains unclear.
Trichomes and Pavement Cells
Many plant cells exhibit specialized forms of cell expansion (divergent from a simple spherical or cylindrical form). Epidermal
cells, such as fibers, trichomes, and pavement cells, exhibit differential growth rates in different regions of the cell, resulting in the
variation in cell morphology observed in these cells. Although the directions of growth vary in these cells (trichomes grow as
a single projection, perpendicular to the epidermal plane, whereas pavement cells extend multiple lobes horizontally within
the epidermal plane), the mechanism by which these cells grow is essentially the same. Wall loosening in discrete locations
of the cell results in a turgor pressure driven bulge in that region of the cell. Wall loosening co-localizes with phospholipase
D(PLD) activity in the newly forming bulge area. Following the loss of the cortical attachments that hold the cytoskeleton to
the plasmalemma, wall weakening occurs and the turgor pressure driven bulge occurs. PLD activity results in the rearrangement
of CMTs. The initiation of polarized growth from this region of wall loosening seems to be intimately linked to the appearance of
an actin patch. Actin patches have been observed during the initiation of a variety of specialized cell morphologies (pollen tubes,
root hairs, leaf trichomes, and lobe-forming epidermal pavement cells. Disruption of the actin patches results in aberrant cell
morphologies. Although the precise organization of the actin in these regions is unclear, APR2/3 proteins (MF polymerizing
proteins) accumulate in the bulge. APR2/3 activity is modulated by Rho-GTPase activity and such proteins in plants (ROP –
Rho-GTPase in plants) accumulate during bulge initiation. Actin polymerization could contribute directly to membrane protru-
sion in plants. Actin aggregates coincide with the regional clustering of CMTs, suggesting cooperation between MFs and MTs.
MTs, with their own set of interactors and regulators, might act to localize and focus the actin-based cell expansion. A ROP
signaling network controls the development of lobes in pavement cells. Local activation of ROP increases MF dynamics to
promote localized growth. In addition, ROP activity also promotes the disorganization of MT arrays in this region (Figure 10).
Pavement cells form tight associations due to the precise alignment of lobes (convex regions) from one cell with invaginations
Figure 9 Cellular organization of MTs and MFs in a typical tip-growing cell. An idealized tip-growing cells, showing CMTs as linear red structures.
MF bundles (thick blue lines) also appear as longitudinal structures in the shank region of tip-growing cells. Fine MF arrays (thin blue lines) are
distributed both cortically and throughout the cytosol, but most prominently appear as the “apical actin structure”just behind the cell apex. Reprinted
from Rounds, C.M., Bezanilla, M., 2013. Growth mechanisms in tip-growing plant cells. Annu. Rev. Plant Biol. 64, 243–265.
The Plant Cytoskeleton 19
Reference Module in Food Science, First Edition, 2016, 1–22
Author's personal copy
(concave regions) in the adjacent cell. This alignment requires the coordination of cytoskeletal elements in the adjoining cells.
Lobe (convex) regions are enriched in fine F-actin and poor in microtubules, thus promoting cell expansion, while invaginated
areas (concave regions) in adjacent cells are rich in microtubule arrays that restrict cell expansion. This interdigitated architecture
is finely regulated by auxin-induced activation of different ROPs in the two adjacent pavement cells, resulting in the promotion of
either MT or MF arrays in areas of invagination or lobes, respectively.
Tracheary Elements
Tracheary elements (TEs) function in water (and dissolved nutrient) transport from roots to the upper portions of the plant. To
withstand the negative water potential pressures, TEs develop secondary wall thickenings which reinforce the wall and prevent
cell collapse. First formed TEs (protoxylem) develop wall thickenings in spiral/helical pattern around the circumference of the
cell. Later formed TEs (metaxylem) form more intricate wall thickenings with a reticulate or pitted pattern of wall thickenings.
MTs have long been proposed to be the causal agents responsible for the deposition of the thickened wall band observed in xylem
cells. As elongated cells begin the transition to a xylem cell, CMTs, originally evenly dispersed perpendicular to the long axis of the
cell, begin to laterally associate into CMT bundles. As cells progress from a helical to a reticulate pattern of wall thickening, changes
in MT patterns predict the locations of new or subsequent wall thickenings. This reorganization of CMTs precedes the actual depo-
sition of new wall components.
The mechanism by which CMTs reorganize remains unclear. Some evidence indicates that this reorganization of CMTs does not
require MT turnover. Cells with taxol-stabilized MTs still form CMT bundles and secondary wall thickenings. CMT bundling
involves MT bundling proteins, such as MAP65 to cross-link CMTs into bundles. Other data suggests that MIDD1 (a þend MT
binding protein) is involved in depolymerizing CMTs in pit areas (regions of the wall where no secondary wall deposition occurs).
MIDD1 specifically co-localizes with CMTs in pit regions and not with CMTs outside of pit regions. Loss of MIDD1 results in
secondary cell wall lacking pits. MIDD1 binds to the plasmalemma, independent of either CMTs or secondary cell wall. ROPs
bind to the plasmalemma creating a template of ROP domains. Once the template is established, MIDD1 accumulates at the plas-
malemma domains created by ROPs. Kinesin (an MT depolymerizing kinesin) binds to MIDD1, resulting in the depolymerization of
the CMTs on the template. Subsequently, MAP70 functions to define the boundary for secondary wall deposition by adjusting the
patterns of CMTs to shift the secondary wall pattern. Thus, MIDD1 promotes local depletion of CMTs while MAP70 modifies the
shape of the MT-depleted domains. Subsequent cell wall thickenings are defined through the activities of katanin and MAP65
which organize the CMTs beneath the future secondary wall thickening. Once the new CMT array is established, FRA1 (kinesin
Figure 10 Interdigitated Growth of Plant Pavement Cells. Pavement cells contain alternating lobes and indentations. The auxin efflux carrier PIN1,
which specifically localizes to the plasmalemma of lobes, promotes auxin accumulation in the cell wall between lobes and indentations. This auxin
activates two small GTPases (ROPs) that propagate different responses in adjacent pavement cells. At the lobe of one cell, ABP (Auxin binding
protein) switches on ROP2, which promotes the assembly of cortical MFs through RIC4. In the adjacent cell, auxin activates ROP6, which promotes
the binding of RIC1 to CMTs. This results in the formation of well-ordered bundles of MTs that restrict expansion of the cell and generate an indenta-
tion. The two pathways function in an antagonistic manner; ROP2 suppresses RIC1 activity, whereas well-ordered MTs repress the interaction of
ROP2 and RIC4. From Pietra, S., Grebe, M., 2010. Auxin paves the way for planar morphogenesis. Cell 143, 29–31.
20 The Plant Cytoskeleton
Reference Module in Food Science, First Edition, 2016, 1–22
Author's personal copy
motor protein) and MAP20 (may promote the activity of cellulose synthesis along CMTs) control the orientation of the cellulose
microfibrils along the CMTs.
The production of secondary wall thickenings requires the localized deposition of secondary cell wall components into discrete
locations. These components are delivered to the plasmalemma via Golgi-derived vesicles. Two arrays of MFs (thick bundles and
fine filaments) appear to be required for the deposition of cell wall material to the secondary wall thickenings of TEs. Thick cables of
MFs, running along the length of the cell, perpendicular to the MT arrays, direct the delivery of vesicles containing secondary wall
components to the plasmalemma. Movement of vesicles along these cables is interrupted at the CMT bundles associated with
secondary wall thickening. Attachment of vesicles to the plasmalemma occurs at CMT bundles, however removal of these MTs using
oryzalin does not prevent vesicle insertion at this site. The existence of transversely oriented fine MF arrays in the developing xylem
has been confirmed in young developing xylem at an early stage of secondary wall synthesis. These fine arrays co-localized with the
MT bands and remain in place if MTs are removed. Interaction between MFs and the plasmalemma requires membrane bound poly-
phosphate phosphatase. Mutations in this protein (FRAGILE FIBERS3 [FRA3]) alters actin organization (CMT patterns remain unal-
tered) which causes a reduction in secondary wall synthesis.
Conclusion and Future Perspectives
The combination of molecular genetic analysis, live-cell imaging, biochemistry and computer simulations have provided new
insight to the roles of MTs and MFs in cell and plant development. Further technological advances in these techniques will poten-
tially push forward the investigation of cytoskeletal components, distribution and function in plants.
Many processes in plant morphogenesis (such as cell cycle, cell expansion, cell differentiation, responses in to internal and external
signals) are now know to require the proper functioning of both MTs and MFs. Processes that, in the past, were thought to involve only
MTs (such as mitosis, cytokinesis, directed cell expansion) have now been shown to also require properly functioning MFs.
Plants have evolved the use of the highly conserved structure of the MT in numerous, novel ways. In addition to MT arrays
common to other eukaryotes (mitotic spindles), plants have evolved other, unique division-specific arrays such as the PPB, phrag-
mosome and phragmoplast. These novel MT arrays most likely evolved in response to the presence of the rigid cellulose wall, itself
a product of MT function during interphase.
Significant progress has been made regarding the mechanisms by which plant cells, lacking recognizable MTOCs, organize
the various MT arrays found during the cell cycle and subsequent morphogenesis. New details regarding the location, compo-
sition, dynamics and function of MT nucleation complexes (g-TuRCs) has resulted in testable hypotheses regarding these
processes. In addition, a better understanding of the roles of various associated proteins (MAPs and ABPs) has helped to clarify
how very similar structures (MTs and MFs) can be modified to produce the dichotomy of cytoskeletal arrays and functions. In
plants, many MAPs (such as kinesin,MAP215,CLASP,Augmin) that share homology of structure and function with similar
proteins in animals and yeasts have been well-studied and characterized. However studies of plant-specificMAPs (such as
MAP65,MAP70,AIR9,CSI) are fewer and work at every level is needed to elucidate the molecular mechanisms by which these
proteins modify cytoskeletal function. In addition, the number of plant-specificMAPs currently documented most likely repre-
sents only a fraction of the total number of such proteins functioning in plants. The cytoskeleton is regulated by MAPs acting
together and a future challenge will be to sort out the productive (functional) interactions from incidental associations.
Although many MAPs have been characterized in plants, it is largely unknown how these proteins spatially and temporally
modulate MT dynamics in living cells. Future studies should examine the role of MAPs in the spatial regulation of MT
dynamics. Posttranslational modification of tubulins will mostly likely play a more prominent role in the study of the inter-
actions between MTs and MAPs.
Significant advances have been made in our understanding of the function of MF arrays in plants. Large-scale sequencing efforts
have led to the identification of genes encoding various ABPs that determine how MFs are organized in a particular cell type. Under-
standing the mechanisms thatgovern actin organization is important for ultimately deciphering its function in plant cell development.
Although recent progress has revealed the importance of MF dynamics in cell division, and cell elongation/expansion in plants, there
are still “missing links”in our understanding of their detailed mechanisms. On the other hand, actin MFs are known to interact with
other cellular components, such as organelles, MTs and plasmalemma. Thus, extensive studies on the dynamics of various cellular
components would undoubtedly contribute to our further understanding of the roles of MF dynamics and organization in plant devel-
opment. Until recent times, actin was all but ignored in treatments of the expansion of diffusely growing cells. It has now become clear
that actin MFs can promote outgrowth in plant cells, as seen in pavement cells, and that in cooperation with MTs, plays a major role in
cell expansion and the construction of complex cell shapes. However, we have only just begun to discover the diverse roles of actin MFs
and the myriad signaling pathways that govern their organization and function.
MTs and MFs are often co-distributed during various stages of the cell development. These two cytoskeletal components may
cooperate in a spatially and temporally coordinated manner through specificMAP and ABP proteins or multi-protein complexes.
Although some have been identified, many more probably wait discovery. Over the past years, a growing number of proteins or
protein complexes that bridge these cytoskeletal systems have been identified, including MAP190,MAP18, various KCH proteins
and EB1. To elucidate the mechanisms of MT and MF interaction and the regulation of these interactions, the precise function
of cross-linking proteins (both currently known and those yet to be discovered) needs to be clarified and identified via proteomics
and creative genetic strategies.
The Plant Cytoskeleton 21
Reference Module in Food Science, First Edition, 2016, 1–22
Author's personal copy
References
Blume, Y.B., Baird, W.V., Yemets, A.I., Breviario, D. (Eds.), 2008. The Plant Cytoskeleton: A Key Tool for Agro-Biotechnology. Springer, Dordrecht.
Fosket, D.E. (Ed.), 1994. Plant Growth and Development: A Molecular Approach. Academic Press, San Diego, New York.
Hamada, T., 2014. Microtubule organization and microtubule-associated proteins in plant cells. Int. Rev. Cell Mol. Biol. 312, 1–52.
Henty-Ridilla, J.L., Li, J., Blanchoin, L., Staiger, C.J., 2013. Actin dynamics in the cortical array of plant cells. Curr. Opin. Plant Biol. 16, 678–687.
Li, S., Sun, T., Ren, H., 2015. The functions of the cytoskeleton and associated proteins during mitosis and cytokinesis in plant cells. Front. Plant Sci. 6, 282.
Nick, P. (Ed.), 2000. Plant Microtubules: Potential for Biotechnology. Springer-Verlag, Berlin, Heidelberg.
Nick, P. (Ed.), 2008. Plant Microtubules: Development and Flexibility. Plant Cell Monographs, vol. 11. Springer-Verlag, Berlin Heidelberg.
Oda, Y., 2015. Cortical microtubule rearrangements and cell wall patterning. Front. Plant Sci. 6, 236.
Rounds, C.M., Bezanilla, M., 2013. Growth mechanisms in tip-growing plant cells. Annu. Rev. Plant Biol. 64, 243–265.
Staigler, C.J., Baluska, F., Volkmann, D., Barlow, P.W. (Eds.), 2000. Actin: A Dynamic Framework for Multiple Plant Cell Functions, vol. 89. Kluwer Academic Publishers, Dordrecht,
Boston, London.
Struk, S., Dhonukshe, P., 2014. MAPs: cellular navigators for microtubule array orientations in Arabidopsis. Plant Cell Rep. 33, 1–21.
Verma, D.P.S., Hong, Z. (Eds.), 2007. Cell Division Control in Plants. Springer Science. https://books.google.com/books?isbn¼3540734872.
Wasteneys, G.O., Brandizzi, F., 2013. A glorious half-century of microtubules. Plant J. 75, 185–188.
22 The Plant Cytoskeleton
Reference Module in Food Science, First Edition, 2016, 1–22
Author's personal copy
