Emerging roles of cortical microtubule–membrane interactions
- 362 Downloads
Plant cortical microtubules have crucial roles in cell wall development. Cortical microtubules are tightly anchored to the plasma membrane in a highly ordered array, which directs the deposition of cellulose microfibrils by guiding the movement of the cellulose synthase complex. Cortical microtubules also interact with several endomembrane systems to regulate cell wall development and other cellular events. Recent studies have identified new factors that mediate interactions between cortical microtubules and endomembrane systems including the plasma membrane, endosome, exocytic vesicles, and endoplasmic reticulum. These studies revealed that cortical microtubule-membrane interactions are highly dynamic, with specialized roles in developmental and environmental signaling pathways. A recent reconstructive study identified a novel function of the cortical microtubule-plasma membrane interaction, which acts as a lateral fence that defines plasma membrane domains. This review summarizes recent advances in our understanding of the mechanisms and functions of cortical microtubule-membrane interactions.
KeywordsRho-related GTPases from plants (ROP) IQ67-domain 13 (IQD13) Vesicle tethering (VETH) Microtubule depletion domain 1 (MIDD1) Cellulose synthase-microtubule uncoupling (CMU) Kinesin
Cortical division zone
Cellulose synthase complex
Microtubule-associated cellulose synthase compartment
Plant-specific Rho/Rac GTPase
Small CESA compartment
Plant cells are enclosed within rigid and extensible cellulosic cell walls. Plants undergo anisotropic cell growth, which is accompanied by a dramatic increase in cell volume and leads to tissue and organ development. Cellulose microfibrils embedded in the cell wall matrix determine the direction of cell growth by restricting cell expansion, which in turn determines the cell shape. Therefore, cellulose microfibril patterning is a critical factor that determines cell growth and tissue and organ development (McFarlane et al. 2014).
Land plants evolved a unique microtubule (MT) structure, the cortical microtubule (cMT) array. The cMT regulates the deposition pattern of cellulose microfibrils. In flowering plants, the transverse cMT array is observed in most interphase bipolar cells except for tip growing cells and highly specialized cells such as dead xylem cells. Electron micrographs reveal that cMTs occur in the vicinity of the plasma membrane (PM), with a constant distance from the membrane (Ledbetter and Porter 1963). Cross bridges are often observed linking the MT and PM (Hardham and Gunning 1978; Lancelle et al. 1986), leading to the hypothesis that cMTs are anchored to the PM by linkers. This hypothesis is supported by the facts that cMTs are preserved on membrane ghosts (Akashi and Shibaoka 1991), and cMTs rarely undergo lateral displacement (Shaw et al. 2003).
Cellulose microfibrils are synthesized by large cellulose synthase complexes (CSCs), which are composed of 18–36 cellulose synthase subunits and their accessory proteins (McFarlane et al. 2014). Cellulose synthases are assembled and matured to form active CSCs in the Golgi (Zhang et al. 2016), and then secreted to the PM through the trans-Golgi network (TGN) and small vesicles named SmaCC/MASC in the vicinity of cMTs (Crowell et al. 2009; Gutierrez et al. 2009; Sampathkumar et al. 2013). CSCs begin to synthesize cellulose in the space outside of the PM, which is thought to propel the CSC (Morgan et al. 2013). CSCs are tethered to cMTs through CSI1 (Bringmann et al. 2012; Gu et al. 2010; Li et al. 2012; Mei et al. 2012), which determines their trajectory along the cMTs (Paredez et al. 2006). The close interaction between cMT and PM is apparently indispensable for CSC guidance and proper patterning of cell wall deposition.
The cMTs are preferentially nucleated adjacent to preexisting cMTs, which generates a branch or bundle of cMTs. The cMTs are thought to be self-organized into arrays through their dynamic behaviors, including growth, shrinkage, and severing, as well as interactions between cMTs that lead to MT bundling or collision-induced catastrophe (Wasteneys and Ambrose 2009). Well-ordered cMTs are closely anchored to the PM, whereas discordant cMTs tend to be located at the cytoplasmic side of the PM. The parallel (ordered) cMTs preferentially serve as tracks for CSC movement, highlighting the importance of cMT-PM interactions (Barton et al. 2008). The mechanism underlying cMT-PM interactions has not been elucidated.
Microtubule-associated proteins (MAPs) mediate MT dynamics, interactions between MTs, and MT-membrane interactions. Membrane-protein interactions require distinct protein moieties such as lipid modifications, transmembrane domains, or lipid-binding domains (Cho and Stahelin 2005). Thus, cMT-membrane interactions are typically mediated by protein complexes that include MAPs, lipid-binding proteins, and scaffold proteins, although some MAPs can bind directly to membrane. The number of identified MAPs is continually increasing in plants (Hamada 2014). These MAPs are largely involved in MT dynamics and organization; only a few MAPs are involved in MT-membrane interactions. Recent studies identified additional MAPs and membrane proteins that function in cMT-membrane interactions, including the PM, endosome, exocytotic vesicles, and endoplasmic reticulum (ER). The activity and expression of these proteins are controlled in response to cellular and environmental signals, and they regulate cellular events such as cell wall deposition, cell polarity, and organelle organization. Recent studies revealed novel roles of cMT-membrane interactions, such as lateral stabilization of cMTs and lateral fences constraining PM proteins. This review summarizes recent work on the mechanisms and functions of cMT-membrane interactions in flowering plants.
PLDs regulate cMT stability in response to abiotic stress
Phospholipase D (PLD) hydrolyzes phosphatidylcholine to produce choline and phosphatidic acids (PAs), a major membrane constituent that acts as a signaling molecule. Two Arabidopsis PLD isoforms, PLDα1 and PLDδ, were reported to indirectly and directly regulate cMT-PM interactions, respectively. PLDα1 is activated by salt stress to produce PA (Yu et al. 2010), which can bind to MAP65-1 and promote its MT bundling activity, thereby stabilizing cMTs. Loss of PLDα1 or MAP65-1 enhances plant susceptibility to salt stress. Thus, PLD1α1 indirectly regulates the cMT-PM interaction through PA and MAP65-1 to increase salt stress tolerance (Zhang et al. 2012). PLDα1 has an opposing activity to cMT in guard cells, where it promotes MT disorganization in response to ABA treatment (Jiang et al. 2014). These results suggest that PLDα1 may mediate cMT-PM interactions through different PA-MAP complexes in different tissues.
PLDδ was reported to directly bind to MTs (Gardiner et al. 2001; Zhang et al. 2017). PLDδ is preferentially detected in the PM fraction (Gardiner et al. 2001; Wang and Wang 2001) and localized at the PM in vivo (Andreeva et al. 2009; Pinosa et al. 2013; Zhang et al. 2017), suggesting that PLDδ is a stable PM protein. A lack of PLDδ causes cMT hyperstabilization, suppressing heat stress-induced MT depolymerization. Consistent with this phenotype, PLDδ promotes MT destabilization in vitro (Zhang et al. 2017). Thus, it is likely that PLDδ directly mediates cMT-PM interactions to destabilize cMTs in response to heat stress.
ROPs regulate cMT organization
ROPs are plant-specific Rho/Rac GTPases. ROPs activate cellular signaling by changing the GTP/GDP state; GTP-bound ROP can interact with effector proteins to activate downstream signaling (active form), whereas GDP-bound ROP cannot interact with effector proteins and thereby inactivates downstream signaling (inactive form). The active form of ROP GTPase is anchored to the PM via C-terminal lipid modification and interacts with effectors to regulate a broad range of cellular events (Yalovsky 2015). Eleven Arabidopsis ROPs have been identified; ROP6 and ROP11 are involved in cMT organization. ROP6 recruits RIC1 effector protein, which directly binds to MTs (Fu et al. 2009). RIC1 also interacts with a katanin p60 subunit, KTN1, and promotes its MT severing activity. This signaling pathway promotes MT severing at MT branches, which consequently promotes parallel ordering of cMTs (Lin et al. 2013).
ROP11 regulates cMT organization in xylem vessels. ROP11 is locally activated to recruit MIDD1 scaffold protein, which in turn recruits a MT-depolymerizing kinesin, Kinesin-13A. This ROP11-MIDD1-Kinesin-13A pathway promotes local depletion of cMTs, resulting in the formation of pits in secondary cell walls (Oda and Fukuda 2012, 2013; Oda et al. 2010). Local activation of ROP11 in xylem cells is mediated by the ROP guanine nucleotide exchange factor ROPGEF4 (Oda and Fukuda 2012). In root hairs, some ROPGEFs including ROPGEF4 are activated by the FERONIA receptor-like kinase at the PM (Duan et al. 2010; Huang et al. 2013). Thus, receptor-like kinases might be involved in regulating the ROP11-MIDD1-Kinesin-13A pathway.
PHGAPs may be involved in preprophase band positioning
ROPs may be involved in cMT-PM interactions in mitotic cells. During preprophase, cMTs are arranged into a preprophase band (PPB), which establishes the cortical division zone (CDZ). The CDZ directs the fusion of the newly formed cell plate with the PM by attracting the phragmoplast MTs (Muller and Jurgens 2016; Schaefer et al. 2017). Putative ROPGAP proteins, PHGAP1 and PHGAP2, are required for accurate orientation of the PPB, phragmoplast, and cell plate. PHGAPs contain a pleckstrin homology (PH) domain, which typically mediates membrane interactions, and a putative GTPase-activating (GAP) domain. The putative GAP domain is assumed to inactivate ROP GTPases and is required for normal orientation of PPB and phragmoplast, suggesting that PHGAPs function by inactivating ROP GTPases. The TTP complex is required for PPB formation (Azimzadeh et al. 2008; Drevensek et al. 2012; Spinner et al. 2013); this complex contains TON1 proteins, TRM scaffold proteins, and PP2A phosphatases including a TON2/FASS regulatory subunit, all of which localize to PPB during preprophase. In contrast, PHGAPs are not specifically localized at the PPB but are widely localized at the PM, probably via the PH domain of PHGAPs. Thus, PHGAPs may indirectly regulate TTP position through ROP inactivation.
PPB recruits distinct proteins such as TAN (Walker et al. 2007), RanGAP (Xu et al. 2008), and POK2 kinesins (Lipka et al. 2014; Muller et al. 2006). PPB disappears during entry into the metaphase whereas these proteins remain in the CDZ. Then, PHGAPs are localized to the CDZ in a POK2-dependent manner. A feedback mechanism has been proposed, in which PHGAPs direct PPB to establish CDZ, which in turn recruits PHGAPs (Stockle et al. 2016).
IQD13 establishes a lateral fence for ROP GTPase
Studies on xylem vessel cells revealed that cMT acts as a lateral fence that restricts the localization of ROP GTPases. In differentiating metaxylem cells, PM domains containing activated ROP11 (hereafter referred to as the ROP domain) depolymerize cMTs, whereas cMTs surrounding the ROP domain restrict the area of the ROP domain (Oda and Fukuda 2012, 2013; Oda et al. 2010). The mechanism by which cMTs restrict the ROP domain is unknown. Recently, an IQD family member (Abel et al. 2005), IQD13, was identified as a xylem-specific MAP (Sugiyama et al. 2017). IQD13 is associated with both PM and cMT in vivo, suggesting that IQD13 could be a scaffold linking the cMT and ROP domains (Fig. 1a).
To test this hypothesis, the authors examined the effect of IQD13 on the shape of ROP domains. Coexpression of ROPGEF4, ROPGAP3, and ROP11 enables ROP11 to be locally activated on the PM, which in turn generates ROP domains even in non-xylem cells (Oda and Fukuda 2012). Using this approach, ROP domains were generated in tobacco leaf epidermal cells in the presence or absence of IQD13. In the presence of IQD13, ROP11 forms narrow, elongated ROP domains within the lattice of cMTs. In the absence of IQD13, ROP11 forms round ROP domains that are independent of cMTs. Truncated IQD13 lacking the PM-associated N-terminal domain failed to narrow the PM domains or restrict them within the cMT lattice, suggesting that PM association of IQD13 is required for shaping ROP domains. Thus, IQD13 likely functions in cMT-PM linkage, which inhibits the lateral diffusion of activated ROP11 on the PM. Consistent with this hypothesis, IQD13 overexpression causes narrower secondary cell wall pits, whereas IQD13 knockout causes rounder secondary cell wall pits (Fig. 1a). IQD13 promotes MT growth, which consequently increases the amount of cMTs. Thus, IQD13 synergistically fences the ROP domains by simultaneously promoting cMT polymerization and cMT-PM interaction (Sugiyama et al. 2017).
ROPs form PM domains in various cell types including leaf epidermal cells, root hairs, and pollen tubes (Yang and Lavagi 2012). IQD14, which redundantly functions with IQD13 (Sugiyama et al. 2017), is broadly expressed in plants (Burstenbinder et al. 2017b). Most of the 33 IQD members in Arabidopsis are localized to cMTs (Burstenbinder et al. 2017b; Hamada et al. 2013). Therefore, IQD14 and potentially other IQD members may regulate ROP domains in non-xylem cells by promoting cMT-PM interactions. The cell wall affects the trajectory and speed of PM protein diffusion (Martiniere et al. 2012). The results of recent studies indicate that cMTs also affect the behavior of PM proteins depending on cMT-PM linkers.
MAPs involved in cellulose synthesis
The cMT rarely displays lateral displacement due to its tight association with PM (Shaw et al. 2003). Recent work reported that cellulose-microtubule uncoupling (CMU) 1 and CMU2 are novel MAPs that regulate the lateral stability of cMTs on the PM (Liu et al. 2016). The cmu1 cmu2 double mutant plants fail to guide the trajectory of CSCs along the cMT, and display abnormal lateral displacement of cMTs. Loss of CSI1, which mediates interactions between the CSC and cMTs, rescued the lateral cMT displacement phenotype of cmu1 cmu2 plants. Because CMUs are localized along cMTs but not colocalized with CSI1 or CSC, CMUs likely stabilize cMT-PM interactions rather than mediate MT-CSC interactions (Liu et al. 2016). A lack of CMUs apparently has no effect on the amount of cMTs, suggesting that CMUs specifically regulate cMT lateral stability. Thus, cMT anchorage to the PM is likely achieved independently of CMUs. How CMUs associate with the PM is unknown. IQD1 interacts with CMU1 (Burstenbinder et al. 2013), and IQD1 overexpression promotes CMU1 localization to cMTs, suggesting that IQD1 is involved in cMT-PM interactions through CMU1 (Burstenbinder et al. 2017a, 2013).
The transmembrane proteins companion of cellulose synthase (CC) 1 and CC2 are essential factors for sustained cellulose synthesis under salt stress (Endler et al. 2015). CC1 and CC2 colocalize with CSCs in TGN and with moving CSC particles along cMTs. CCs interact with cellulose synthase subunits in yeast; thus CCs are proposed to be a component of CSCs. CC1 also interacts with MTs and promotes MT polymerization in vitro. The cMTs and CSC activity are hypersensitive to salt stress in cc1 cc2 double mutants (Endler et al. 2015). These recent studies show that CCs together with CSCs contribute to cMT-PM interactions and promote cMT formation under salt stress.
CLASP mediates endosome targeting
CLASP is a MAP involved in cMT organization and cell morphology (Ambrose et al. 2007; Ambrose and Wasteneys 2008; Kirik et al. 2007). Loss of CLASP causes the partial detachment of cMTs from the PM and the increased lateral movement and co-alignment of cMTs in leaf epidermal cells, suggesting that CLASP activity affects cMT-PM interaction or cMT–cMT interactions (Ambrose and Wasteneys 2008). CLASP also mediates cMT-endosome interactions (Ambrose et al. 2013). Sorting Nexin1 (SNX1) is a component of the retromer complex that regulates recycling of the auxin efflux carrier PIN-FORMED2 (PIN2), which maintains the polarized accumulation of PIN2 at the PM in Arabidopsis roots (Cui et al. 2010; Jaillais et al. 2006; Kleine-Vehn et al. 2008). CLASP interacts with SNX1, and the clasp null mutant reduces polar PIN2 accumulation and displays auxin-related root defects. Thus, CLASP is proposed to recruit SNX1 endosomes to cMTs, which consequently facilitates PIN2 accumulation at the PM (Ambrose et al. 2013). SNX1 has a PX domain that interacts with phosphoinositide in the membrane. Knockdown of formation of aploid and binucleate cells 1 (FAB1), which encodes a phosphatidylinositol 3-phosphate 5-kinase that produces phosphatidylinositol 3,5-bisphosphate [PtdIns(3,5)P2] from phosphatidylinositol 3-phosphate [PtdIns(3)P], impairs SNX1 loading to the endosome and causes mislocalization of PIN2. Thus, FAB1-generated PtdIns(3,5)P2 may mediate SNX1 loading to endosome (Hirano et al. 2011, 2015).
In meristematic cells, CLASP accumulates at the PM along the cell edge and promotes cMT bundling, which enables cMT elongation across the cell edges (Ambrose et al. 2011). During mitosis, CLASP preferentially associates with the PM rather than the spindle and phragmoplast (Kirik et al. 2007). This localization pattern suggests that CLASP can be anchored to the PM. SABRE encodes a PM transmembrane protein that is required for cMT organization, cell polarity, and cell fate determination, and is epistatic to CLASP (Pietra et al. 2013, 2015). SABRE might be involved in PM-CLASP interactions.
VETH mediates exocytotic vesicle targeting
Tethering protein complexes regulate the fusion of vesicles with the target membrane. The conserved oligomeric Golgi (COG) tethering complex directs retrograde trafficking to the Golgi in yeast. The exocyst tethering complex controls vesicle fusion with the PM. These tethering complexes are known to function independently (Vukasinovic and Zarsky 2016). Recent studies revealed that a COG subunit interacts with exocyst to mediate cMT-vesicle interactions (Oda et al. 2015; Vukasinovic et al. 2017). The novel coiled-coil proteins vesicle tethering 1 (VETH1) and VETH2, which are up regulated during xylem vessel differentiation, are localized at vesicle-like compartments moving along cMTs in cultured xylem cells. VETHs interact and colocalize with COG2, which is a COG complex subunit. VETH1 and COG2 also partially colocalize with the EXO70A1 subunit of the exocyst complex (Oda et al. 2015). Another study reported that COG2 interacts with EXO70A1 in vivo (Vukasinovic et al. 2017). During the early stage of xylem vessel cell differentiation, exocyst subunits are localized at numerous PM foci independently of cMTs. As cell differentiation progresses, exocyst subunits translocate to cMT bundles. The translocated exocyst subunits are dramatically lost in response to drug-mediated cMT disruption, whereas MF disruption does not affect exocyst localization, suggesting that exocyst localization depends on cMTs in differentiating xylem vessel cells (Vukasinovic et al. 2017). Ectopic expression of VETH1 together with COG2 is sufficient to induce EXO70A1 translocation to cMTs (Oda et al. 2015), demonstrating that VETHs recruit the exocyst to cMTs through COG2.
EXO70A1 is broadly expressed in plants, and the exo70a1 loss-of-function mutant displays pleiotropic defects in cell development (Fendrych et al. 2010; Hala et al. 2008; Synek et al. 2006). In xylem vessel cells, loss of EXO70A1 causes deformed secondary cell wall thickenings and abnormal accumulation of cytoplasmic vesicles (Li et al. 2013; Tu et al. 2015). Loss of the EXO84B subunit reduces secondary cell wall thickening in xylem vessels (Vukasinovic et al. 2017). Thus, the exocyst complex likely directs exocytic vesicles to deliver cell wall materials that is required for secondary cell wall thickening (Fig. 1b). In xylem vessel cells, some exocyst loci still remain even after cMT disruption by drug treatment. Exocyst-PM interactions may be established after translocation to cMTs, which enables vesicles fused with the PM to secrete cell wall materials outside of the PM. Exocyst subunits localize to the PM independently of cMTs in epidermal cells (Fendrych et al. 2013). The cMT-targeting of the exocyst complex is specifically observed in xylem vessel cells. Xylem vessel cells form denser and faster CSCs along cMT bundles than in epidermal cells, suggesting that cellulose synthesis is more active in xylem vessel cells than in epidermal cells (Watanabe et al. 2015; Wightman et al. 2009). Given that VETH is preferentially expressed in xylem cells, the VETH-COG-exocyst pathway may specifically target exocytic vesicles to cMT bundles to facilitate active secondary cell wall development.
Kinesin motors transporting vesicles
The class-4 kinesin FRAGILE FIBER1 (FRA1) was first identified as a regulator of cellulose microfibril alignment in secondary cell walls (Zhang et al. 2010; Zhong et al. 2002). However, recent studies show that FRA1 is a processive motor (Ganguly et al. 2017; Zhu and Dixit 2011) involved in vesicle trafficking of non-cellulosic materials along cMTs in both primary and secondary cell walls (Kong et al. 2015; Zhu et al. 2015). The fra1 null mutant displays little defects in cMT organization, alignment, amount of cellulose microfibrils, and CSC velocity. Instead, fra1 plants display reduced cell wall thickness and slightly reduced pectin content in cell walls. Live imaging of fully functional fluorescently tagged FRA1 revealed that FRA1 is localized to particles that move along cMTs at much faster velocities than CSCs, supporting the hypothesis that FRA1 is not involved in cellulose microfibril synthesis (Kong et al. 2015; Zhu et al. 2015).
Recently, a kinesin-14 motor protein, kinesin-like calmodulin binding protein (KCBP), was found to serve minus-end-directed transport of nuclei and chloroplasts along MTs in the moss Physcomitrella patens (Jonsson et al. 2015; Yamada et al. 2017). Phospholipid liposomes are directly transported by KCBP without any cross linkers. KCBP likely has affinity to membrane and may interact with cargo other than nuclei and chloroplasts (Yamada et al. 2017). It is not known whether KCBPs in flowering plants transport vesicles or organelles along cMTs. During mitosis, Arabidopsis KCBP is localized to the PPB at the distal end of the phragmoplast and to CDZ together with its interactor AIR9. Based on this localization, it is proposed that KCBP reels the phragmoplast MTs into the CDZ by minus-end-directed motility rather than transporting any cargo (Buschmann et al. 2015). In the developing trichome, Arabidopsis KCBP does not show processive movement but rather shows only short movement on the cMTs. KCBP is localized at the tip of growing branches, and loss of KCBP causes deformation of cMTs and MF organization, suggesting that KCBP orchestrates cMT and MF organization (Tian et al. 2015).
cMT-ER membrane interactions
NET indirectly connects cMTs with ER
In plant cells, ER shape, localization, and dynamics are regulated primarily by MFs together with myosin motor proteins and MF-binding proteins (Cao et al. 2016; Griffing et al. 2014; Sparkes et al. 2009; Ueda et al. 2010; Wang et al. 2014; Wang and Hussey 2017). Several studies suggest that cMTs and ER interact in plants; cMTs promote ER branching even when MFs are removed. ER tubules slowly elongate along cMTs in the absence of MFs (Hamada et al. 2012, 2014).
Networked (NET) is an actin-binding protein family that is unique in plants (Deeks et al. 2012). Of the 12 NET proteins in Arabidopsis, NET3B and NET3C interact with PM, ER membrane, and MF, and are localized at ER-PM contact sites where ER membrane is closely attached to the PM (Wang et al. 2014; Wang and Hussey 2017). NET3C also interacts with VAP27-1, a homologue of a yeast ER-PM contact site protein. VAP27-1 interacts with MTs in vivo and in vitro, and VAP27-NET3C often colocalizes with cMTs at ER-PM contact sites, suggesting that NET3C mediates cMT-ER interactions through VAP27-1 at ER-PM contact sites (Wang et al. 2014). It was hypothesized that cMT-ER contact sites function as a hub that organizes delivery and exchange of macromolecules and organelles (Hamada et al. 2012; Pena and Heinlein 2013). The precise role of the NET3C-VAP27-mediated MT-ER interaction is still unknown. The net3b net3c double mutant is gametophore lethal. VAP27-1 overexpression or knockdown causes pleiotropic phenotypes including root hair deformation. Therefore, the MF- and MT-ER interaction mediated by NET3C and VAP27-1 have essential roles in plant development (Wang et al. 2016).
Formin may connect cMTs with ER
Formins are the conserved FH2 domain-containing proteins involved in MF nucleation, and MF and MT organization. Formins from flowering plants contain the transmembrane domain (class I) or PTEN-like domain (class II) and are predicted to localized to PM. Some Arabidopsis formins indeed localize at the PM (Cvrckova 2013). An Arabidopsis class I formin, AtFH4, has MF polymerization activity and also has a MT-binding domain (Deeks et al. 2010). Transiently overexpressed AtFH4 localizes to ER membrane and induces co-alignment of ER and cMTs in tobacco epidermal cells, suggesting that AtFH4 mediates MT-ER interaction (Deeks et al. 2010). Contrary to this, however, immunostaining showed that AtFH4 localizes at the PM within the cell–cell boundary in mesophyll cells (Deeks et al. 2005). Although the precise localization of AtFH4 is still controversial, AtFH4 likely mediates cMT-membrane interaction and possibly organizes MFs beside the cMT-membrane connection.
In the past decade, numerous MAPs and membrane proteins regulating cMT-membrane interactions have been identified in plants, revealing the complexity of the cMT-membrane network and the diversity of cMT-membrane interactions (Fig. 2). In particular, the discovery of new players in cMT-PM interactions, such as CMUs and IQDs, highlights the importance of cMT-PM interactions and demonstrates that various linkers have specialized roles in regulating PM protein behavior. It is likely that these proteins are only a small part of the protein network that mediates MT-membrane interactions. Linkers that completely anchor cMTs to the PM are still missing, probably due to their redundancy and insolubility. Technical advances in high throughput genome editing, proteomic analysis, and imaging techniques such as super resolution microscopy and correlative light electron microscopy will accelerate studies of cMT-membrane interactions. Reconstructive approaches such as ROP domain reconstruction will help to study how cMT-PM complex influences the behavior of PM proteins. Many of the identified proteins are plant-unique proteins. Our understanding of cMT-membrane interactions will be facilitated by the identification of additional MAPs and membrane proteins, particularly plant-unique proteins.
This work was supported by grants from MEXT KAKENHI (grant no. 16H01247 to YO) and JSPS KAKENHI (Grant no. 16H06172 to YO).
- Akashi T, Shibaoka H (1991) Involvement of transmembrane proteins in the association of cortical microtubules with the plasma membrane in tobacco BY-2 cells. J Cell Sci 98:6Google Scholar
- Buschmann H, Dols J, Kopischke S, Pena EJ, Andrade-Navarro MA, Heinlein M, Szymanski DB, Zachgo S, Doonan JH, Lloyd CW (2015) Arabidopsis KCBP interacts with AIR9 but stays in the cortical division zone throughout mitosis via its MyTH4-FERM domain. J Cell Sci 128:2033–2046PubMedCrossRefGoogle Scholar
- Deeks MJ, Cvrckova F, Machesky LM, Mikitova V, Ketelaar T, Zarsky V, Davies B, Hussey PJ (2005) Arabidopsis group Ie formins localize to specific cell membrane domains, interact with actin-binding proteins and cause defects in cell expansion upon aberrant expression. New Phytol 168:529–540PubMedCrossRefGoogle Scholar
- Drevensek S, Goussot M, Duroc Y, Christodoulidou A, Steyaert S, Schaefer E, Duvernois E, Grandjean O, Vantard M, Bouchez D, Pastuglia M (2012) The Arabidopsis TRM1-TON1 interaction reveals a recruitment network common to plant cortical microtubule arrays and eukaryotic centrosomes. Plant Cell 24:178–191PubMedPubMedCentralCrossRefGoogle Scholar
- Hamada T, Tominaga M, Fukaya T, Nakamura M, Nakano A, Watanabe Y, Hashimoto T, Baskin TI (2012) RNA processing bodies, peroxisomes, Golgi bodies, mitochondria, and endoplasmic reticulum tubule junctions frequently pause at cortical microtubules. Plant Cell Physiol 53:699–708PubMedCrossRefGoogle Scholar
- Lipka E, Gadeyne A, Stockle D, Zimmermann S, De Jaeger G, Ehrhardt DW, Kirik V, Van Damme D, Muller S (2014) The phragmoplast-orienting kinesin-12 class proteins translate the positional information of the preprophase band to establish the cortical division zone in Arabidopsis thaliana. Plant Cell 26:2617–2632PubMedPubMedCentralCrossRefGoogle Scholar
- Martiniere A, Lavagi I, Nageswaran G, Rolfe DJ, Maneta-Peyret L, Luu DT, Botchway SW, Webb SE, Mongrand S, Maurel C, Martin-Fernandez ML, Kleine-Vehn J, Friml J, Moreau P, Runions J (2012) Cell wall constrains lateral diffusion of plant plasma-membrane proteins. Proc Natl Acad Sci USA 109:12805–12810PubMedPubMedCentralCrossRefGoogle Scholar
- Sampathkumar A, Gutierrez R, McFarlane H, Bringmann M, Lindeboom J, Emons AM, Samuels L, Ketelaar T, Ehrhardt D, Persson S (2013) Patterning and life-time of plasma membrane localized cellulose synthase is dependent on actin organization in Arabidopsis interphase cells. Plant Physiol 162:675–688PubMedPubMedCentralCrossRefGoogle Scholar
- Spinner L, Gadeyne A, Belcram K, Goussot M, Moison M, Duroc Y, Eeckhout D, De Winne N, Schaefer E, Van De Slijke E, Persiau G, Witters E, Gevaert K, De Jaeger G, Bouchez D, Van Damme D, Pastuglia M (2013) A protein phosphatase 2A complex spatially controls plant cell division. Nature Commun 4:1863CrossRefGoogle Scholar
- Zhang Y, Nikolovski N, Sorieul M, Vellosillo T, McFarlane HE, Dupree R, Kesten C, Schneider R, Driemeier C, Lathe R, Lampugnani E, Yu X, Ivakov A, Doblin MS, Mortimer JC, Brown SP, Persson S, Dupree P (2016) Golgi-localized STELLO proteins regulate the assembly and trafficking of cellulose synthase complexes in Arabidopsis. Nature Commun 7:11656CrossRefGoogle Scholar