Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Characterization of New Players in Planar Polarity
Transcript
Characterization of New Players in Planar Polarity Establishment in Arabidopsis Stefano Pietra Umeå Plant Science Centre Fysiologisk Botanik Umeå 2014 Characterization of New Players in Planar Polarity Establishment in Arabidopsis Stefano Pietra Umeå Plant Science Centre Fysiologisk Botanik Umeå 2014 This work is protected by the Swedish Copyright Legislation (Act 1960:729) ISBN: 978-91-7601-050-1 Front cover by: Stefano Pietra E-version available at http://umu.diva-portal.org/ Printed by: Servicecenter KBC, Umeå University Umeå, Sweden 2014 Table of Contents Table of Contents i Abstract iii Sammanfattning iv List of Papers v Author contributions vi Abbreviations vii Introduction 1 A common paradigm for cell polarity establishment 1 Cell polarity in unicellular organisms: fission and budding yeast 1 Planar polarity, the coordinate polar organization of a tissue layer 2 The frizzled planar polarity pathway in Drosophila 3 Planar polarity in plants: coordinate positioning of root hairs 6 An auxin gradient directs root hair polarity 6 Ethylene affects root hair polarity upstream of auxin 7 Auxin transport fine-tunes the auxin gradient and affects planar polarity 8 Effects of intracellular vesicle trafficking and GNOM on planar polarity 9 Polarly localized proteins in the root epidermis 10 The actin cytoskeleton is not only an effector of planar polarity 10 Rho-of-plant GTPases are the earliest markers of root hair planar polarity 11 Additional factors influencing planar polarity of root hairs 12 Root hair initiation and tip growth 13 Initiation and morphogenesis of leaf trichomes 15 Planar morphogenesis of pavement cells 17 Microtubules and their organization at the cell cortex 19 Organization and dynamics of the actin cytoskeleton 22 The plant cytoskeleton during cell division 25 The microtubule-associated protein CLASP 28 Patterning of the root and leaf epidermis 30 Aim 34 Results and Discussion 35 The kreuz und quer mutant 35 kreuz und quer is mutated in the SABRE gene 35 Planar polarity defects of sab mutants 36 Cell polarity defects of sab trichomes 37 The SAB protein 38 Homologs of the SAB protein 38 Subcellular localization of the SAB protein 39 SAB interaction with the CLASP gene 41 SAB and CLASP interact genetically 41 SAB also has a role in PPB orientation 41 i SAB is not involved in mitotic progression 42 Complex interaction of SAB and CLASP in the control of distinct processes 43 SAB function in the organization of cortical microtubules 44 Role of SAB in elongating epidermal cells 44 Bipolar growth of longitudinal microtubule arrays 45 Cortical microtubule organization at root hair initiation sites 45 Actin is an additional player in the planar polarity pathway 47 ACTIN7 genetically interacts with SAB 47 The actin cytoskeleton affects planar polarity of root hair positioning 48 AIP1-2 is an actin interactor in plants 49 AIP1-2 expression is restricted to hair-producing cells 50 Interplay of ethylene signalling and actin cytoskeleton in planar polarity 51 Patterned expression of AIP1-2 depends on WER function 51 SAB has a role in root epidermal patterning 52 sab mutants differentiate excessive ectopic root hairs 52 Longitudinal cell divisions are not responsible for sab patterning defects 53 Destabilized expression of fate markers in the sab root epidermis 53 SAB affects epidermal cell fate upstream of the core patterning pathway 54 SAB does not control patterning of leaf trichomes 55 SAB may act on transduction of patterning cues from the cortex 56 Conclusions and Future Perspectives 57 Acknowledgements 59 References 60 ii Abstract Coordinated polarity and differentiation of cells in the plane of a tissue layer are essential to the development of multicellular organisms. Arabidopsis thaliana root hairs and trichomes provide model systems to study the pathways that control planar polarity and cell fate specification in plants. A concentration gradient of the plant hormone auxin provides an instructive cue that coordinates polar assembly of signalling complexes at plasma membranes of root epidermal cells; however, knowledge about additional players and cytoskeletal effectors driving cell polarization prior to hair emergence remains limited. On the other hand, epidermal cell fate specification is controlled by a well-characterized gene network of transcription factors that translate positional signals and cell-to-cell communication into tissue-wide patterning. Yet, new components are continuously found to interact with the patterning pathway, shedding light on its connections with diverse developmental processes. This thesis presents the SABRE (SAB) gene as a novel player in planar polarity establishment and root epidermal patterning. SAB is a large protein with sequence similarity to proteins present in all eukaryotes and affects planar polarity as well as orientation of cell divisions and cortical microtubules. Genetic interaction with the microtubule-associated protein gene CLASP further supports involvement of SAB in microtubule arrangement, suggesting a role for this gene in cytoskeletal organisation. Strikingly, SAB also interacts genetically with ACTIN7 (ACT7), and both ACT7 and its modulator ACTIN INTERACTING PROTEIN 1-2 (AIP1-2) contribute to planar polarity of root hair positioning. Cell-file specific expression of AIP1-2 depends on the epidermal-patterning regulator WEREWOLF (WER), revealing a connection between actin organization, planar polarity and cell fate specification. Consistent with this finding, SAB also functions in patterning of the root epidermis by stabilizing cell fate acquisition upstream of the core patterning pathway. These results unveil new roles for SAB in planar polarity and epidermal patterning and suggest that organization of the microtubule and the actin cytoskeleton are important to both planar polarity establishment and cell fate specification. iii Sammanfattning Samordning av polaritet och differentiering av celler inom ett vävnadslager är avgörande för utvecklingen av multicellulära organismer. Rothår och bladhår hos Arabidopsis thaliana utgör modellsystem för att studera signalvägar som kontrollerar planpolaritet och specifikation av cellers öde hos växter. En koncentrationsgradient av växthormonet auxin ger en instruktiv signal som koordinerar polär hopsättning av signalkomplex vid plasmamembranet i rotepidermisceller; dock är kunskapen om ytterligare aktörer och hur cytoskelettets aktörer påverkar cellpolaritet innan rothår bildas begränsad. Vad gäller differentieringen av epidermala cellers öde kontrolleras dessa genom ett väl karakteriserat nätverk av transkriptionsfaktorer som överför positionssignaler och cell-till-cell kommunikation till vävnadsomfattande mönsterbildning. Fortfarande hittas dock nya komponenter som interagerar med signalvägarna för mönsterbildning, vilket ger nya insikter om dess förbindelser med diverse utvecklingsprocesser. Denna avhandling presenterar genen SABRE (SAB) som en ny aktör i etableringen av planpolaritet och mönsterbildning av rotepidermis. SAB är ett stort protein som har sekvenslikhet med proteiner som finns i alla eukaryoter och det påverkar planpolaritet, orientering av celldelning och kortikala mikrotubler. Genetisk interaktion med genen för det mikrotubuliassocierade proteinet CLASP stärker ytterligare inblandningen av SAB i organiserandet av mikrotubler och antyder att denna gen har en roll i organiserandet av cytoskelettet. Slående är att SAB även interagerar genetiskt med ACTIN7 (ACT7) och att både ACT7 och dess modulator ACTIN-INTERACTING PROTEIN1-2 (AIP1-2) bidrar till planpolaritet vid positionering av rothår. Cellfils-specifikt uttryck av AIP1-2 beror på den epidermala mönsterbildande genen WEREWOLF (WER), vilket påvisar ett samband mellan organisationen av aktin, planpolaritet och specifikationen av cellers öde. SAB fungerar även i mönsterbildning av rotens epidermis och stabiliserar förvärvet av cellöde uppströms av den centrala signalvägen för mönsterbildning. Dessa resultat visar på nya roller för SAB i planpolaritet och mönsterbildning av epidermis och indikerar att organiseringen av mikrotubler och aktin-cytoskelettet är viktiga både för etablerandet av planpolaritet och för specificeringen av cellers öde. iv List of Papers I Stefano Pietra, Anna Gustavsson, Christian Kiefer, Lothar Kalmbach, Per Hörstedt, Yoshihisa Ikeda, Anna N. Stepanova, Jose M. Alonso and Markus Grebe (2013). Arabidopsis SABRE and CLASP interact to stabilize cell division plane orientation and planar polarity. Nature Communications 4 II Christian S. Kiefer, Andrea R. Claes, Jean-Claude Nzayisenga, Stefano Pietra, Thomas Stanislas, Anke Hüser, Yoshihisa Ikeda and Markus Grebe. Arabidopsis AIP1-2 restricted by WER-mediated patterning modulates planar polarity. Submitted manuscript III Stefano Pietra*, Patricia Lang* and Markus Grebe. SABRE is required for stabilization of root hair patterning in Arabidopsis thaliana. Submitted manuscript * These authors contributed equally to this work. The papers will be referred to by their Roman numbers in the text. Publications by Stefano Pietra NOT included in this thesis: Stefano Pietra and Markus Grebe (2010). Auxin paves the way for planar morphogenesis. Cell 143, 29-31 v Author contributions I Stefano Pietra performed the fine-mapping and the map-based identification of the kuq mutation (summarized in Figure 2), characterization of the kuq mutant and other sab alleles necessary for Figure 1a-c and Figure 3a-e, RTPCRs shown in Figure 3f-h, analyses for Figure 4, quantifications for Figure 5a-e, live imaging for Figure 6c,d,r-t, Figure 7a,b,d-p, Supplementary Figure S4d-k and Supplementary Movies 1-6, recombineering work to generate the SAB-3xYpet and 3xYpet-SAB complemented sab-5 lines, live imaging for Figure 8j-l and Figure 9a-h,v,w. He additionally co-supervised Lothar Kalmbach during his Diploma thesis work that produced Figure 1h-m and Supplementary Figure S2e,f. Stefano Pietra generated all used transgenic lines expressing fluorescent markers in the sab background, contributed to analysis and statistical analysis of all experiments and wrote the manuscript together with Markus Grebe. II Stefano Pietra performed the ROP immunolocalization experiments, the imaging, the quantification and the statistical analysis required for Figure 5. He read, commented on and improved the manuscript draft. III Stefano Pietra initially observed the patterning defects in sab-mutant roots. He subsequently co-supervised and guided Patricia Lang during her master thesis work that produced the results required for all Figures, Tables 1 and 2. Stefano Pietra performed all quantitative and statistical analysis of epidermal cell patterning displayed in Table 3 including analyses of the sab5;wer-1 and sab-5;cpc-1 mutants. He generated experimental materials, contributed to analysis and interpretation of all results and wrote the manuscript together with Markus Grebe. Stefano Pietra also performed part of the experiments, imaging and quantification required for Figure 10 presented in the results and discussion section of this thesis. The remaining part was performed by Anna Gustavsson. vi Abbreviations 2,4-D +TIP ABP1 ABP ACT2 ACT7 ACT8 ADF AIP1 AIR9 AN APT1 ARF1 ARF-GEF ARP2/3 AUX1 AVG axr2-1 BFA CAP CDK CDS CESA6 CLASP CLC2 CLC3 CP CPC CtBP/BARS CTR1 Dgo DIS1 Ds Dsh EB1 EF-1α EGL3 EIN1 2,4-dichlorophenoxyacetic acid plus-end tracking protein AUXIN BINDING PROTEIN 1 actin-binding protein ACTIN2 ACTIN7 ACTIN8 ACTIN DEPOLYMERIZATION FACTOR ACTIN INTERACTING PROTEIN 1 AUXIN INDUCED IN ROOTS 9 ANGUSTIFOLIA ABERRANT POLLEN TRANSMISSION 1 ADP-RIBOSYLATION FACTOR 1 guanine exchange factor for adenosyl ribosylation factor actin-related protein 2/3 AUXIN RESISTANT 1 aminoethoxyvinylglycine auxin resistant 2-1 Brefeldin A CYCLASE ASSOCIATED PROTEIN cyclin-dependent kinase cortical division site CELLULOSE SYNTHASE 6 CLIP170-associated protein CLATHRIN LIGHT CHAIN 2 CLATHRIN LIGHT CHAIN 3 capping protein CAPRICE C-terminal binding proteins/brefeldin A ADPribosylated substrates CONSTITUTIVE TRIPLE RESPONSE 1 Diego DISTORTED 1 Dachsous Dishevelled END BINDING 1 ELONGATING FACTOR-1α ENHANCER OF GLABRA3 ETHYLENE INSENSITIVE 1 vii EIN2 ERH2 ERH3 ETC1 ETO1 ETR1 F-actin Fj Flr Fmi Frtz Ft Fy Fz G-actin GA GCP2 GCP3 GEF GL1 GL2 GL3 GST H IBA In JA JKD KCA1 KIP KTN1 kuq MAP MOR1 Mwh N NADPH oxidase P PDR8 PEN3 PIN ETHYLENE INSENSITIVE 2 ECTOPIC ROOT HAIR 2 ECTOPIC ROOT HAIR 3 ENHANCER OF TRY AND CPC 1 ETHYLENE OVERPRODUCER 1 ETHYLENE RESPONSE 1 Filamentous actin Four-jointed Flare Flamingo Fritz Fat Fuzzy Frizzled Globular actin gibberellin γ-tubulin complex protein 2 γ-tubulin complex protein 3 guanine nucleotide exchange factor GLABRA 1 GLABRA 2 GLABRA 3 Glutathione-S-Transferase hair indole-3-butyric acid Inturned jasmonic acid JACKDAW KINESIN CDKA;1-ASSOCIATED 1 KINKY POLLEN katanin kreuz und quer microtubule-associated protein MICROTUBULE ORGANIZATION 1 Multiple Wing Hairs non-hair nicotinamide adenine dinucleotide phosphate oxidase phosphorus PLEIOTROPIC DRUG RESISTANCE 8 PENETRATION 3 PIN-FORMED viii PIP5K3 Pk PLD PLDζ1 POK1 POK2 POM1 PPB PRC1 PtdIns(4)P PtdIns(4,5)P2 RanGAP1 RHD2 RHD4 RHD6 Rho-GDI ROP ROS RSH RT-PCR SAB SCAR SCM SCN1 SCZ SIM SMT1 SNX1 SPK1 SPR1 Stan Stbm TAN1 TMK TON1A TON1B TON2 TRY Tsr TTG1 Vang VLN4 PHOSPHATIDYLINOSITOL PHOSPHATE 5KINASE 3 Prickle PHOSPHOLIPASE D PHOSPHOLIPASE Dζ1 PHRAGMOPLAST ORIENTING KINESIN 1 PHRAGMOPLAST ORIENTING KINESIN 2 POM POM 1 preprophase band PROCUSTE 1 phosphatidylinositol 4-phosphate phosphatidylinositol 4,5-bisphosphate Ran GTPase ACTIVATING PROTEIN 1 ROOT HAIR DEFECTIVE 2 ROOT HAIR DEFECTIVE 4 ROOT HAIR DEFECTIVE 6 Rho-GDP dissociation inhibitor Rho-of-plants reactive oxygen species ROOT-SHOOT-HYPOCOTYL-DEFECTIVE Reverse transcription PCR SABRE suppressor of cAMP receptor defects SCRAMBLED SUPERCENTIPEDE 1 SCHIZORIZA SIAMESE STEROL METHYLTRANSFERASE 1 SORTING NEXIN 1 SPIKE 1 SPIRAL 1 Starry Night Strabismus TANGLED 1 TRANSMEMBRANE KINASE TONNEAU 1A TONNEAU 1B TONNEAU 2 TRIPTYCHON Twinstar TRANSPARENT TESTA GLABRA 1 Van Gogh VILLIN 4 ix WAVE WEI2 WEI7 WER Wg ZWI Wiskott-Aldrich Syndrome protein family verprolin-homologous protein WEAK ETHYLENE INSENSITIVE 2 WEAK ETHYLENE INSENSITIVE 7 WEREWOLF Wingless ZWICHEL x Introduction A common paradigm for cell polarity establishment Cell polarity can be defined as an asymmetry in the shape, organization, and distribution of structures within a cell (Cove, 2000). Polarized cell growth is essential for the development of both unicellular and multicellular organisms (Nelson, 2003). In simple single-cell organisms, like bacteria and yeast, cell polarity is fundamental for morphogenesis, proliferation, and ultimately survival. Similarly, multicellular organisms that are unable to set up a correct polarity face early death or severe perturbation in their development: their cells need to polarize in order to acquire a specific fate or morphology and carry out specialized functions (Yang, 2008). Although some systems are able to establish their polarity in an autonomous way (Cove, 2000), in most cases polarization requires a specific internal (biological) or external (environmental) signal to break cell symmetry and initiate asymmetrical organization and dynamics of cellular components. Despite the great diversity among cell types that display polar growth, a common set of mechanisms is used by evolutionarily distant organisms, and it is possible to outline a paradigm that can be applied to all eukaryotic systems (Nelson, 2003; Yang, 2008): in response to an initial cue, a polarity landmark is established on the cell surface and determines the localized aggregation of signalling complexes of Rho-family small GTPases (Perez and Rincón, 2010; Nelson, 2003; Yang, 2008). These central signalling components initiate cascades that activate the mobilization of proteins from intracellular pools and direct the recruitment and oriented assembly of cytoskeletal structures, which in turn lead to targeted vesicle secretion. Polarized protein transport in combination with endocytic recycling generates a positive feedback loop able to reinforce the initial signal and sustain polar growth (Geldner, 2009; Grebe, 2004). The simplest examples of how polarity is generated and translated into polar growth are unicellular organisms. Studies from the model systems fission and budding yeast provide excellent platforms to understand and explore similar conserved mechanisms that are employed by more complex multicellular organisms. Cell polarity in unicellular organisms: fission and budding yeast Cells of the fission yeast Schizosaccharomyces pombe have a cylindrical rod shape and grow by polarized tip extension until they enter mitosis and divide in two shorter symmetrical cells (Piel and Tran, 2009). Microtubules play an essential role in directing growth at the cell tips: they are organized in 1 bundles parallel to the long axis of the cell and are uniformly polarized with their plus ends growing away from the geometrical cell centre (Nelson, 2003). Dynamic microtubules deliver to cell ends the microtubule plus end proteins Tea1p and Tea4p, essential for marking the growth site (Mata and Nurse, 1997; Martin et al., 2005). Tea1p is anchored to the cell cortex through the membrane-associated protein Mod5p (Snaith and Sawin, 2003) and the complex then recruits the polarisome complex, which includes the actin-binding protein Bud6p and the formin For3p (Martin et al., 2005). The Rho GTPase Cdc42p activates For3p (Martin et al., 2007), which nucleates actin cables thus establishing tracks for myosin-driven vesicular transport of materials for cell growth and of additional polarity determinants (Piel and Tran, 2009). The imbalance between this polarized delivery and their rate of dispersion allows accumulation of the polarity factors, generating a feedback loop that reinforces the initial polarity cue (Piel and Tran, 2009). The budding yeast Saccharomyces cerevisiae has cells with round shape that divide by formation on their surface of a single bud, which grows polarly and then separates at maturation. Positional information from the former budding cycle of the yeast cell establishes a landmark (bud scar), determining anisotropic growth of the plasma membrane. Also in this case, specific localization and activation of Cdc42p is crucial for oriented assembly of the actin cytoskeleton and targeting of vesicles to the bud site (Nelson, 2003; Perez and Rincón, 2010). Initial stochastic increases in Cdc42p clustering are reinforced by actin filaments through their effects on targeted secretion and endocytosis, in a positive feedback loop that counteracts lateral diffusion and stabilizes growth polarity (Marco et al., 2007; Valdez-Taubas and Pelham, 2003). Hence, these are perfect examples from unicellular organisms of how an initial polarity cue, mediated by the microtubule cytoskeleton in one case and by cell-membrane landmarks in the other, can induce polar activation of Rho GTPase molecular switches and in turn (re)organization of actin filaments, driving targeted secretion of more polarity determinants to the growth site. The cytoskeleton and Rho-type GTPases are central actors in polarization of all eukaryotic systems, and it is therefore fascinating to compare the mechanisms regulating such an essential process in very distantly related species. Planar polarity, the coordinate polar organization of a tissue layer Setting up and regulation of polarity can gain additional levels of complexity in multicellular systems. Polar growth of individual cells often needs to be coordinated in the context of tissues and of the whole organism. This implies that each cell can be polarized along several axes, and needs mechanisms to distinguish multiple membrane domains and organize them differently 2 (Nelson, 2003; Geldner, 2009). In addition, polarity can be coordinated among cells within a single tissue layer, determining alignment of the whole plane to the same polarity cues. This process is referred to as planar polarity (Nübler-Jung et al., 1987). Planar polarity is most obvious in animal epithelia, where structures like scales, feathers and hairs offer a very convenient read-out to analyze the mechanisms that regulate their orientation. One of the most commonly used models for research on planar polarity is hair orientation on the wings of the fruitfly Drosophila melanogaster (Adler, 2002). Genetic studies in this system have accumulated a large body of knowledge over the past decades, revealing many of the molecular components that control planar polarity establishment in animals (Zallen, 2007). The frizzled planar polarity pathway in Drosophila Cells of the Drosophila wing epithelium each produce a single hair that emerges at the distal end of the cell and points towards the tip of the wing (Wong and Adler, 1993). Initial studies identified a set of genes that form the core planar polarity pathway. The name reflects that mutations in any of the components of the pathway lead to non-polar cells, with hairs initiating in their centre (Goodrich and Strutt, 2011). These components control planar polarity of wing hairs through asymmetrical subcellular localization in both cell-autonomous and non-autonomous ways (Adler, 2002). A key member of this pathway is the frizzled (fz) gene, encoding a sevenpass Wnt transmembrane receptor and lending its name to the whole gene network, which is commonly termed “the frizzled pathway” (Vinson and Adler, 1987; Adler, 2002). From early stages of wing development, Fz localizes distally on apical cell surfaces and recruits the cytoplasmic proteins Dishevelled (Dsh), required for signal transduction (Strutt, 2009), and Diego (Dgo). At the same time, the four-pass transmembrane protein Strabismus (Stbm, also known as Van Gogh; Vang) and the cytosolic protein Prickle (Pk) associate and accumulate specifically at proximal cell ends (Fig. 1; Zallen, 2007). Asymmetrical protein distribution can be achieved through intracellular antagonistic interactions, which stabilize the formation of mutually exclusive complexes at the two ends of a cell, and intercellular contacts between distal and proximal complexes in adjacent cells (Tree et al., 2002). These latter are mediated by the formation of homodimers of the transmembrane cadherin Flamingo (Fmi, also known as Starry Night; Stan), which localizes to both poles of the cells and allows heterologous interaction between Fz and Stbm (Fig. 1; Strutt and Strutt, 2009). Cell-cell communication establishes a feedback loop that reinforces the asymmetrical subcellular localization of core proteins in neighbouring cells and is likely to coordinate intercellular polarity, as demonstrated by the fact that mutations 3 Figure 1. The Drosophila frizzled pathway. Core planar polarity proteins localize asymmetrically across the junction between two wing epidermal cells, forming a proximal and a distal complex that interact to coordinate planar polarity of wing hairs. Frizzled (Fz), Dishevelled (Dsh) and Diego (Dgo) localize to the distal end of cells, whereas Strabismus (Stbm) and Prickle (Pk) accumulate at proximal ends and Flamingo (Fmi) forms homodimers over the cell junction. Image adapted from Strutt and Strutt (2009) with permission of the publisher. in fz and stbm have non-cell autonomous effects on surrounding cells (Adler, 2002). Asymmetrical localization of the core proteins of the planar polarity pathway is dependent on each other’s presence, in a system capable of selforganizing polarity locally (Strutt and Strutt, 2005). However, the nature of the global signal that initiates polarization of the core module and aligns it across an entire tissue has been the object of a long debate (Zallen, 2007). One candidate for such a role has for several years been the Fat/Dachsous system (Strutt and Strutt, 2005). Fat (Ft) and Dachsous (Ds) are both large atypical cadherins that localize to the apical ends of epithelial cells and interact heterophilically across cell boundaries (Thomas and Strutt, 2012). Their activity and accumulation are affected by a Golgi kinase, Four-jointed (Fj), which is expressed in a gradient along the wing and modulates their binding, generating a bias in the orientation of the heterodimers and thus a directional cue (Matis and Axelrod, 2013). The findings that loss of Ft, Ds, or Fj activity caused the core proteins to lose their ability to coordinate polarity at long distance suggested that they could function as the factor dictating global polarity upstream of the core pathway (Yang et al., 2002). Further studies however revealed discrepancies within this linear three-tiered model and nowadays the relationship between the Ft/Ds and core systems is controversial. Currently, the Ft/Ds module is thought to most likely act in parallel to or downstream of another global signal (Strutt, 2009; Goodrich and Strutt, 2011; Matis and Axelrod, 2013). From the very beginning of studies on planar polarity, the presence of a gradient of Fz activity along the wing has been an attractive candidate for the role of global signal imposing and maintaining directional polarity. Because fz expression is homogeneous though, the model required the presence of a morphogen whose concentration gradient could modulate Fz activation over the whole tissue layer (Lawrence, 1966). Taking into account that in several 4 signalling pathways proteins of the Frizzled family acted as receptors for secreted Wnt ligands and that Dsh was already part of a signalling cascade in a different Wnt pathway (Thomas and Strutt, 2012), it was easy to presume that the long-range signal providing a polarizing cue could be a Wnt ligand (Adler et al., 1997). However, despite several Wnts being implicated in planar polarity signalling in vertebrates and Wnt gradients being present in other polarized tissues, evidence of a Wnt ligand controlling planar polarity of Drosophila wing hairs has been lacking for many years (Goodrich and Strutt, 2011). Only recently the two redundant Wnt proteins Wingless (Wg) and dWnt4 have been shown to modulate local intercellular interactions between Fz and Stbm. Through their graded distribution Wg and dWnt4 generate an aligned gradient of Fz activity providing the long-sought global cue that orients planar cell polarity in the wing (Wu et al., 2013). In addition to its role in coordinating tissue polarity, Fz also functions in the initiation and placement of wing hairs (Adler, 2002). This cellautonomous activity relies on signalling through tissue-specific effectors acting downstream of the core pathway and specifying the site of hair growth. Three effector proteins, Inturned (In), Fritz (Frtz) and Fuzzy (Fy) localize proximally in wing cells in a Stbm-dependent manner and direct accumulation as well as activation of the formin-homology domain protein Multiple Wing Hairs (Mwh) to proximal cell edges (Strutt and Warrington, 2008). Mwh functions as a repressor of actin filament formation and restricts actin polymerization and hair initiation to the opposite, distal region of the cell. Mutations in any of these effectors lead to multiple hairs emerging in various locations around the cell periphery (Strutt and Strutt, 2009). The importance of actin dynamics in wing hair initiation is confirmed by the role of Rho family GTPases, modulators of the actin cytoskeleton, downstream of Dsh: mutations in RhoA or its effector Drok, which acts on the myosin II regulatory light chain in the wing, determine the formation of multiple hairs pointing distally (Strutt et al., 1997; Winter et al., 2001), and expression of a dominant negative version of Rac1 gives rise to duplicated or triplicated wing hairs (Eaton et al., 1996). Furthermore, actin dynamics are not only an output of the planar polarity pathway, limiting the number and position of hair outgrowths, but are also an essential component required for the pathway to function (Ren et al., 2007). In fact, mutants of either of the two actin severing factors twinstar (tsr) and flare (flr), which encode the Drosophila homologs of Cofilin/Actin Depolymerization Factor (ADF) and Actin Interacting Protein 1 (AIP1), display hair polarity defects and aberrant Fmi localization (Blair et al., 2006; Ren et al., 2007). These findings strongly suggests that actin remodelling has a role in establishing the polarity of core components of the pathway, possibly affecting their intracellular transport or localized stabilization (Blair et al., 2006). 5 Planar polarity has been studied for the longest time and in greatest depth through analyses of the growth of hairs on the wing epithelium (Adler, 2002); however, the frizzled pathway indeed also has a role in establishing planar polarity of other Drosophila structures, like bristles and eye ommatidia (Goodrich and Strutt, 2011). Orthologs of the frizzled core and global components are present in many other organisms, including vertebrates, and regulate diverse processes from convergent extension during gastrulation to patterning of hairs in mammals (Simons and Mlodzik, 2008). Strikingly, however, no homologs of fz or any other planar polarity core protein are present in plants. Planar polarity in plants: coordinate positioning of root hairs Plant species certainly also possess an array of polarized tissues, and must thus have evolved different mechanisms to coordinate their polarity. One of the systems offering the clearest read-out of polarization over the plane of a whole tissue in plants is the root epidermis with its pattern of outgrowing hairs (Grebe, 2004). Root hairs are unicellular structures that distantly resemble the Drosophila wing hairs in the fact that they emerge distally in hair-bearing cells (trichoblasts), at a conserved position close to the root tip (Schiefelbein and Somerville, 1990). The uniform polarity of these structures has been already observed in the middle of last century (Sinnott and Bloch, 1939). When powerful genetic tools in the model system Arabidopsis thaliana became available, researches started to dissect the mechanisms that regulate polarization of root hair placement (Masucci and Schiefelbein, 1994). However, studies specifically focusing on how planar polarity of hair positioning is established and maintained in plants have been conducted only throughout the last decade (Grebe, 2004; Nakamura et al., 2012). An auxin gradient directs root hair polarity In Arabidopsis the plant hormone auxin has a central role in the control of hair polarity in the root (Fig. 2; Masucci and Schiefelbein, 1994; Sabatini et al., 1999; Grebe, 2004). An auxin gradient with its maximum in the root tip provides vectorial information for the orientation of root hairs (Fig. 3; Fischer et al., 2006), and modulating this gradient or its perception can affect epidermal cell polarity (Masucci and Schiefelbein, 1994; Grebe, 2004; Fischer et al., 2006). Exogenous auxin supply or auxin overproduction induce hair initiation at a more basal (rootward) position within trichoblasts, while mutants affecting the response of cells to auxin, as the auxin resistant 2-1 (axr2-1) mutant, display an apical (shootward) shift of hair positioning (Masucci and Schiefelbein, 1994). In addition, application of a membranepermeable auxin directly to the root epidermis of a mutant lacking an 6 endogenous auxin gradient is able to orient root hair growth towards the auxin source, demonstrating that a local auxin concentration gradient is sufficient to direct root hair polarization (Fischer et al., 2006). Figure 2. Genetic framework of planar polarity in the Arabidopsis root. Ethylene signalling through CONSTITUTIVE TRIPLE RESPONSE 1 (CTR1) and ETHYLENE INSENSITIVE 2 (EIN2) affects local auxin biosynthesis by WEAK ETHYLENE INSENSITIVE 2 (WEI2) and WEAK ETHYLENE INSENSITIVE 7 (WEI7). Auxin is transported polarly by the influx carrier AUXIN RESISTANT 1 (AUX1) and the efflux carrier PIN-FORMED 2 (PIN2) and forms a root-tip oriented gradient that guides positioning of Rho-of-plant (ROP) small GTPases. ROP positioning is further affected by the GNOM guanine exchange factor, which also acts upstream of PIN2. ROP accumulation directs polar initiation of root hairs. Diagram based on data from Fischer et al. (2006) and Ikeda et al. (2009). Ethylene affects root hair polarity upstream of auxin The hormone ethylene is known to interact with auxin in a multitude of processes (Muday et al., 2012), and also has a role in planar polarity of root hair positioning (Fig. 2; Masucci and Schiefelbein, 1994; Fischer et al., 2006; Ikeda et al., 2009). Exogenous application of the ethylene precursor 1Aminocyclopropane-1-carboxylic acid (ACC) polarizes hair initiation towards the basal cell ends, and a similar hyperpolarization is determined by increases in endogenous ethylene biosynthesis in mutants like the dominant ethylene overproducer 1 (eto1; Masucci and Schiefelbein, 1994). On the contrary, inhibiting ethylene production with the chemical 7 aminoethoxyvinylglycine (AVG) or ethylene perception with mutations in ETHYLENE RESPONSE 1/ETHYLENE INSENSITIVE 1 (ETR1/EIN1) or ETHYLENE INSENSITIVE 2 (EIN2) induces an apical shift in root hair positioning (Masucci and Schiefelbein, 1994; Fischer et al., 2006). Ethylene signalling acts upstream of auxin biosynthesis in the planar polarity pathway, as demonstrated by the inhibitory effect of the negative regulator of ethylene signalling CONSTITUTIVE TRIPLE RESPONSE 1 (CTR1) on the local expression of the auxin biosynthesis genes WEAK ETHYLENE INSENSITIVE 2 (WEI2) and WEAK ETHYLENE INSENSITIVE 7 (WEI7) in the root tip (Ikeda et al., 2009). Mutations in CTR1 increase auxin biosynthesis and consequently the auxin gradient in the root, resulting in a basal shift of root hair positioning (Ikeda et al., 2009). Figure 3. Polar root hair positioning directed by an auxin gradient. Root hairs initiate close to the basal end of epidermal cells. Arrows indicate trichoblast polarization towards the auxin concentration maximum in the root tip. Image adapted from Grebe (2004) and Fischer et al. (2006) with permission of the publishers. Auxin transport fine-tunes the auxin gradient and affects planar polarity Generation and maintenance of the root tip-oriented auxin gradient does not, however, only depend on auxin biosynthesis in shoot or root, but also strongly relies on redistribution of the hormone via membrane-localized specialized transporters (Kramer and Bennett, 2006). These include among others the auxin influx carrier AUXIN RESISTANT 1 (AUX1; Bennett et al., 1996) and auxin efflux carriers of the PIN-FORMED (PIN) family (Petrášek et al., 2006), responsible for the acropetal transport (downward through the 8 root centre toward the tip) and basipetal recycling (upward through the epidermal layer) of auxin in the root (Kramer and Bennett, 2006). Polar auxin transport is indeed essential for the fine-tuning of the auxinresponse gradient in the root apex (Blilou et al., 2005) and a crucial component of the non-cell autonomous effect of auxin on root hair planar polarity (Fig. 2; Grebe et al., 2002; Ikeda et al., 2009). Loss-of-function aux1 mutations induce hair initiation at a more apical position and insensitivity to the effects of the auxin influx carrier substrate 2,4-dichlorophenoxyacetic acid (2,4-D) on root hair polarity (Grebe et al., 2002). By contrast, single mutants of the genes encoding the PIN1 and PIN2 auxin efflux carriers do not show any planar polarity defects, and only the quadruple pin1;pin2;pin3;pin7 mutant displays a weak but significant apical shift, suggesting that polar hair positioning does not rely strongly on the activity of PIN proteins (Grebe et al., 2002; Fischer et al., 2006). However, mutation of PIN2 partially suppresses hair hyperpolarization of ctr1 mutants and enhances the apical shift of ein2 mutants, revealing a subtle contribution of PIN2 to planar polarity establishment (Ikeda et al., 2009). Effects of intracellular vesicle trafficking and GNOM on planar polarity Polar membrane localization of auxin transporters is based on polar targeting, endocytic recycling, and transcytosis (Kleine-Vehn and Friml, 2008). AUX1 localizes to the apical membrane of protophloem cells and to apical and basal ends of epidermal cells (Swarup et al., 2001), PIN1 localizes basally in vascular cells (Gälweiler et al., 1998), and PIN2 accumulates on the apical membrane of epidermal cells and elongating cortical cells (Müller et al., 1998), while it is situated on the basal end of meristematic cortical cells (Blilou et al., 2005). Delivery of auxin carriers to correct membrane domains is dependent on a subclass of guanine exchange factors for adenosyl ribosylation factor small G proteins (ARF-GEFs) that mediate intracellular vesicle trafficking (Geldner et al., 2003). The fungal toxin Brefeldin A (BFA) specifically inhibits the ARF-GEFs responsible for basal targeting of auxin transporters (Steinmann et al., 1999; Kleine-Vehn et al., 2008), abolishing for instance membrane localization of PIN1 and AUX1 (Steinmann et al., 1999; Grebe et al., 2002). BFA treatment thus induces a defect in auxin distribution and polar localization of root hairs, which emerge at a more apical position (Grebe et al., 2002). One of the direct targets of BFA is the ARF-GEF GNOM, which mediates basal membrane localization of PIN1 and PIN2 but not AUX1 (Steinmann et al., 1999; Geldner et al., 2003; Fischer et al., 2006). Hypomorphic gnom mutants display an apical shift in root hair initiation, and combining a hypomorphic gnom transheterozygote mutant combination with mutations 9 in AUX1 and EIN2 reveals synergistic activity of the three genes in modulating the auxin gradient and polar hair positioning in the root (Fischer et al., 2006). GNOM acts upstream of PIN2, as confirmed by genetic analyses and perturbed basal PIN2 localization in the cortex of weak gnom mutants (Kleine-Vehn et al., 2008; Ikeda et al., 2009). However, its function in planar root hair polarity is likely not limited to an effect on PIN proteins, because the defects in hypomorphic gnom mutants are much stronger than those detected in knock-outs of multiple PIN genes (Fig. 2; Fischer et al., 2006). Polarly localized proteins in the root epidermis Auxin transporters are not the only proteins that localize polarly in the Arabidopsis root epidermis. Several proteins are subject to specific trafficking that allows them to accumulate at distinct membrane domains in epidermal cells. An example is the PENETRATION 3/PLEIOTROPIC DRUG RESISTANCE 8 (PEN3/PDR8) ATP binding cassette transporter, which is involved in defence from pathogens and efflux of the auxin precursor indole3-butyric acid (IBA), and localizes to outer membranes at the root-soil interface (Stein et al., 2006; Strader and Bartel, 2009; Łangowski et al., 2010). On the opposite side of the cells, the boric acid/borate exporter BOR1 localizes polarly to the inner lateral membrane domain and translocates boron from the epidermis toward the xylem in a tightly controlled manner (Takano et al., 2010). Finally, the AGC protein kinase D6PK, which controls efficient auxin transport and is able to phosphorylate PIN proteins, localizes to the basal membrane of several meristematic root cell types, including epidermal cells (Zourelidou et al., 2009). At the moment however, none of these polarly localized proteins has been characterized for having any effect on establishment or regulation of root hair planar polarity. The actin cytoskeleton is not only an effector of planar polarity Despite fundamental differences in the global polarizing cues as well as the core genes that determine planar polarity of hairs between Arabidopsis and Drosophila, the final effectors that control hair assembly and initiation at polar sites appear surprisingly similar. The actin cytoskeleton is the main candidate for initiating the final morphological events that lead to emergence of root hairs, as filamentous actin (F-actin) accumulates in root hair bulges just after hair initiation (Baluška et al., 2000), although data on early stages have not been published yet (Grebe, 2004). Furthermore, although no detailed experiments have focused on placing actin within the planar polarity pathway, mutations in the ACTIN2 (ACT2) gene have been shown to affect not only root hair growth 10 but also positioning of the hair initiation site, leading to slightly apicallyshifted hairs in addition to causing the occasional formation of multiple hairs per trichoblast (Ringli et al., 2002). Rho-of-plant GTPases are the earliest markers of root hair planar polarity Actin organization is regulated in other systems by small GTPases of the Rho family, and indeed in root hair initiation as well members of the Rho-ofplants (ROP) family seem to play a central role in coordinating actin at sites of hair formation. ROPs relocalize polarly to hair initiation sites even prior to any visible sign of hair outgrowth (Molendijk et al., 2001; Jones et al., 2002), and ROP2 overexpression induces additional and often misplaced hairs on single cells (Jones et al., 2002). Loss-of-function evidence to complement these data is missing, but knock-out of the gene encoding the Rho-GDP dissociation inhibitor (RhoGDI) SUPERCENTIPEDE 1 (SCN1) that limits ROP2 activation also leads to multiple sites of hair growth on individual trichoblasts (Carol et al., 2005), functionally supporting a role for ROPs in hair initiation site selection. ROP patches are the earliest markers of future sites of hair formation, and their positioning is dependent on the combined action of AUX1, of ethylene signalling and of GNOM on the generation of an auxin gradient in the root tip (Fig. 2; Fischer et al., 2006; Ikeda et al., 2009). Furthermore, mathematical modelling studies of auxin-dependent ROP activation support this view: given a specific cell length, ROP levels and active versus inactive ROP diffusion rates, an intracellular gradient of auxin concentration is sufficient to spontaneously drive polar placement of ROP patches (Payne and Grierson, 2009). In addition, auxin exogenously applied to leaf protoplasts can activate ROP2 and ROP6 through AUXIN BINDING PROTEIN 1 (ABP1) and members of the TRANSMEMBRANE KINASE (TMK) family of kinases, leading to the reorganization of the actin and microtubule cytoskeleton (Xu et al., 2010; Xu et al., 2014). This shows that auxin can indeed positively regulate ROPs and their downstream effectors. Whether the same mechanism exists in root cells and how it can generate a polarity cue from a concentration gradient remains to be investigated. Polar localization of ROP2 to hair initiation sites also depends on function of ADP-RIBOSYLATION FACTOR 1 (ARF1) family members, central components of the intracellular vesicle trafficking machinery whose activity can be inhibited by BFA (Molendijk et al., 2001; Xu and Scheres, 2005). This suggests another possible regulatory mechanism upstream of ROPs, but the exact functional link between the polarizing signal and potential downstream effectors of root hair polarity remains to be found. 11 Additional factors influencing planar polarity of root hairs Additional processes and genes affect planar polarity of the root epidermis, although it is not clear yet what their role is and at which stage they come into play. For instance, clathrin-mediated endocytosis, a mechanism controlling membrane internalization and protein recycling from the plasma membrane and the trans-Golgi network, has been proposed to be important for polar positioning of root hairs. Double mutants of CLATHRIN LIGHT CHAIN 2 (CLC2) and CLATHRIN LIGHT CHAIN 3 (CLC3) initiate their hairs at a more apical position in trichoblasts (Wang et al., 2013), and this defect has been suggested to depend on aberrant auxin transport or signalling (Wang et al., 2013). However, no direct evidence that the phenotype is caused by altered auxin distribution is currently available, and endocytosis may as well have non-auxin-related effects on planar polarity. Furthermore, an important factor in modulating differential endocytosis of polarly distributed proteins is membrane sterol composition (Men et al., 2008). The sterol biosynthesis mutant sterol methyltransferase 1 (smt1) has defects in cell polarity and polar auxin transport, and in addition displays more randomized positioning of hairs in the root epidermis, suggesting that sterols do as well have a role in planar polarity establishment (Willemsen et al., 2003). This hypothesis is supported by the detection of sterols accumulating to sites of root hair formation at very early stages of bulge initiation, already before any structural sign of outgrowth is visible (Ovecka et al., 2010). It is thus tempting to speculate that sterols are involved in selection of the bulging site. One of the first steps required for bulge formation in root hair cells is local loosening of the cell wall. Altering the cell wall structure by mutating a gene responsible for cellulose biosynthesis, as in the procuste 1 (prc1) mutant of CELLULOSE SYNTHASE 6 (CESA6), can affect planar polarity of ROP and root hair positioning (Singh et al., 2008). In prc1 mutants, hairs emerge from both more basal and more apical positions along the trichoblasts, and a similar defect is observed in localization of the ROP patches that mark the future sites of hair initiation. It could be that cell wall composition influences selection of the polar site for hair outgrowth, but the molecular mechanisms that connect this contribution to the planar polarity pathway are currently elusive. PRC1 affects hair formation downstream of a transcription factor acting very early in root hair initiation, ROOT HAIR DEFECTIVE 6 (RHD6; Masucci and Schiefelbein, 1994; Singh et al., 2008). Mutants defective in RHD6 display a reduced number of root hairs and were also the first Arabidopsis mutants characterized for having defects in root hair polar positioning: trichoblasts of rhd6 roots produce hairs from a more apical 12 position compared to wild type (Masucci and Schiefelbein, 1994). Auxin and ACC application can rescue the rhd6 mutant phenotypes, indicating that RHD6 function is most likely mediated by auxin and ethylene. However, the exact role of RHD6 in planar polarity establishment is still unknown. Root hair initiation and tip growth Root hairs are a system of choice for studying planar polarity in plants. They are long tubular projections that grow on the outer surface of epidermal cells, and represent the main point of contact between roots and soil. Their various roles include water and nutrient uptake, anchorage, and interaction with microorganisms (Schiefelbein, 2000; Dolan, 2001; Grierson et al., 2001). Specification of root hairs takes place in the meristem based on positional information from the underlying cortex, and epidermal cell fate is acquired already during embryonic root development, although it is still possible for it to change postembryonically (Berger et al., 1998a). Cells destined to produce a hair have a denser cytoplasm and smaller vacuole compared to non-hair cells, have a higher division rate but elongate more slowly than non-hair cells (Dolan et al., 1994; Berger et al., 1998b). Initiation of hairs in elongating cells is under the control of signalling from ethylene and depends on auxin, whose transport from the root tip to differentiating hair cells is facilitated by specific expression of the AUX1 auxin influx transporter in non-hair cells (Jones et al., 2009). Several mutants related to ethylene or auxin perception show that hairs can be initiated on cells with cytological characteristics of non-hair cells, and vice versa. Regulation through ethylene and auxin may allow plants to modulate root hair density in response to environmental stimuli and thus override the positional cues (Schiefelbein, 2000). The first morphological sign of hair outgrowth is the formation of a bulge at a specific site on the lateral wall of an almost fully elongated epidermal cell. Size and number of swellings are under tight genetic control (Grierson et al., 2001), and bulge outgrowth results from an initial acidification of the cell wall (Bibikova et al., 1998), which locally activates wall-loosening proteins allowing directional turgor-driven deformation of the cell surface (Rounds and Bezanilla, 2013). Within a growing hair bulge, rearrangement of the cytoskeleton and polarization of the cytoplasm prepare the cell for tip-oriented hair elongation. Dense and dynamic F-actin meshworks assemble just below the growing tip (Baluška et al., 2000) and long actin bundles at the cell cortex and in the central region of the hair orient axially and are essential to drive hair growth and reverse-fountain cytoplasmic streaming (Rounds and Bezanilla, 2013). Accordingly, mutation of ACT2 (Ringli et al., 2002) and 13 mutation or inhibition of genes encoding proteins involved in actin filament initiation like DISTORTED 1 (DIS1; Mathur et al., 2003a) and CROOKED (Mathur et al., 2003b), in actin turnover like ACTIN INTERACTING PROTEIN 1-1 (AIP1-1) and ACTIN INTERACTING PROTEIN 1-2 (AIP1-2; Ketelaar et al., 2004), or in actin bundling like VILLIN 4 (VLN4; Zhang et al., 2011) affect tip growth, leading to shorter root hairs. Cortical microtubule arrays, which are transversal to the root axis in elongating trichoblasts, reorganize at the onset of hair outgrowth (Van Bruaene et al., 2004) and align to the root-hair shank to control growth direction and limit the number of growth sites (Bibikova et al., 1999). Shortly after bulge formation, the root hair starts elongating by tip growth and the nucleus is driven into the hair by actin-regulated processes (Ketelaar et al., 2002). Soon after tip growth initiated, a Ca2+ gradient focused towards the apical end of the hair is established (Wymer et al., 1997), and reactive oxygen species (ROS) accumulate close to the growing region (Foreman et al., 2003). Calcium ions and ROS are part of a positive feedback loop including the ROOT HAIR DEFECTIVE 2 (RHD2) nicotinamide adenine dinucleotide phosphate oxidase (NADPH oxidase), which localizes to growing root hair tips in an actin-dependent manner (Jones et al., 2007; Takeda et al., 2008) and there produces ROS. These molecules activate hyperpolarizationactivated Ca2+ channels facilitating influx of Ca2+ (Foreman et al., 2003), which in turn induces RHD2 activity (Takeda et al., 2008) and ensures a robust mechanism for maintaining polar hair elongation. While elongating, root hairs oscillate between periods of fast and slow growth that are tightly correlated with oscillations in extracellular pH, extracellular ROS accumulation, and cytosolic Ca2+ concentration (Monshausen et al., 2007; Monshausen et al., 2008). Cyclic increases in pH and ROS locally stiffen the cell wall, limiting tip growth and contributing to restricting it spatially, while at the same time affecting organization of the actin cytoskeleton (Ishida et al., 2008, Cárdenas, 2009). Ca2+ as well plays a crucial role in fine-tuning actin dynamics, and in addition is implicated in regulating exocytosis near the hair apex (Monshausen et al., 2008). During root hair growth a large number of vesicles form close to the hair tip in what is referred to as the clear zone, a region rich of endoplasmic reticulum and Golgi bodies (Rounds and Bezanilla, 2013). Vesicles are polarly secreted and fuse with the plasma membrane, delivering new cell wall material to the growing tip in a process strictly dependent on a functional actin cytoskeleton (Dolan, 2001; Grierson et al., 2001). Phospholipids, and in particular phosphatidylinositols, also contribute to the control of root hair elongation, acting as second messengers in cytoskeletal reorganization and membrane trafficking: PHOSPHATIDYLINOSITOL PHOSPHATE 5-KINASE 3 (PIP5K3), an enzyme involved in the synthesis of phosphatidylinositol 4,5-bisphosphate 14 [PtdIns(4,5)P2], localizes to sites of root hair initiation even prior to bulge formation, and pip5k3 mutants do not accumulate PtdIns(4,5)P2 at root hair growing tips and display much shorter root hairs (Kusano et al., 2008). Furthermore, the phosphatidylinositol 4-phosphate phosphatase ROOT HAIR DEFECTIVE 4 (RHD4) regulates polarized secretion at growing root hair tips, and rhd4 mutants accumulate phosphatidylinositol 4-phosphate [PtdIns(4)P] and have short and bulged root hairs (Thole et al., 2008). Wild-type root hairs rarely branch, but several conditions including interference with the microtubule cytoskeleton (Bibikova et al., 1999) and mutations of the Rho-GDI gene SCN1 (Parker et al., 2000) have been shown to strongly induce hair branching. Hair elongation proceeds until the hair is about 1 mm long and is promoted by auxin and ethylene signalling, as demonstrated by the long hairs of seedlings treated with 2,4-D or ACC and shorter hairs of mutants impaired in auxin or ethylene signal transduction (Pitts et al., 1998). Initiation and morphogenesis of leaf trichomes Leaf trichomes represent a fine example of cells that coordinate their polarity within the plane of a tissue layer: their branches orient along the leaf proximo-distal axis over the whole leaf surface, offering a clear polar readout (Fig. 4; Hülskamp et al., 1994). These characteristics, and the possibility for mutations to affect their coordinated orientation (Kirik et al., 2007), make trichomes good potential models for studying planar polarity in the leaf epidermis. Figure 4. Trichome organization on the leaf epidermis. Trichomes emerge at regular intervals from the leaf surface and undergo two branching events during their maturation, orienting their proximal side stem to the leaf proximo-distal axis. Images adapted from Yang (2008) and Hülskamp et al. (1994) with permission of the publishers. 15 In Arabidopsis, trichomes are large unicellular structures that differentiate on the epidermis of several organs, including leaves, and their functions are related to protection from insect herbivores, excessive transpiration, and UV light (Hülskamp, 2004). Leaf trichome development initiates near the distal end of young leaves from single protodermal cells spaced at regular intervals of three to four cells from each other (Hülskamp, 2004). Specification of the trichome fate is carried out through a complex activator-inhibitor mechanism that by intercellular communication is able to reinforce stochastic fluctuations of trichome promoting and suppressing factors thus establishing de novo patterning (Pesch and Hülskamp, 2004; Schellmann and Hülskamp, 2005). Hormone signalling also affects trichome density, as application of gibberellin (GA) and jasmonic acid (JA) stimulates trichome formation, while salicylic acid decreases their number (An et al., 2013). Once trichomes are initiated they undergo four rounds of endoreduplication, resulting in cells with increased size and 16 times the DNA content of a diploid cell. The switch from mitotic division to endoreduplication is controlled by the cyclin-dependent kinase (CDK) inhibitor SIAMESE (SIM; Churchman et al., 2006), whose loss-of-function induces multicellular trichomes. The number of endoreduplication cycles is tightly regulated by several mechanisms, including protein degradation, hormone signalling, DNA decatenation and cell-death regulation (Schellmann and Hülskamp, 2005). Defects in trichome ploidy have profound effects on trichome morphology, as the number of branches positively correlates with the cellular DNA content (Hülskamp, 2004). Surprisingly, recent studies revealed that endoreduplication also plays a role in maintaining trichome cell identity, thus confirming the importance of a precise control of the cell cycle (Bramsiepe et al., 2010). Enlarged trichome cells with an increased DNA content start growing out of the leaf surface and undergo two consecutive branching events (Hülskamp et al., 1994). The first event results in a proximal side stem and a distal main stem, which both align to the leaf proximo-distal axis; subsequently the main stem splits into two branches, in a plane perpendicular to the primary branching plane (Fig. 4; Folkers et al., 1997). Trichome branch initiation depends on the microtubule cytoskeleton: cortical microtubules reorient when a branch is formed, and mutants or treatments that affect de novo synthesis or fragmentation and realignment of microtubules display reduced branching or even completely isotropic growth (Mathur et al., 1999; Kirik et al., 2002a; Kirik et al., 2002b; Webb et al., 2002). On the contrary, drug treatments that stabilize microtubules are able to induce new branch points in branching-defective mutants (Mathur and Chua, 2000). In addition, specific transport along microtubules also seems to have an important role in branch formation: the branching-related gene ZWICHEL 16 (ZWI) encodes a kinesin-like microtubule motor protein (Oppenheimer et al., 1997) that interacts with the C-terminal binding proteins/brefeldin A ADP-ribosylated substrates (CtBP/BARS) related protein ANGUSTIFOLIA (AN; Folkers et al., 2002). Both zwi and an mutants display only two branches as a consequence of no secondary branching event (Hülskamp et al., 1994). AN may be involved in regulating vesicle budding from Golgi stacks, and another kinesin protein, KINESIN-13A, has been found to play a role in trichome branching and distribution of Golgi stacks at the cell periphery (Lu et al., 2005). These findings have led to the formulation of a “Golgi delivery model” (Smith and Oppenheimer, 2005) where KINESIN13A directs transport of Golgi stacks along cortical microtubules to branch initiation sites and AN, which is targeted to the branching sites by ZWI, facilitates fission of vesicles containing material for branch initiation. After branches are initiated, trichomes continue increasing their size by expanding along the whole cell surface, in a process that strictly depends on an intact actin cytoskeleton (Hülskamp, 2004). Mutants affected in the DISTORTED class of genes all display defects in the directionality of trichome expansion, similar to those of plants treated with drugs interfering with actin function (Mathur et al., 1999). Several of these genes encode subunits of two complexes involved in control of the actin cytoskeleton: the actin-related protein 2/3 (ARP2/3) complex and the suppressor of cAMP receptor defects/Wiskott-Aldrich Syndrome protein family verprolinhomologous protein (SCAR/WAVE) complex, which regulates the activity of ARP2/3 (Mathur et al., 2003a; Mathur et al., 2003b; Basu et al., 2005; Szymanski, 2005). The ARP2/3 complex promotes nucleation of new F-actin from pre-existing filaments, and could therefore be responsible for organization of the cytoplasmic actin cables that extend through the elongating trichome branches and contribute to cytoplasm organization, organelle motility and spatial regulation of cell expansion (Szymanski, 2005; Smith and Oppenheimer, 2005). Recent results have shown that activation of the SCAR/WAVE complex is mediated by signalling from the ROP guanine nucleotide exchange factor (GEF) SPIKE 1 (SPK1; Basu et al., 2008). Mutation of SPK1 also affects trichome branching and arrangement of the microtubule cytoskeleton (Qiu et al., 2002), thus suggesting interdependence and integration of microtubules and F-actin organization in trichome morphogenesis. Planar morphogenesis of pavement cells Pavement cells are unspecialized epidermal cells that coordinate their polar growth within the plane of the leaf surface (Yang, 2008). Polarity of pavement cells is multidirectional and the acquisition of a specific cell morphology is based exclusively on cell-cell coordination of growth. 17 Therefore, interconnected shaping of pavement cells is referred to as planar morphogenesis (Nakamura et al., 2012). Pavement cells possess a characteristic jigsaw puzzle shape consisting of interdigitating lobes and indentations whose development is spatiotemporally mutually regulated in neighbouring cells (Fig. 5; Yang, 2008). The mechanisms that establish and modulate cell polarity in pavement cells are strikingly similar to those that operate in other types of polarly growing cells: they are based on organization of the actin and microtubule cytoskeleton through ROP small GTPases, which in turn transduce signals from the hormone auxin (Smith and Oppenheimer, 2005). Feedback regulation within the signalling pathway ensures stability of the system and allows for polarity to be integrated between adjacent cells (Xu et al., 2011). Figure 5. Interdigitated morphogenesis of pavement cells. The polarly localized auxin efflux carrier PIN-FORMED 1 (PIN1) facilitates auxin accumulation in the cell wall between two developing pavement cells. At the lobe of one cell, AUXIN BINDING PROTEIN 1 (ABP1) senses auxin and activates Rho-of-plant 2 (ROP2), which promotes assembly of cortical actin filaments (F-actin) through the RIC4 effector. In the adjacent cell, auxin signalling activates Rho-of-plant 6 (ROP6), which triggers the association of the effector RIC1 with cortical microtubules. This association promotes the formation of microtubule bundles that restrict cell expansion and generate an indentation. The two pathways antagonize each other: ROP2 suppresses RIC1 activity in lobes, whereas microtubule bundles repress the interaction between ROP2 and RIC4 in indentations. Image reproduced from Pietra and Grebe (2010) with permission of the publisher. Recent findings suggest that in pavement cells extracellular auxin is sensed by ABP1 and induces interaction between ABP1 and the TMK receptor at the plasma membrane (Xu et al., 2014). This complex transduces the auxin signal to activate two antagonistic membrane-associated ROPs (Fig. 5; Xu et al., 2010): ROP2, through its effector RIC4, promotes the assembly of cortical F-actin and induces polar outgrowth and the formation of a lobe (Fu et al., 2005); ROP6 on the contrary, through its effector RIC1, organizes 18 parallel bundles of cortical microtubules that locally restrict cell expansion and generate an indentation (Fu et al., 2009; Smith, 2003). Inhibition of RIC1 by ROP2 and repression of ROP2 activation through RIC1-dependent microtubule organization (Fu et al., 2005) ensure that only one of the two antagonistic pathways is activated in a specific cortical region of a pavement cell. This and another feedback mechanism, ROP2 stimulating PIN1 polar accumulation at lobe tips to export more auxin into the cell wall (Xu et al., 2010), enforce growth coordination between neighbouring cells and determine the growth of lobes where the contiguous cell is forming indentations, and indentations where there are lobes (Xu et al., 2011). The example of pavement cells morphogenesis illustrates how tight regulation of cytoskeletal organization through ROPs can modulate anisotropic cell expansion to generate functional shapes. ROPs are essential for the coordinated establishment of cell polarity in adjacent cells and are key elements for transduction of polarizing signals like auxin and other hormones (Li et al., 2013). Microtubules and their organization at the cell cortex Microtubules are multimeric structures composed of heterodimers of the globular proteins α- and β-tubulin. One α-tubulin and one β-tubulin molecule, of which the Arabidopsis genome encodes six and nine isoforms respectively (Ishida et al., 2007), each bind a GTP molecule and then associate forming active dimers that assemble into hollow cylindrical structures. After polymerization, β-tubulin monomers hydrolyze their GTP nucleotide to GDP, losing their ability to reassemble onto a microtubule until GDP is exchanged for GTP. Subunit asymmetry determines the polar nature of microtubules, whose two ends have different assembly and disassembly rates: the faster growing end is designated as the plus or leading end, while the less active one is referred to as the minus or lagging end (Desai and Mitchison, 1997). Microtubules are highly dynamic polymers, with the ability at steady state to constantly add and lose subunits at each of their two ends. Initiation of new microtubules requires nucleation complexes, whose role is to lower the energetic barrier for assembling new polymers. In plants, these complexes are composed of two molecules of γ-tubulin associated with one molecule each of γ-tubulin complex protein 2 (GCP2) and γ-tubulin complex protein 3 (GCP3; Seltzer et al., 2007). While nucleation complexes of animal cells are located on centrosomes during interphase, plant cells lack centrosomes, and their nucleation sites localize at the nuclear surface and at the cell cortex (Ehrhardt and Shaw, 2006). Cortical microtubules can initiate either from widely dispersed locations devoid of other polymers or, more commonly, branching at a defined angle from sites residing along existing microtubules 19 (Fig. 6; Nakamura et al., 2010; Murata et al., 2005). As microtubules elongate, their minus end is released from the nucleation site through action of a katanin-containing severing complex, and new dimers are added to the plus end (Sedbrook and Kaloriti, 2008). New polymers that are created in proximity to the cell cortex soon establish cortical association (Ehrhardt, 2008). While microtubules grow, they keep being cross-linked to the cell membrane through a mechanism that may depend on the lipidhydrolyzing enzyme PHOSPHOLIPASE D (PLD; Gardiner et al., 2001; Dhonukshe et al., 2003). At this stage both the plus and minus end of microtubules undergo dynamic instability, a behaviour that alternates moments of growth and periods of depolymerization (Fig. 6; Chan et al., 2003). The presence of a cap of GTP-bound tubulin subunits at the plus end favours the addition of new tubulin dimers, while the GDP-bound β-tubulin subunits at the minus end tend to disassemble, promoting depolymerization at this end. This causes microtubules to relocate by continuous addition and removal of dimers, through a mechanism that is termed hybrid treadmilling Figure 6. Plant microtubule dynamics. Modes (Fig. 6; Shaw et al., 2003). of microtubule initiation, growth and relocation Association to the plasma behaviours, and possible outcomes of microtubule encounters. Image adapted from Müller et al. (2009) membrane ensures that with permission of the publisher. microtubules do not slide laterally or leave the cell cortex. Thus, microtubules advance by treadmilling and explore the two-dimensional cortical space until they encounter other microtubules. Encounter of a growing microtubule end with another microtubule leads to cross-over or catastrophe when the two meet at an angle greater than 3040°, and induces bundling when the angle of incidence is lower (Fig. 6; Dixit 20 and Cyr, 2004). Cross-over often determines severing of microtubules by the severing protein katanin (KTN1; Wightman and Turner, 2007), while catastrophe consists in the rapid switch from growth to shrinking. Loss of the GTP-cap initiates depolymerization of the GDP-bound subunits from the plus end and can only be reversed by addition of new GTP-activated tubulin dimers, an event termed “rescue” (Sedbrook and Kaloriti, 2008). Bundling after sharp-angle encounters is at the base of the self-organization of microtubules into arrays of higher-order structures: collision at low angles induces realignment of the microtubule trajectory leading the growing microtubule to continue elongating parallel to the encountered microtubule (Dixit and Cyr, 2004). While the growing microtubule advances by treadmilling, its non-aligned minus end depolymerizes or “zips” to the encountered microtubule, thus leading to perfectly coaligned structures that are likely to induce other inciding microtubules to realign, and generate progressively larger bundles (Ehrhardt, 2008). Formation of stable bundles is essential for microtubules to provide an enduring cortical scaffold for biochemical complexes that require support over an extended period of time (Ehrhardt and Shaw, 2006). The main example of such structures are cellulose synthase complexes, which need to be guided while they deposit cellulose microfibrils in the cell wall (Paredez et al., 2006). By reorienting microtubule bundles at different developmental stages, in specific locations within the cell, or in response to specific environmental signals (Lindeboom et al., 2013), cells are able to affect the cell wall growth pattern and regulate their own morphology and growth anisotropy. For instance, cortical microtubules of newly divided root meristematic cells are relatively randomly arranged, and they organize into a transversal pattern of bundles to drive directional cell expansion until the cell reaches maturity, at which point the bundles slowly reorient and become parallel to the root axis (Ehrhardt and Shaw, 2006). Despite their higher stability compared to individual microtubules, cortical microtubule bundles are still highly dynamic structures and are constantly disassembled and recreated by means of the self-organizing dynamic interactions of the individual microtubule polymers. In order to be able to self-organize, microtubule arrays also require the activity of a large variety of microtubule-associated proteins (MAPs). MAPs are defined by their ability to bind directly to the microtubule lattice and to decorate microtubules (Sedbrook and Kaloriti, 2008). They have roles that range from nucleating to stabilizing or destabilizing, cross-linking, severing and anchoring microtubules. Some MAPs are conserved between animals and plants, while others belong to families that are only found in plant species (Gardiner, 2013). Prominent MAPs that are involved in the organization of cortical microtubules include the already mentioned GCP2 and GCP3 (Nakamura 21 and Hashimoto, 2009), required for microtubule nucleation, and KTN1, the homolog of the animal p60 katanin ATPase subunit that severs microtubules releasing their minus ends (Burk et al., 2001; Bichet et al., 2001). MICROTUBULE ORGANIZATION 1 (MOR1) is a MAP required for construction of the microtubule array (Whittington et al., 2001), while MAP65 is part of a family that creates bridges between microtubules and is essential for their stability and bundling (Smertenko et al., 2000). A group of MAPs localize to growing ends of microtubules and dissipate upon microtubule shrinking, and are therefore called plus-end tracking proteins (+TIPs). They include the plant homologs of END BINDING 1 (EB1; Chan et al., 2003) and CLIP170-associated protein (CLASP; Kirik et al., 2007; Ambrose et al., 2007), and the plant-specific SPIRAL 1 (SPR1; Sedbrook et al., 2004; Nakajima et al., 2004). In animal cells, +TIPs stabilize plus ends of microtubules in the so-called “search and capture” process, where they promote recognition and association of microtubules with specific structures (Sedbrook and Kaloriti, 2008). A similar role could be hypothesized in plants, where +TIPs might regulate microtubule stability and mediate binding of plus ends as well as association with the cell cortex (Sedbrook and Kaloriti, 2008). Motor proteins are a specific class of MAPs whose role is to utilize the energy released by ATP hydrolysis to transport cargoes along microtubules. Animals possess two classes of motor proteins, kinesins and dyneins. While apparently the Arabidopsis genome does not encode for any dynein heavy chain gene (Wickstead and Gull, 2007), it possesses a large variety of genes encoding kinesins, both able to travel towards the plus and the minus end of microtubules (Gardiner, 2013; Struk and Dhonukshe, 2014). Organization and dynamics of the actin cytoskeleton The plant actin cytoskeleton is an array of long polymers of the globular protein actin, called filaments. Actin filaments possess diverse functions and in interphase cells are involved in cytoplasmic streaming, intracellular movement of organelles, and trafficking of secretory and endocytic vesicles (Hussey et al., 2006; Henty-Ridilla et al., 2013). By acting on these different processes, actin controls cellular architecture and cell shape, polar growth, and responses to external stimuli (Wasteneys and Galway, 2003; Hussey et al., 2006). Actin filaments can form a diffuse meshwork of fine strands or organize in thick parallel or antiparallel bundles, which serve as tracks for motor proteins like myosins that hydrolyze ATP to transport organelles or vesicles moving along the cytoskeletal scaffold (Staiger and Blanchoin, 2006). Arabidopsis possesses eight actin isoforms that can be divided in two classes according to their expression pattern: ACT2, ACT7 and ACT8 are 22 vegetative actins, while ACT1, ACT3, ACT4, ACT11 and ACT12 are preferentially expressed in reproductive tissues (Kandasamy et al., 2009). Different isoforms vary in their biochemical properties and ability to interact with actin-binding proteins (ABPs), indicating a marked functional specificity within each family (Wasteneys and Galway, 2003). Globular actin (G-actin) binds one ATP nucleotide and polymerizes forming actin filaments (F-actin). Because of the endogenous asymmetry of actin monomers and of the lag between incorporation of an ATP nucleotide, hydrolysis to ADP and release of inorganic phosphate, actin filaments are polar and have two ends, named the barbed (or plus) end and the pointed (or minus) end. The barbed end has a higher affinity for actin subunits and grows faster, while the pointed end does not usually extend and displays a prevalence of ADP-actin (Hussey et al., 2006). The actin cytoskeleton needs to continuously rearrange in order to be functional, and the highly dynamic actin filaments base their rapid turnover on a mechanism termed stochastic dynamics (Staiger et al., 2009). Unlike microtubule plus ends, growing ends of actin filaments almost never arrest their growth or depolymerize, but rather keep elongating at high rates; disassembly of the filaments does not depend on the loss of monomers from the minus end as in treadmilling microtubules, but is instead achieved through stochastic and frequent severing and loss of the aged region of the filament (Staiger et al., 2009; Henty-Ridilla et al., 2013). In all eukaryotic organisms, actin dynamics and remodelling strictly depend on the coordinated activity of multiple classes of ABPs. Plants possess a great variety of ABPs, diversified by their roles, expression patterns and biochemical properties (Kandasamy et al., 2009). Their multiple functions can often be inferred through sequence similarity with ABPs from other organisms; however, in some cases their specific activities can be substantially different (Staiger and Blanchoin, 2006). Profilin is a small ABP that binds monomeric actin and has a central role in organizing the actin cytoskeleton. Profilin forms 1:1 complexes with most of the free G-actin monomers present in the cell (which represents over 90% of the actin pool) and prevents both their incorporation at the pointed ends of actin filaments and the spontaneous assembly of new filaments (Staiger and Blanchoin, 2006). Nucleation of actin filaments initiates either freely in the cytosol or, in most of the cases, on the side of another filament (Hussey et al., 2006). Both these modes require the action of specific ABPs: formins are able to bind profilin-actin monomers and generate new filaments, while the ARP2/3 complex attaches to the side of existing filaments, initiating a new F-actin branch (Henty-Ridilla et al., 2013). Activity of the ARP2/3 complex is modulated by the SCAR/WAVE complex (Szymanski, 2005). The SCAR/WAVE complex may be an effector of ROP2, and therefore this could be a pathway that integrates signals to organize the actin cytoskeleton (Basu 23 et al., 2008). Surprisingly though, while the ARP2/3 complex is essential in yeast and animals, mutations in Arabidopsis genes encoding for ARP2/3 subunits only determine defects in the shape of trichomes and pavement cells, suggesting that the complex may have a more limited role in plants than in animals (Hussey et al., 2006; Mathur et al., 2003a; Mathur et al., 2003b; Basu et al., 2005). The rate of actin filament elongation is modulated by several ABPs that bind with high affinity to barbed ends and block the incorporation of new monomers. One of these ABPs is the heterodimeric capping protein (CP), which ensures a small population of short F-actin is maintained in each cell and regulates the availability of barbed ends controlling the number of growing filaments (Hussey et al., 2006). CP inactivation by phosphoinositide lipids or phosphatidic acid releases the growing end of a filament and allows polymerization to proceed (Staiger and Blanchoin, 2006). Disassembly of the filaments is driven by the severing activity of several ABPs in the aged region, where all actin subunits are bound to ADP. The most prominent severing ABPs are ACTIN DEPOLYMERIZATION FACTORS (ADFs), also known as cofilins, and villins (Henty-Ridilla et al., 2013). Activity of ADFs is modulated by several factors, including the ACTIN INTERACTING PROTEIN 1 (AIP1) plant orthologs. Plant AIP1 has been shown to enhance the activity of ADFs both in vitro (Allwood et al., 2002) and in vivo (Augustine et al., 2011). Whereas ADFs remain attached to the severed filaments but leave the barbed end free to potentially reinitiate growth, villins and AIP1 cap the filament end, inhibiting further elongation (Henty-Ridilla et al., 2013). The small fragments resulting from filament severing mostly undergo depolymerization, leading to single ADP-bound ADF-actin complexes. The exchange of ADP to ATP on actin subunits is most likely favoured in plants by the CYCLASE ASSOCIATED PROTEIN (CAP), which regenerates profilin-actin monomers capable of assembling to the growing end of a new filament (Staiger and Blanchoin, 2006). In most plant cells, a large part of the actin filaments organize in robust bundles or cables that are generally less dynamic and longer-lived than actin filaments (Wasteneys and Galway, 2003). Such structures can be initiated by formins or generated by several classes of bundling ABPs, which collect single filaments giving rise to higher-order structures (Hussey et al., 2006). The actin filament cross-linkers fimbrins and villins, as well as ELONGATING FACTOR-1α (EF-1α), can bind actin and bundle it (Wasteneys and Galway, 2003; Hussey et al., 2006). Binding of fimbrins and villins to actin filaments also prevents modification by other ABPs, protecting the bundles from severing and disassembly and making them more stable (Staiger and Blanchoin, 2006). 24 The plant cytoskeleton during cell division In addition to their roles in providing a dynamic intracellular scaffold and controlling cell growth and polarity, the microtubule and actin cytoskeleton carry out essential functions also during cell division. Plant cells are surrounded by a rigid cell wall and thus divide based on very different mechanisms compared to animal cells and require a set of unique cytoskeletal structures that enable mitosis and cytokinesis to occur. The first sign of an approaching mitosis in plants is the appearance of the preprophase band (PPB), a cortical band of microtubules and actin filaments that mark the site of the future division plane (Fig. 7; Wasteneys, 2002). In cells dividing symmetrically, the PPB is located at the cell equator just beneath the plasma membrane and encircles the nucleus, which is centred before mitosis through the action of actin filaments and microtubules radiating from the nuclear surface (Rasmussen et al., 2011). Formation of the PPB begins in the G2 phase of the cell cycle, when cortical microtubule bundles become restricted to the cell midplane via selective depolymerization and/or local stabilization (Dhonukshe and Gadella, 2003; Vos et al., 2004). Accumulation of PPB actin filaments depends on an intact microtubule cytoskeleton and contributes to constraining width of the PPB, which becomes narrower as the cell approaches prophase (Rasmussen et al., 2013). Several proteins localize to the PPB during prophase and are required for correct PPB formation. Among these are the regulatory B’’ subunit of the protein phosphatase 2A complex TONNEAU 2 (TON2)/FASS (Camilleri et al., 2002), the protein products of the tandem duplicated genes TONNEAU 1A (TON1A) and TONNEAU 1B (TON1B; Azimzadeh et al., 2008), as well as the microtubule-binding proteins MOR1 (Whittington et al., 2001), CLASP (Ambrose et al., 2007; Kirik et al., 2007) and MAP65 (Smertenko et al., 2004). Mutations in TON2/FASS, TON1A/TON1B or MOR1 completely or partially abolish PPB formation and result in cells dividing in abnormal orientations (Camilleri et al., 2002; Azimzadeh et al., 2008; Whittington et al., 2001). At the onset of prometaphase, once chromosomes are condensed, the nuclear envelope breaks down and the PPB dissolves leaving the cell cortex without any microtubules (Müller et al., 2009). A memory of the position of the PPB at the cell periphery is, however, preserved throughout mitosis as the cortical division site (CDS) that during cytokinesis guides the orientation of the newly formed cell division plane (Fig. 7; Rasmussen et al., 2011). After the PPB is disassembled, a region with very low or no actin accumulation and flanked by two thin bands of high actin density remains on the cell cortex at the former location of the PPB and is termed the actin-depleted zone (Müller et al., 2009). The actin-depleted zone coincides with a zone 25 Figure 7. Cell division in plant cells. Microtubules (green) at the cell cortex become restricted to the cell equator creating the preprophase band (PPB). When the mitotic spindle is formed, the PPB disassembles leaving a mark (red semicircles) referred to as the cortical division site (CDS). As chromosomes later decondense forming daughter nuclei (blue), microtubules of the phragmoplast guide vesicles (white circles) to the cell division plane marked by the CDS and assemble the cell plate. The cell plate expands from the centre of the cell until it fuses to the plasma membrane (PM) to complete cytokinesis. Preprophase is illustrated as whole cell view, while other stages are sections through the cell mid-plane. Image adapted from Barr and Gruneberg (2007) with permission of the publisher. devoid of the KINESIN CDKA;1ASSOCIATED 1 (KCA1), which starts to be depleted from the CDS even before disappearance of the PPB (Vanstraelen et al., 2006). Localized removal of actin remodelling proteins and KCA1 from the plasma membrane involves mechanisms that are still unknown, but could be mediated by clathrin-mediated endocytosis (Karahara et al., 2009). In addition to negative CDS markers, there are also positive markers of the CDS during mitosis and cytokinesis: one of these proteins is TANGLED 1 (TAN1), which is recruited by PPB microtubules and persists at the CDS throughout completion of cytokinesis (Walker et al., 2007). Also the Ran GTPase ACTIVATING PROTEIN 1 (RanGAP1) localizes to the PPB and stays at the CDS even after the PPB has disassembled (Xu et al., 2008). The role of TAN1 and RanGAP1 is to provide guidance for the new cell wall that will form during cytokinesis; accordingly, mutation of TAN1 or depletion of RanGAP1 cause the formation of cell wall stubs and misoriented cell division planes (Cleary and Smith, 1998; Xu et al., 2008). Furthermore, analogous defects in cell division orientation are found in pok1;pok2 double mutants, defective in the two kinesins PHRAGMOPLAST ORIENTING KINESIN 1 (POK1) and PHRAGMOPLAST ORIENTING KINESIN 2 (POK2), which are required for keeping TAN1 and RanGAP1 at the CDS during telophase (Müller et al., 2006). The involvement of two kinesins suggests that continuous transport via microtubules may be necessary to maintain the specific localization of CDS components. 26 Already prior to complete disassembly of the PPB, cells begin to form the mitotic spindle, the only mitotic structure whose function is to a large extent conserved between plants and animals (Rasmussen et al., 2013). As opposed to animal cells though, plants do not have a centrosome, and thus their spindles have diffuse, broad poles that contain numerous microtubule foci (Wasteneys, 2002). During prometaphase highly dynamic microtubules radiate from each pole, with their plus ends attaching to the chromosomes via kinetochores or overlapping at the spindle equator with microtubules coming from the opposite pole (Fig. 7). These interactions align the chromosomes during metaphase and, together with a cage of actin bundles that surround the spindle and connect it to the cell periphery, help stabilizing the spindle structure (Rasmussen et al., 2011). During mitosis spindle orientation sometimes changes with respect to the pre-existing PPB; however, as long as reorientation is subtle, it does not determine any alteration in the final division plane, as the newly forming cell wall is later able to track back to the CDS and align to it (Rasmussen et al., 2013). At the anaphase-telophase transition, once chromosomes have been pulled towards the poles and start to decondense, remnants of the disassembling mitotic spindle begin to form the phragmoplast in the central region of the cell, between the two sets of daughter chromosomes (Fig. 7; Wasteneys, 2002). The spindle microtubules maintain their polarity and are joined by new microtubules emanating from the poles and by newly polymerized actin filaments. The phragmoplast consists of two compact cylindrical bundles of parallel microtubules and actin filaments that face each other with opposite polarities at the plane of cell division (Jürgens, 2005). Plus ends of the phragmoplast microtubules interdigitate, while actin filaments do not overlap (Jürgens, 2005). Microtubules that form the phragmoplast direct the delivery of Golgi-derived vesicles to the plane of cell division, where the vesicles fuse synthesizing a transient membrane compartment called cell plate. The cell plate initiates as a disc at the centre of the cell division plane and progressively expands laterally and matures, giving rise to the new cell wall that separates the two daughter cells (Fig. 7; Samuels et al., 1995; Jürgens, 2005). As the cell plate develops, excessive membrane is removed from its centre by endocytosis (Samuels et al., 1995; Seguí-Simarro et al., 2004) and microtubules begin disassembling in the central region of the phragmoplast and translocating to the outer edges, where new actin filaments are polymerized as well. This centrifugal growth drives the cell plate towards the CDS, where eventually cell plate and parental membrane fuse completing cytokinesis (Jürgens, 2005). Microtubules and F-actin, in addition to the already mentioned TAN1 and RanGAP1, may have a role in directing the phragmoplast to the CDS (Rasmussen et al., 2013). Moreover, cell plate maturation and attachment to the mother cell wall is controlled by several 27 additional factors, including the microtubule-associated protein AUXIN INDUCED IN ROOTS 9 (AIR9; Buschmann et al., 2006), the adaptor/coatomer-like protein TPLATE (Van Damme et al., 2006), and the extensin ROOT-SHOOT-HYPOCOTYL-DEFECTIVE (RSH; Hall and Cannon, 2002). These proteins localize to the cell plate and/or the CDS shortly before fusion of the two membranes, and mutation of RSH or knockdown of TPLATE cause misoriented and incomplete cell walls, implicating them in the final steps of cytokinesis (Hall and Cannon, 2002; Cannon et al., 2008; Van Damme et al., 2006; Van Damme et al., 2011). Recently, downregulation of the TPLATE interactor TML has also been shown to induce ectopic accumulation of the KNOLLE cell plate syntaxin at the CDS and plasma membrane, further supporting a role for the TPLATE complex in correct maturation and fusion of the cell plate (Gadeyne et al., 2014). As cells complete division and enter interphase, microtubules first appear as a transient array radiating from the nucleus towards the cell periphery and soon afterwards repopulate the cell cortex and associate with the plasma membrane (Wasteneys, 2002). The microtubule-associated protein CLASP CLASP1 and CLASP2 are two mammalian MAPs that have been identified as interactors of the microtubule-binding protein CLIP170 (Akhmanova et al., 2001). They belong to the class of +TIPs and are essential for controlling the attachment of microtubule plus ends to various cellular structures both in interphase and during cell division (Lansbergen et al., 2006; Maiato et al., 2005). CLASPs stabilize subsets of microtubules by selectively binding to their plus ends in response to positional cues and by promoting pause state or rescue, thus effectively reducing microtubule dynamicity (Galjart, 2005; Sousa et al., 2007; Al-Bassam et al., 2010). During mitosis instead, CLASPs regulate kinetochore fibre dynamics by adding tubulin subunits at microtubule plus ends (Maiato et al., 2005). CLASPs may also be hubs that integrate cytoskeletal signalling, as they have been shown to associate directly with actin filaments (Tsvetkov et al., 2007) and to be regulated by processes dependent on Rho GTPases (Lansbergen et al., 2006). Arabidopsis only possesses one CLASP gene and analyses of clasp loss-offunction mutants have shown that, similar to its non-plant orthologs, CLASP in plants functions in promoting microtubule stability (Ambrose et al., 2007; Kirik et al., 2007). Consistent with these findings, clasp mutants are hypersensitive to microtubule-depolymerizing drugs, and overexpression of CLASP leads to strong bundling of microtubules (Ambrose et al., 2007; Kirik et al., 2007). In plants, CLASP is constitutively expressed at low levels in all tissues (Ambrose et al., 2007; Kirik et al., 2007). During mitosis, CLASP localizes to 28 the PPB, accumulating almost exclusively at the cell edges intersected by the band (Ambrose et al., 2011), and is present on spindles and phragmoplasts (Ambrose et al., 2007; Kirik et al., 2007). In newly divided root cells, CLASP is enriched along the apical and basal cell sides, while it becomes distributed along longitudinal edges in cells undergoing elongation (Ambrose et al., 2011). CLASP localization on cortical microtubules is not uniform, but rather consists of distinct domains at the cell periphery (Ambrose et al., 2011). In plants, the CLASP protein labels the full length of microtubules, and despite being a +TIP shows only a weak enrichment at the plus end; there it maps just behind the +TIP EB1b (Ambrose et al., 2007; Kirik et al., 2007). A prominent role of the Arabidopsis CLASP is to coordinate the interaction of microtubules with the plasma membrane: CLASP mediates close association of cortical microtubules with the cell cortex, and in mutants defective for CLASP the microtubules detach more frequently from the plasma membrane and subsequently form hyper-parallel and more prominent bundles (Ambrose and Wasteneys, 2008). Another role for CLASP at the cell periphery is to promote the growth of cortical microtubules around sharp cell edges, counteracting barrier-induced microtubule catastrophes (Ambrose et al., 2011). Enrichment of CLASP along specific cell edges allows the extension of microtubules into adjacent cell faces and the formation of transfacial bundles in meristematic root cells (Ambrose et al., 2011). Regulation of local CLASP accumulation at cell edges can thus modulate cell-wide organization of the cortical microtubule array, including formation and orientation of the PPB (Ambrose et al., 2011). Recently CLASP has also been connected to tethering of endosomal compartments to cortical microtubules through direct interaction with the retromer-complex component SORTING NEXIN 1 (SNX1; Ambrose et al., 2013). SNX1 is involved in endosomal recycling of membrane proteins, including the auxin transporter PIN2 (Jaillais et al., 2006). CLASPdependent recruitment of SNX1 vesicles to cortical microtubules modulates targeting and abundance of PIN2 at the plasma membrane (Ambrose et al., 2013). A role for CLASP in regulating PIN2 polarity remains controversial and awaits further analyses (Brandizzi and Wasteneys, 2013). Consistent with an effect of CLASP on PIN2 localization and auxin transport, clasp mutants display an array of auxin-related defects: plants have reduced apical dominance and abundant lateral roots, roots are shorter and have smaller apical meristems, aberrant cellular organization, and delayed gravitropic responses (Ambrose et al., 2013). Furthermore, root and hypocotyl cells of clasp mutants are shorter, there are fewer cells in the root meristem, rosette leaves are twisted, pavement cells are less interdigitated and trichomes have less branches, indicating that CLASP in multiple ways affects cell expansion and morphogenesis as well as cell division (Ambrose et al., 2007; Kirik et al., 2007). Interestingly, trichome branches of clasp 29 mutants are also misoriented with respect to the leaf proximo-distal axis (Kirik et al., 2007). This observation, together with the more frequent occurrence of PPBs that are misaligned relative to the cell axis in clasp mutants (Ambrose et al., 2007), suggests a potential role for CLASP in cell and planar polarity. Patterning of the root and leaf epidermis The root and leaf epidermis of Arabidopsis are models for studying planar polarity in plants. In addition, they are optimal systems also to analyze the specification of cell fate in the context of a tissue layer (Ishida et al., 2008). During their development, root and leaf epidermal cells are faced with the option of whether to differentiate or not specialized structures, root hairs and trichomes respectively. Cells do not choose their fate individually, but rather receive specific signals that coordinate patterning of cell fate over the whole tissue, transmitting positional cues and exploiting cell-to-cell communication (Schiefelbein et al., 2009; Hülskamp, 2004). In plants as in other organisms, body pattern is organized during embryogenesis, where the apical-basal body axis is established, principal tissues and organs are formed, and different cell types are specified (Jürgens et al., 1991). This also holds true for epidermal cell patterning, which is set up in the embryo and then influences postembryonic development, when fates are translated into morphological outputs (Berger et al., 1998a). In roots, fate patterning generates a characteristic alternation of parallel cell files that produce either hair or non-hair cells (Galway et al., 1994), while on the leaf surface trichomes occur almost never in clusters but rather initiate at a fixed distance from each other (Hülskamp, 2004). The longitudinal files that align over the root surface are derived from 16 meristematic initials that attain a specific fate according to the position they occupy with respect to the underlying layer of 8 cortical cells (Dolan et al., 1993). Cells located over the anticlinal walls of two cortical cells will usually grow a hair and are thus referred to as hair (H) cells, or trichoblasts; on the other hand, the cells lying over the outer periclinal wall of a single cortical cell do not generally grow any hairs and are referred to as non-hair (N) cells, or atrichoblasts (Fig. 8; Dolan and Costa, 2001). Cell files propagate by transversal cell divisions, and files of hair-forming trichoblasts are normally separated by either one or two files of hairless atrichoblasts (Dolan et al., 1994). Rare occurrence of longitudinal cell divisions can alter the number of epidermal cell files and possibly the fate of daughter cells; changes in patterning can in fact still take place postembryonically and cells are constantly assessing the positional information they receive from the underlying cortex (Berger et al., 1998a; Costa and Shaw, 2006). 30 Figure 8. Cellular organization and epidermal cell differentiation in the root of Arabidopsis thaliana. Coloured transverse section of a root displaying epidermal hair (H) cells that lie over the cleft between two cortical cells in red and non-hair (N) cells located on top of one single cortical cell in blue. Both the positional cues that direct root hair fate acquisition and the intercellular communication that spaces trichomes over the leaf surface converge onto a similar set of transcription factors that organize epidermal patterning both in the root and in leaves (Fig. 9). The central component of this pathway is a protein complex capable of activating the homeobox transcription factor GLABRA 2 (GL2), which leads to non-root-hair and trichome cells (Di Cristina et al., 1996, Masucci et al., 1996, Rerie et al., 1994). Expression of GL2 represses transcription of the hair-initiating gene RHD6 (Menand et al., 2007) and of PHOSPHOLIPASE Dζ1 (PLDζ1; Ohashi et al., 2003), a gene controlling root hair initiation and morphogenesis. The core components of the activator complex have been uncovered over the years, and include the WD40-repeat protein TRANSPARENT TESTA GLABRA 1 (TTG1; Galway et al., 1994), the homologous bHLH transcription factors GLABRA 3 (GL3) and ENHANCER OF GLABRA3 (EGL3; Bernhardt et al., 2003), and the R2R3 MYB transcription factor WEREWOLF (WER; Lee and Schiefelbein 1999) or GLABRA 1 (GL1; Oppenheimer et al., 1991). WER and GL1 are tissue-specific components and are expressed exclusively in roots and leaves, respectively. A group of proteins that include CAPRICE (CPC; Wada et al., 1997), TRIPTYCHON (TRY; Schellmann et al., 2002) and ENHANCER OF TRY AND CPC 1 (ETC1; Kirik et al., 2004) have the role of inhibiting the activity of the complex (Simon et al., 2007; Wester et al., 2009). Synthesis of these small MYB-like DNA-binding proteins takes place in N and trichome cells and they later move to adjacent hair and trichome-less cells where they compete with WER or GL1 for binding to the bHLH transcription factors (Wada et al., 2002; Kurata et al., 2005). Lack of a transcriptional activation domain in CPC, TRY and ETC1 means that whenever they bind to GL3/EGL3 the complex does not activate GL2 transcription, thus determining root-hair 31 or trichome-less fate (Ishida et al., 2008). In a mechanism of lateral inhibition aimed at stabilizing fate determination and patterning in the epidermis, transcription of CPC, TRY and ETC1 is induced by the activator complex, which therefore both supports non-hair/trichome fate in a cell and promotes hair/no-trichome fate in its neighbours (Koshino-Kimura et al., 2005, Ryu et al., 2005). In addition to this, transcription of GL3 and EGL3 is repressed by the activator complex in non-hair cells and stimulated by CPC and TRY in hair cells; once expressed, the bHLH proteins move to non-hair cells where they reinforce fate selection by generating more activator complexes (Bernhardt et al., 2005). In root epidermal cells, an additional level of feedback regulation is provided by the MYB23 gene, encoding a transcription factor closely related to WER (Kang et al., 2009). MYB23 is able to substitute for WER in the activator complex, which in turn can induce expression of MYB23, therefore generating a positive loop that stabilizes accumulation of the patterning determinants (Kang et al., 2009). Figure 9. Simplified representation of the gene networks controlling patterning of the root and leaf epidermis. The activator complex including the WD40-repeat protein TRANSPARENT TESTA GLABRA 1 (TTG1), the homologous bHLH transcription factors GLABRA 3 (GL3) and ENHANCER OF GLABRA3 (EGL3), and either the MYB transcription factor WEREWOLF (WER) or its homolog GLABRA 1 (GL1) promotes expression of the GLABRA 2 (GL2) homeobox transcription factor and induces differentiation into non-hair (N) cells or trichome cells. The activator complex also activates expression of the CAPRICE (CPC) or TRIPTYCHON (TRY) MYB-like DNA-binding proteins, which move to neighbouring cells and compete with WER or GL1, respectively, for binding to GL3/EGL3. In root hair (H) cells, a positional cue from the receptor-like kinase SCRAMBLED (SCM) represses WER expression. Image adapted from Ishida et al. (2008) with permission of the publisher. In trichome patterning, the so-called activator-depletion model described by Bouyer et al. (2008) provides an additional option to the activator-inhibitor model presented so far. The authors show that the activator TTG1 is ubiquitously expressed in the leaf epidermis and its protein product is capable of moving between cells. The GL3 transcription factor is present at 32 high concentrations in trichome cells and can interact directly with TTG1 and trap it in the nucleus, efficiently depleting a positive regulator of trichome initiation from cells adjacent to trichomes (Balkunde et al., 2011). This mechanism of lateral inhibition does, however, not exclude functioning of the core patterning pathway and is instead likely to operate in parallel with it (Pesch and Hülskamp, 2009; Balkunde et al., 2011). Moreover, a similar process has been recently proposed also for epidermal cells in roots, where CPC, an inhibitor of non-hair fate, could get trapped by EGL3 in H cells, thus ensuring accumulation of a promoter of H fate and providing stability to cell fate patterning (Kang et al., 2013). Upstream of the core patterning pathway, specification of cell fate in the root epidermis relies on signals from the underlying cortical layer. The molecular nature of these positional cues and how exactly they are generated and transmitted to epidermal cells still remains elusive; however, they likely require signalling through the receptor-like kinase SCRAMBLED (SCM; Kwak and Schiefelbein, 2007). SCM action inhibits transcription of WER preferentially in H cells, on account of the higher amount of SCM in this position (Kwak and Schiefelbein, 2008). This differential accumulation is achieved through inhibition of SCM expression in N cells by the activator complex, which in this manner negatively feeds on its own inhibitor and gives additional robustness to cell fate decisions (Kwak and Schiefelbein, 2008). Additional factors non-cell-autonomously contribute to epidermal fate specification prior to the patterning gene network. The zinc finger protein JACKDAW (JKD) modulates epidermal patterning by acting from the underlying cortical layer and asymmetrically distributing cell fate determinants in an SCM-dependent manner (Hassan et al., 2010). As a consequence, in jkd mutants the fate of epidermal cells is less tightly connected with their position relative to the underlying cells, and ectopic hairs and non-hairs form more frequently (Hassan et al., 2010). The heat shock transcription factor SCHIZORIZA (SCZ) suppresses epidermal fate in cortical cells and controls cell fate separation in ground tissue; mutants defective in SCZ develop root hairs also from the cortical cell layer and, moreover, display perturbations in the establishment of epidermal hair and non-hair cell fates (Mylona et al., 2002; ten Hove et al., 2010). Patterning of epidermal cell fate in Arabidopsis thaliana has been extensively studied over the past 20 years, and many of the molecular mechanisms and transcriptional networks that control cell fate specification in the root and leaf epidermis have been elucidated (Schiefelbein et al., 2014). However, new players are continuously discovered and challenge the present knowledge about the establishment of cell differentiation patterns in the epidermis. Adding new components to these pathways and investigating their role will provide a clearer and more complete picture of the complex regulation behind fate patterning. 33 Aim The aim of this study was to employ polar initiation of Arabidopsis root hairs as a model system to identify new components of the plant planar polarity pathway. The work focussed on characterizing the novel planar polarity actor SABRE and analyzing its functions in cell division orientation and stabilization of microtubule arrays. Additional research concentrated on the role that the actin cytoskeleton and a modulator of its organization have in planar polarity of root hair positioning. Finally, further investigation addressed the connection between planar polarity establishment and cell fate patterning in the root epidermis. 34 Results and Discussion Root hair positioning in Arabidopsis thaliana provides an easy-to-analyze phenotype for studying the establishment of planar polarity in plants. Genetic dissection of the mechanisms that control this process has helped outlining a pathway that describes how plants set up and maintain polar hair initiation (Masucci and Schiefelbein, 1994; Fischer et al., 2006; Ikeda et al., 2009). Nonetheless, we still know relatively little about how polarity cues are generated and translated into morphological outputs. This prompted us to investigate the presence of additional players involved in root hair planar polarity. To this end, we performed a genetic screen and isolated several mutants affected in the polar outgrowth of hairs on their root epidermis. One of the mutants identified was named kreuz und quer (kuq) [German for criss-cross] on account of its intricate placement of hairs that emerged at very diverse positions along trichoblasts, displaying clear defects in their polar positioning (Paper I, Fig. 1a-c). The kreuz und quer mutant kreuz und quer is mutated in the SABRE gene Positional cloning of the kuq mutation revealed that it was located in the large SABRE (SAB) gene and consisted of a single nucleotide deletion likely to introduce a premature stop codon in the last sixth of the coding sequence (Paper I, Fig. 2). Analysis of T-DNA insertion lines in the SAB gene revealed homozygous mutants with phenotypes identical to those of kuq and unable to complement the kuq mutation, thus confirming that the planar polarity defects of kuq were caused by disruption of SAB function (Paper I, Fig. 3a-e). Reverse transcription PCR (RT-PCR) analysis of three sab mutant lines confirmed that they did not produce any full-length SAB transcript (Paper I, Fig. 3f-h). This finding, together with the observation that 9 out of 11 T-DNA insertion lines analyzed had indistinguishable phenotypes, suggested that they were null alleles. We thus adopted one of these lines, named sab-5, for further analyses. Two of the analyzed T-DNA insertion lines in the SAB gene displayed phenotypes similar to, but weaker than kuq and the other sab alleles. One of them carried a T-DNA insertion in the promoter region of SAB, suggesting that it might affect SAB expression rather than protein functionality. In the other line the T-DNA was inserted only 300 nucleotides away from the end of the 7824 nucleotides coding sequence, indicating a hypomorphic mutation that might be producing a truncated, yet still partially functional version of the SAB protein (Paper I, Fig. 2 and Fig. 3a,b,e). These two weak alleles have 35 not been further analyzed yet, but in the future could prove valuable tools for dissecting SAB function: in addition to opening the way for studies on dosage-dependence of SAB function and for serial-deletion analyses, they are also the only two self-fertile sab mutants identified in this study and could be used in genetic screens for SAB interactors. Mutants defective in SAB have first been described in 1993 (Benfey et al., 1993) and were isolated in a genetic screen for genes affecting root morphogenesis. sab-mutant seedlings displayed shorter roots with increased diameter, and more detailed analyses revealed that their increased radial expansion when compared to the wild type was mainly a consequence of defects in anisotropic growth of the cortical layer (Benfey et al., 1993; Aeschbacher et al., 1995). Pharmacological and genetic interference with ethylene biosynthesis or perception could partially rescue the sab phenotype and this led the authors to hypothesize that SAB was involved in limiting radial cell expansion by antagonizing ethylene action (Aeschbacher et al., 1995). Antagonistic interaction of SAB with ethylene signalling was also suggested recently in a study that identified sab mutants as hypersensitive to phosphate starvation (Yu et al., 2012). Both studies report that ethylene biosynthesis is not altered by mutations in SAB. However, the molecular nature of the interplay between SAB and this hormone remains elusive and requires further investigation. Planar polarity defects of sab mutants Quantitative analysis of root hair positioning in kuq and other sab mutants revealed an irregular distribution of hair position compared to wild-type seedlings, with hairs initiating more frequently both at the very basal end and towards the middle of trichoblasts (Paper I, Fig. 1a-c and Fig. 3c-e). We observed similar defects in the localization of ROPs at hair initiation sites prior to root hair emergence: patches of ROP proteins accumulated either at the basal-most end of the cells or at a more apical position with respect to the wild type (Paper I, Fig. 1d-g and Fig. S2a-d). ROPs are the earliest markers of polar hair initiation (Molendijk et al., 2001; Jones et al., 2002), and these findings are consistent with an involvement of SAB in planar polarity establishment upstream of ROP positioning. The root-hair and ROP phenotypes of sab mutants are intriguing concerning their display of both a basal and an apical shift in hair initiation. Most of the mutants with planar polarity defects described so far in plants have either a basal or an apical shift: for instance enhanced biosynthesis or signalling of either ethylene or auxin determine hyperpolarization of root hair initiation, while defects in perception of the hormones shift hair outgrowth towards a more apical position (Masucci and Schiefelbein, 1994; Grebe et al., 2002; Fischer et al., 2006; Ikeda et al., 2009). The fact that 36 mutations in SAB provoke a shift in both directions suggests that the effect of this gene cannot be limited to a linear interaction with one of the two main hormone pathways. This hints towards a more complex involvement of SAB in planar polarity establishment, possibly through multiple contributions to independent pathways. In addition to the positioning of hairs in the root epidermis, we analyzed the alignment of trichomes on leaves of sab mutants as a second system to study polar orientation of cells within the plane of a tissue layer. The proximal side stem, which in wild type aligns to the proximo-distal axis of the leaf (Folkers et al., 1997), had a much more variable orientation in both kuq and sab-5 mutants (Paper I, Fig. 1h-j and Fig. S1e,f). A similar misalignment had only been reported for mutants defective in the MAP gene CLASP (Kirik et al., 2007), and indicates that the effect of SAB on planar polarity is not limited to the root but extends to aerial tissues. Trichoblasts align to an auxin gradient in the root; however, a similar uniform auxin gradient over the whole leaf surface has not yet been described (Mattsson et al., 2003). Auxin certainly has instructive roles in leaf initiation and vascularization (Leyser, 2005; Scarpella et al., 2006), and it will be interesting to investigate if it also contributes to planar polarity of leaf trichomes or whether this tissue relies on different, yet to be identified polarizing cues. The discovery that SAB functions in establishment of polarity in both the root and the leaf epidermis suggests that the mechanisms are at least in part conserved between the two tissues. Cell polarity defects of sab trichomes When observing the morphology of leaf trichomes in kuq and sab-5 mutants, we noticed a higher frequency of trichomes with only two branches, and a higher incidence of trichomes with abnormal distance between the two branching points and/or aberrant orientation of the branches with respect to the leaf surface (Paper I, Fig. 1k-m and Fig. 5a). Initiation of trichome branches depends on the microtubule cytoskeleton (Mathur et al., 1999), and these defects indicated that SAB could act on microtubules to control the correct cell polarity of leaf trichomes. Trichome branching of sab mutants is not as severely inhibited as in seedlings treated with microtubuledepolymerizing drugs (Mathur et al., 1999) or mutants defective in the assembly of tubulin subunits (Kirik et al., 2002a; Kirik et al., 2002b), but rather resembles mutants affected in genes responsible for the microtubule stabilization or severing and realignment (Kirik et al., 2007; Burk et al., 2001). Therefore the role of the SAB gene might be in reorganization rather than assembly of the microtubule cytoskeleton. 37 The SAB protein Homologs of the SAB protein The SAB gene encodes a protein of 2607 amino acids with a predicted molecular weight of 290 kDa. The Arabidopsis genome harbours a gene for one other protein of similar size with high sequence similarity to SAB, named KINKY POLLEN (KIP) after the defects in pollen tube elongation of mutants lacking its function (Procissi et al., 2003). Similarly, a Zea mays homolog of SAB, the ABERRANT POLLEN TRANSMISSION 1 (APT1) gene, is involved in growth of pollen tubes after pollen germination (Xu and Dooner, 2006). Whereas APT1 is only expressed in mature pollen and its loss-of-function does not affect any other tissues, KIP is required for tip growth of both pollen tubes and root hairs (Procissi et al., 2003; Xu and Dooner, 2006). Pollen tubes of kip mutants are twisted and branched, and root hairs are shorter and thicker. Defects in root hair elongation are sometimes related to incorrect functionality of the actin cytoskeleton (Baluška et al., 2000; Ringli et al., 2002), and hair waving or branching is normally caused by interference with the microtubule cytoskeleton (Bibikova et al., 1999) or ROP signalling (Parker et al., 2000). Thus, although actin filaments and microtubules of kip pollen tubes are reported not to differ from wild-type (Procissi et al., 2003), it is difficult to rule out an involvement of KIP in cytoskeletal organization. The morphology of sab pollen tubes and root hairs does not differ from wild type, and on the other hand kip seedlings do not display the strong growth impairment of sab mutants (Procissi et al., 2003). However, despite apparently controlling unrelated cellular processes, SAB and KIP seem to at least partially interact genetically because sab;kip double mutants have more severe defects both in seedling development and tip growth compared to the two single mutants (Procissi et al., 2003). The expression pattern of SAB and KIP is also relatively similar, with mRNA of both genes expressed in all plant tissues and KIP enriched in tipgrowing cells (Procissi et al., 2003). The low and mostly homogeneous levels of SAB expression are in agreement with previous findings (Aeschbacher et al., 1995), with data from Genevestigator (Zimmermann et al., 2004), and with a recent report using quantitative real-time PCR to analyze levels of SAB transcripts (Yu et al., 2012). Transcriptional fusion of SAB regulatory sequences with GUS further confirmed SAB expression in all the analyzed tissues and revealed strongest signals in hypocotyl, cotyledons and leaves (Yu et al., 2012). Intriguingly, either one or two proteins with similar size and high sequence similarity to SAB are present in most plant species whose genome has been sequenced (Paper I, Fig. 4). SAB-related proteins can also be found 38 in most other eukaryotes, even distantly related ones such as yeasts and primates. Sequence homology among plants is distributed over the whole SAB protein, while in non-plant species the similarity is restricted to two regions of about 700 and 300 amino acids in the central and C-terminal parts of the protein. Such level of conservation strongly points towards the requirement for SAB in an essential process, one needed for organisms with very diverse organization and morphology. The two regions of homology have not been characterized nor are predicted to have any catalytic or structural role, and will indeed be primary targets for understanding the function of the SAB protein. In addition, the presence of these sequences in diverse eukaryotes could allow studies on protein conservation and evolution, performed by exchanging domains between different species and testing for protein functionality. Unfortunately, apart from KIP and APT1 none of the SAB-related proteins have been functionally characterized. SAB does not contain any predicted functional domains or conserved motifs, and the only hint on the role of a SAB-related protein comes from the SAB human homolog KIAA0100, which produces antigens overexpressed in several types of cancer cells (Song et al., 2006; Zhou et al., 2008). Subcellular localization of the SAB protein Analyses with bioinformatic tools gave a range of predictions on the subcellular localization and membrane topology of the SAB protein. All of the algorithms we employed agreed on the presence of a transmembrane domain close to the protein’s N-terminus and following a short stretch of cytosolic amino acids. This indicates that SAB is likely to be a transmembrane protein. Nevertheless, predictions are in disagreement concerning the total number of putative transmembrane domains and consequently the localization of the protein’s C-terminus. The presence of SAB peptides among membrane-associated proteins isolated from Arabidopsis suspension cells has been previously reported and would support SAB membrane localization (Benschop et al., 2007). On the other hand, different studies have suggested that both KIP and SAB, as well as their yeast and Drosophila homologs, do not contain membrane spanning domains but rather a signal peptide directing them towards the secretory pathway and extracellular matrix (Procissi et al., 2003). This discrepancy may be attributed to the evolution of algorithms for membrane topology predictions, which are likely to be more accurate at the time of our analysis. Additionally, fluorescent fusions of C-terminal fragments of the maize APT1 protein have been shown to localize to Golgi bodies, suggesting that the APT1 protein and its homologs may be targeted to these organelles (Xu and Dooner, 2006). However, evidence for localization of the full-length APT1 39 protein in stably transformed plants is lacking and thus these observations are not very informative. In order to directly elucidate the subcellular localization of the SAB protein, we attempted to clone the SAB gene with the aim of fusing it to fluorescent proteins and expressing the resulting constructs in plants. Unfortunately, the extremely large size of the SAB genomic region (about 20 kb including the predicted regulatory sequences) hampered the success of conventional molecular cloning. We thus adopted the alternative strategy of recombination-mediated genetic engineering (recombineering; Zhou et al., 2011) to fuse SAB with three tandem copies of the gene encoding the bright fluorescent protein Ypet and transformed plants with the tagged SAB within its genomic context. This allowed SAB expression under the control of its endogenous regulatory sequences, circumventing problems related to gene overexpression and providing a reliable read-out of its endogenous expression pattern. In order to further identify any possible effect of the fluorescent tag on localization of the SAB protein, we introduced the tag at either end of the gene. Both SAB-3xYpet and 3xYpet-SAB were expressed at very low levels in all cell types of the root meristem and in elongating and differentiating epidermal cells (Paper I, Fig. 8h-l and Fig. 9a,b,v,w). The protein fusions localized to the plasma membrane, to unidentified endomembrane compartments and to the nuclear envelope in interphase cells, whereas during cell division they associated closely with PPBs and mitotic spindles and labelled the cell plate and plasma membranes of the daughter cells (Paper I, Fig. 9a-u). These findings are in agreement with the prediction that SAB could be a membrane-associated or transmembrane protein, and disprove the hypotheses of a secreted or primarily Golgi-localized protein. Additionally, the observation that Ypet fusions at the N- and C-terminus display similar fluorescence intensity is strongly supportive of orientation of both ends of the SAB protein in the cytoplasm; fluorescent proteins are in fact sensitive to acidic conditions and the intensity of their emission would be reduced if they were exposed to the low pH of the root apoplastic space (Swarup et al., 2004). Expression of SAB in all cell types, including cells initiating root hairs, implies that SAB may act directly on polar root hair initiation in a cell-autonomous way. We however cannot exclude non-cellautonomous effects, and expression of SAB under the control of cell-type specific promoters or generation of mosaics of SAB expression would allow to determine whether SAB also contributes to planar polarity establishment in a non-cell-autonomous manner. It is interesting to note that SAB localization at the plasma membrane is strongly enriched in the apical and basal domains of root meristematic cells. This accumulation indicates the presence of a mechanism for polarization of SAB distribution immediately after cell division and supports an involvement of SAB in the generation or 40 maintenance of apical-basal polarity biases in root cells. Finally, SAB is not evenly spread on membranes but rather preferentially confined to patches, dots with relatively stronger signal intensity. Such a pattern suggests either the presence of a yet unknown mechanism for SAB accumulation in specific membrane compartments or association of SAB to discrete structures at the cell cortex, although their identity remains elusive. Uncovering the nature of SAB-labelled endomembrane compartments, clarifying mechanisms and function of SAB membrane localization and revealing the role of SAB in proximity to and in association with mitotic structures present challenges for future functional studies on SAB localization. SAB interaction with the CLASP gene SAB and CLASP interact genetically Since analysis of the SAB protein and its localization did not provide many hints about the role of SAB in plants, we set out to search for possible genetic interactors that could reveal SAB functions. Mutants defective in the MAP gene CLASP are the only ones so far characterized for aberrant trichome orientation and branching similar to sab mutants. Our analyses of sab;clasp double mutants revealed a strong effect on trichome branching and overall development of seedlings and plants, suggesting synergistic interaction between SAB and CLASP (Paper I, Fig. 5a,b and Fig. S3a-c). This prompted us to investigate whether CLASP could also be involved in planar polarity of root hair positioning. Indeed, we uncovered that loss of CLASP function results in a basal shift of both hair initiation and polar ROP positioning (Paper I, Fig. 5c-h). SAB also has a role in PPB orientation The roles of CLASP in plants comprise microtubule stabilization (Ambrose et al., 2007; Kirik et al., 2007) and organization of cortical microtubule arrays (Ambrose and Wasteneys, 2008; Ambrose et al., 2011), including orientation of the PPB microtubules at the cortex of dividing cells (Ambrose et al., 2007). We thus employed tubulin immunolocalization and live imaging to analyze PPBs and other mitotic structures in sab and clasp mutants. Our observations revealed that kuq- and sab-5-mutant PPBs were often misaligned with respect to the cell axis, and that this defect was accompanied by aberrant orientation of mitotic spindles, phragmoplasts and cell division planes (Paper I, Fig. 6a-q and Fig. S3d-h). Live imaging allowed us to assess that mutation of SAB affected positioning of the PPB in the initial phases of cell division, and that the orientation defects were then maintained in the same cell until formation of the cell division plane (Paper I, Fig. 6c,d and 41 Supplementary Movies 1 and 2). These findings unveiled a role for SAB in cell division already at the stage of PPB formation, and suggested that SAB acts on the initial cell division plane alignment and not on maintenance of this alignment during mitosis and cytokinesis. PPB formation is driven by restriction of interphase cortical microtubule bundles to the cell equator, and its correct set up requires a group of proteins that localize to the cell division plane during prophase (Dhonukshe and Gadella, 2003; Vos et al., 2004; Rasmussen et al., 2013). Mutations in the TON2/FASS or TON1A/TON1B genes abolish PPB formation and determine strongly altered cellular organization and positioning of cell division planes (Camilleri et al., 2002; Azimzadeh et al., 2008). SAB function is not equivalent to that of TON2/FASS or TON1A/TON1B, because mutation of SAB does not impair assembly of the PPB but rather its orientation. Similarly, SAB is not analogous to proteins that define the cortical division site like TAN1 or RanGAP1: these markers localize to the PPB and stay at the CDS after PPB disassembly to guide growth of the cell plate towards the parental plasma membrane (Cleary and Smith, 1998; Walker et al., 2007; Xu et al., 2008). First of all SAB does not localize to the PPB, and secondly misoriented cell division planes in sab mutants are the consequence of misoriented PPBs, rather than being caused by lack of guiding signals from correctly aligned PPBs. On the other hand, mutants defective in the gene encoding the MAP katanin display a phenotype more reminiscent of the one observed in sab and clasp mutants: fra2 and lue1 mutants of katanin have correctly oriented PPBs but a high frequency of oblique spindles and phragmoplasts, resulting in oblique cell walls and disordered cell organization (Panteris et al., 2011). Interestingly, the proportion of misaligned spindles in katanin mutants is much higher than the proportion of misoriented phragmoplasts, supporting the notion that phragmoplasts are often able to track back to the CDS and align to it (Rasmussen et al., 2013). No correction of phragmoplast orientation takes place in sab mutants though, confirming that cell division planes are misaligned already at the stage of CDS establishment. Similar to CLASP, katanin colocalizes with PPB, spindle and phragmoplast microtubules (Panteris et al., 2011) and thus differs from SAB also in this regard. Nonetheless, resemblance of sab phenotypes to phenotypes of clasp and katanin mutants points towards a possible role also for SAB in microtubule modification or organization. SAB is not involved in mitotic progression Live imaging of dividing cells additionally offered us the opportunity to compare the timing of mitosis and cytokinesis between wild type and sab-5 mutants. Advancement through mitotic phases was similar in all the cell 42 division events that we followed, in both wild-type and sab-5 cells, and cytokinesis was complete approximately 70 minutes after break down of the nuclear envelope (Paper I, Fig. 6c,d and Supplementary Movies 1 and 2). This result indicates that SAB is not likely to be involved in mitotic progression, although it does not exclude a role for SAB at other stages of the cell cycle. Complex interaction of SAB and CLASP in the control of distinct processes When comparing the defects of sab-5 and clasp-1 in seedling development, in trichome branching, and in misorientation of PPB, spindle and phragmoplast, sab-5 mutants were always more strongly affected (Paper I, Fig. 5a,b and Fig. 6o-q). Mutations in the two genes determined different phenotypes with respect to planar polarity of root hairs and ROPs, with sab5 causing both an apical and a basal shift of hair and ROP positioning and clasp-1 inducing an exclusively basal shift (Paper I, Fig. 3c and Fig. S2a-d and Fig. 5c-h). Combining the two mutations led to strongly enhanced defects in overall growth and in trichome branching (Paper I, Fig. 5a,b and Fig. S3a-c), indicating synergistic action. However, with respect to ROP positioning and orientation of mitotic structures the sab-5;clasp-1 double mutants did not statistically differ from sab-5 (Paper I, Fig. 5k and Fig. 6oq), suggesting that SAB is epistatic to CLASP in the control of these processes. These findings are sign of a complex interaction between SAB and CLASP, which may combine their functions differently in distinct pathways. Synergy between two genes is generally the product of non-linear interplay, whereas epistasis usually indicates that they act sequentially in one pathway. Non-linearity could imply that SAB and CLASP reciprocally feed on each other’s function, act together with additional common partners or form a protein complex to control plant development and trichome morphology. To date, we have no data supporting any of these scenarios and thus further studies are needed to assess the likelihood of each hypothesis. For instance, genetic interaction between SAB and CLASP does not strictly imply direct protein interaction: association between SAB and CLASP could only be confirmed through additional essays, which will however be complicated by the considerable size of both proteins. Different localization also does not suggest direct stable interaction, considering that SAB and CLASP, despite partial potential overlap (at cell edges, in regions at the periphery of mitotic spindles and phragmoplasts), overall display quite dissimilar distribution: CLASP accumulates on PPBs (Ambrose et al., 2007; Kirik et al., 2007) whereas SAB localized only in proximity to them (Paper I, Fig. S5); CLASP labels cortical microtubules (Ambrose et al., 2007; Kirik et al., 2007) whereas the membrane-associated SAB patches did not seem to colocalize 43 with microtubule bundles (Pietra, Gustavsson and Grebe, preliminary observations). In processes where the interaction between SAB and CLASP seemed linear, we studied whether SAB function influenced CLASP expression and vice versa, in order to determine which one of the two genes could act further upstream. GFP-CLASP accumulated on aberrant PPBs in sab-5 mutants, indicating that its localization to mitotic structures was affected by SABdependent orientation of the cortical microtubules in preprophase (Paper I, Fig. 6s,t). On the contrary, SAB-3xYpet and 3xYpet-SAB localization were completely unaltered in clasp-1 mutant seedlings (Pietra and Grebe, data not shown). These results strongly support a role for SAB upstream of CLASP in orientation of cell division planes and planar polarity of the root epidermis. SAB may act through CLASP to control the cortical microtubule array, while playing a broader role in these processes via interaction with other potential partners. SAB function in the organization of cortical microtubules Role of SAB in elongating epidermal cells Considering the role of SAB in orientation of microtubule arrays during mitosis, we decided to investigate whether this gene also has an effect at later stages of differentiation, when cells elongate and start to differentiate a hair bulge. Right after division in the meristem, root epidermal cells have their microtubules distributed randomly at the cell cortex, without a specific arrangement (Ehrhardt and Shaw, 2006). In order to drive anisotropic cell expansion, microtubules organize into cortical bundles that orient transversally to the root axis and to the direction of cell growth (Baskin, 2001). They do so by exploring the two-dimensional cortical environment and modifying their alignment upon encounters with other microtubules at sharp angles (Dixit and Cyr, 2004), and by interacting with MAPs that can finely tune the organization, stability and orientation of the array (Ehrhardt, 2008). When we observed sab-5 epidermal cells that were either beginning to expand or rapidly elongating, we noticed that their cortical microtubule lattice was not regularly transversal as in wild-type cells but rather appeared irregular, more crisscrossed and less aligned to the root apical-basal axis (Paper I, Fig. S4a-c and Fig. 7a-c). Microtubule growth was bidirectional in both wild-type and sab-5 cells (Paper I, Supplementary Movies 3 and 4); however, the orientation of the cortical array of mutant cells did not seem suited for supporting correct directional cell expansion. This could be one of the causes of the reduced final length of sab epidermal cells and overall small size of sab plants (Paper I, Fig. 3a,b). A possible explanation for the aberrant 44 alignment of cortical microtubules in sab elongating cells is that SAB is involved in controlling the dynamics, interactions or bundling of microtubules. Thus, lack of SAB function may not allow formation of the bundles needed for reorganizing the cortical array after cell division. Such a remodelling of cortical microtubule arrays through activity of a protein that affects microtubule interactions would closely resemble the recently described mechanism that is based on the severing protein katanin (Lindeboom et al., 2013). Alternatively, SAB could function in cell-wide arrangement and directionality of microtubule bundles and hence in sab mutants the bundles would not be uniformly aligned and capable of driving expansion in a single direction. Bipolar growth of longitudinal microtubule arrays At later stages, when epidermal cells have fully elongated and cease growth, cortical microtubules reorient from transversal to oblique, and finally form a more longitudinal array (Granger and Cyr, 2001). Longitudinal microtubule arrays have been shown to be strongly bipolarized in hypocotyl cells, both prior to and after bursts of growth (Sambade et al., 2012). We observed a similar bipolar growth in root epidermal cells, with microtubule plus ends elongating towards the shoot apex in the apical part of the cell and towards the root tip in the basal part of the cell (Paper I, Fig. 7d-f and Supplementary Movie 5). A functional explanation for this dynamic behaviour of cortical microtubules is still lacking; however, bipolar growth was maintained also in sab-5 cells, and the main difference compared to the wild type was rather in the longitudinal orientation of microtubules and in their alignment to the growth axis (Paper I, Fig. 7g and Supplementary Movie 6). SAB therefore is not likely to be involved in directionality of microtubule growth, but instead seems to modify alignment of their growth pattern, in meristematic as well as in elongated cells. Cortical microtubule organization at root hair initiation sites Shortly before completing elongation, trichoblasts begin differentiating root hairs on their surface. Bao et al. (2001) reported that reducing α-tubulin expression induces the formation of ectopic root hairs and multiple hairs on individual cells, thus indicating that microtubule dynamics could have a role in establishment of root hair initiation sites. Indeed microtubules have been shown to be disorganized in the apical region of emerging root hairs (Van Bruaene et al., 2004); however, so far no specific microtubule organization seemed to be required for root hair initiation. Our observations of the microtubule cytoskeleton in combination with the root hair initiation site marker PIP5K3-YFP uncovered the presence in 45% of the observed cells of a 45 characteristic radial array of cortical microtubules labelling the future site of hair outgrowth prior to any morphological sign of bulge formation (Paper I, Fig.7h-l and Fig. S4d,e). Radial organization of microtubules has already been described for actively elongating tip-growing cells like moss protonemata and conifer pollen tubes (Doonan et al., 1985; Anderhag et al., 2000). Moreover, microtubules are known to play a role during later phases of root hair tip-growth, as they control growth direction and avoid branching (Bibikova et al., 1999). However, our finding identified specific microtubule reorganization at root hair initiation sites at an earlier stage than described before. The reason why not all of the analyzed cells presented this radial array of microtubules still needs to be investigated, nonetheless one possible explanation is that the radial rearrangement is transient and can only be observed in a limited number of cells unless the hair initiation process is entirely followed over time. Alternatively, the radial array might not be strictly required for root hair initiation, and cells could begin hair formation regardless of this arrangement. Similarly, longitudinal orientation of microtubules might be frequent but not necessary for the earliest stages of hair initiation: 4/71 cells expressing PIP5K3-YFP still had transversally oriented microtubules. This seems to contradict earlier findings stating that microtubule disorganization is necessary for root hair outgrowth in Arabidopsis (Van Bruaene et al., 2004); however, previous studies have employed the formation of a bulge as morphological sign of hair initiation, while we observed the accumulation of a fluorescent marker at the hair initiation site and thus have captured earlier stages of hair formation. When analyzing the organization of cortical microtubules in sab-5 mutant trichoblasts about to initiate a root hair, we could detect sites with radially arranged microtubules that corresponded to PIP5K3-YFP patches (Paper I, Fig. 7m-o). Therefore the mechanisms for generating radial arrays seemed to still be functional in the absence of SAB. However, we could not find any cell where microtubules displayed a clear and unique radial arrangement at the site of hair initiation (Paper I, Fig. 7m-o and Fig. S4h-k). This result indicated that SAB activity could be necessary for directing and limiting the degree of microtubule organization and the formation of radial arrays marking sites of polar root hair initiation. Determining the functional role of microtubule rearrangement at hair initiation sites and the link between radial arrays and establishment of polarity on the lateral membrane of root epidermal cells could give a more precise hint towards SAB function in the planar polarity pathway. 46 Actin is an additional player in the planar polarity pathway ACTIN7 genetically interacts with SAB After finding out that SAB genetically interacted with a MAP gene in planar polarity establishment and affected the orientation and organization of microtubules during cell division and elongation, we decided to investigate whether SAB also interacted with the actin cytoskeleton, which is known to regulate cell shape, architecture and polarity. To this end, we generated double mutants between sab and a mutant defective in the gene encoding the vegetative actin ACTIN7 (ACT7), and obtained seedlings with much shorter, radially swollen roots and delayed growth compared to each of the single mutants (Fig. 10a,b). These severe defects suggested that SAB and ACT7 may interact synergistically in determining overall seedling development, and prompted us to analyze whether ACT7 was also involved in planar polarity of the root epidermis. We thus immunolocalized ROPs, the earliest markers of planar root hair polarity, in trichoblasts of sab-5, act7-6 and sab-5;act7-6 mutants. Both sab-5 and act7-6 single mutants displayed a basal and an apical shift in ROP localization compared to wild-type seedlings (Paper I, Fig. S2a-d and Paper II, Fig. 5a,c,f). However, the apical shift was significantly more pronounced in sab-5;act7-6 double mutants (Fig. 10c). These results indicated that SAB and ACT7 are likely to have a combined effect in the planar polarity pathway, and uncovered a potential, previously unknown, interaction of a SAB-related protein with the actin cytoskeleton. This finding led us to investigate the contribution of the actin cytoskeleton to planar polarity of root hair positioning in more detail. Figure 10. SAB and ACT7 interact genetically to determine seedling development and mediate polar ROP positioning in trichoblasts. (a,b) Strongly impaired development of 5-day-old sab-5;act7-6 seedlings with respect to act7-6 and sab-5 single mutants. Note, extremely short and radially swollen roots. (c) Quantitative analysis of polar positioning of ROP protein patches along the apical-basal axis of epidermal cells in act7-6, sab-5 and sab-5;act7-6 seedlings. Number of cells in classes (frequency) showing ROP positions between basal (0) and apical (1) ends is shown (**P = 0.009 act7-6 versus sab-5, **P = 0.002 sab-5 versus sab-5;act7-6, **P = 0.000 act7-6 versus sab-5;act7-6 by Kolmogorov-Smirnov test, n = 50 cells from 26 to 29 roots per genotype). Data for act7-6 are the same presented in Paper II Fig. 5f. For experimental procedures, see Paper II. Scale bar, 2 mm. 47 The actin cytoskeleton affects planar polarity of root hair positioning Actin is essential for tip growth of root hairs, and thick actin bundles as well as dense F-actin meshworks have been localized to the central region and just below the tip of elongating hairs (Rounds and Bezanilla, 2013; Baluška et al., 2000). However, we could observe F-actin strands already during the first morphological signs of hair initiation and even in proximity to sites of hair emergence prior to formation of a bulge (Paper II, Fig. 1j,k). These localization data support a role for actin organization in early events of root hair initiation and potentially in selection of the polar site of hair outgrowth on the lateral membrane of trichoblasts. In order to examine the effects of interference with the actin cytoskeleton, we analyzed the planar polarity phenotypes of mutants defective in various vegetatively-expressed actins. Previous experiments have shown that mutants of the ACT2 gene display a weak apical shift of root hair initiation compared to wild type (Ringli et al., 2002), and this was confirmed by our analysis of hair positioning in act2 mutants (Paper II, Fig. 1a,b,g). Stronger defects were though uncovered in act7 mutant seedlings, both in distribution of hair initiation sites (Paper II, Fig. 1a,c,g and Fig. S1a,b) and in positioning of ROP patches on the trichoblast lateral membrane (Paper II, Fig. 5a,c,f): in addition to a pronounced apical shift, act7 seedlings also presented a basal shift of hair and ROP positioning, when compared to wild type. These findings, combined with the lack of effects in actin8 (act8) loss-of-function mutants (Paper II, Fig. S1d), suggest that ACT7 is the actin isoform that contributes most strongly to planar polar establishment. Nevertheless, ACT2 also plays a role in this process, as demonstrated by the enhanced root hair positioning defects of act7;act2 double mutants with respect to both single mutants (Paper II, Fig.1b-d,h). Furthermore, the mislocalization of ROPs in act7 mutants indicates that the actin cytoskeleton is likely to have an effect upstream of ROP GTPases in planar polarity of root hairs. This scenario is intriguing because in other processes, like pavement cell morphogenesis and distribution of auxin transporters, the actin cytoskeleton has been suggested as a downstream effector of signals mediated by ROP molecular switches (Fu et al., 2005; Lin et al., 2012). A tempting hypothesis is that the effect of the actin cytoskeleton on ROP localization is part of a feedback meant to reinforce the formation of polar membrane domains by recruiting additional polarity determinants, similar to how polarity establishment is regulated in yeast cells (Piel and Tran, 2009; Marco et al., 2007; Valdez-Taubas and Pelham, 2003). Interference with actin organization may destabilize the set up of hair initiation sites and randomize the positioning of ROPs and eventually root hairs. The observation that mutation of an actin gene results in ROP misplacement is in apparent disagreement with the reported unaffected 48 polar ROP localization upon pharmacological disruption of the actin cytoskeleton (Molendijk et al., 2001). It is thus possible that the role of actin in polar ROP positioning is indirect. Alternatively, interfering with actin organization might not influence the formation of discrete ROP patches at the plasma membrane but only alter their polar positioning along the apicalbasal cell axis: no quantification of ROP distribution following drug treatment has in fact been provided in previous studies (Molendijk et al., 2001). AIP1-2 is an actin interactor in plants In an effort to clarify the mechanisms of actin control over planar polarity, we looked for interactors of plant actins by conducting a yeast two-hybrid screen. Yeast two-hybrid analyses indicated AIP1-2, the product of one of the two isoforms of the AIP1 gene in Arabidopsis, as a strong interactor of all vegetative actins and ACT1 (Paper II, Fig. 2a). Similar results were obtained with the product of the other AIP1 isoform, AIP1-1, and were confirmed in vitro by Glutathione-S-Transferase (GST) pull-down assays with ACT2 and ACT7 (Paper II, Fig. 2b,c and Fig. S2a-f). Significance of the interaction between AIP1-2 and ACT7 was further assessed in planta through analysis of act7;aip1-2 double mutants: germination was significantly reduced and plants that did germinate presented shoot morphological defects that were absent in both of the single mutants (Paper II, Fig. 3c-g and Fig. S3a-f). The synthetic lethality of act7;aip1-2 mutants is indicative of a strong synergistic effect of the two mutations and resembles the lethality that results from combining the yeast homolog of AIP1, whose mutation does not cause severe growth defects on its own, with mutations in ACT1 (Rodal et al., 1999). AIP1 is a modulator of ADF severing activity that was first isolated in yeast in a screen for Act1p interactors (Amberg et al., 1995). Consistent with the role of AIP1 in F-actin disassembly, mutants defective in AIP1 homologs display excessive actin stability and bundling in Drosophila, Physcomitrella patens, and Arabidopsis lines where both AIP1 genes are knocked-down by RNA interference (Ren et al., 2007; Augustine et al., 2011; Ketelaar et al., 2004). Compromised AIP1 activity causes dramatic developmental defects and strongly affects the elongation of tip growing cells in these organisms (Ren et al., 2007; Augustine et al., 2011; Ketelaar et al., 2004). Interestingly, mutants defective in the Drosophila AIP1 homolog Flare show aberrant polarity of wing hair positioning, suggesting that in flies actin remodelling is also involved in setting up planar polarity (Ren et al., 2007). In order to test if Arabidopsis AIP1-2 also had an effect on planar polarity establishment, we analyzed hair and ROP positioning in the root epidermis of aip1-2 mutants, and discovered that hair initiation as well as ROP patches were basally shifted with respect to wild type (Paper II, Fig. 4a,b,d and Fig. 49 5a,b,e and Fig. S4a). The difference in effect between the aip1-2 mutation, causing a basal shift, and mutations in actin genes, act2 determining an apical shift and act7 both an apical and a basal shift, may be attributed to the opposite results that the mutations have on the actin cytoskeleton: whereas aip1-2 likely determines stabilization and bundling of F-actin, although the probable subtlety of this phenotype does not allow detection of any bundles in aip1-2 seedlings, mutations in actin genes destabilize actin filaments and thus reduce the formation of bundles. Considering the effect we observed on plant survival and growth by combining aip1-2 with the act7 mutation, we further investigated whether AIP1-2 interacted synergistically with ACT7 also in planar polarity establishment. Strikingly, positioning of both root hairs and ROPs in act7;aip1-2 double mutants was not different from act7 single mutants (Paper II, Fig. 4e and Fig. 5c,d,g and Fig. S4c). This result indicated that in these processes act7 is epistatic to aip1-2, supporting the idea that ACT7 is the actin isoform providing the strongest contribution to planar polarity and that therefore F-actin stabilization does not affect root hair distribution in the absence of ACT7. AIP1-2 expression is restricted to hair-producing cells With the aim of verifying whether AIP1-2 could interact with actin filaments at root hair initiation sites, we analyzed the localization of a functional AIP12 fluorescent protein fusion. We observed prevalent AIP1-2 expression in the epidermal layer of roots and strong accumulation at future sites of hair emergence and especially in outgrowing hair bulges, where it coincided with an enrichment of F-actin (Paper II, Fig. 4f-h). This finding corroborated the interaction between AIP1-2 and actins, and supported the role of AIP1-2 in planar polarity of root hair positioning. Surprisingly, AIP1-2 expression in epidermal cells was specifically enriched in hair files and displayed a pattern of alternating stripes starting from the early root elongation zone (Paper II, Fig. 4f and Fig. 6a). To test if AIP1-2 enrichment was associated with the differentiation of root hair cells, we employed mutation of the CTR1 gene and treatments with ACC or synthetic auxins to induce the formation of ectopic root hairs. Both ethylene and auxin signalling stimulate root hair initiation (Schiefelbein, 2000), and either ctr1 mutation or application of exogenous ethylene are known to determine an increased proportion of roothair-forming cells by acting downstream of the patterning pathway (Dolan et al., 1994; Tanimoto et al., 1995; Masucci and Schiefelbein, 1996). Intriguingly, we observed enhanced ectopic AIP1-2 expression in cells in the non-hair position and in cells growing ectopic root hairs (Paper II, Fig. 6c-g). This finding demonstrates that heterogeneous expression of AIP1-2 in the root epidermis is affected by ethylene and auxin signalling. 50 Interplay of ethylene signalling and actin cytoskeleton in planar polarity In addition to controlling root hair initiation, ethylene is also one of the main actors in the pathway determining planar polarity of root hair positioning (Masucci and Schiefelbein, 1994; Fischer et al., 2006; Ikeda et al., 2009). The effect of ACC treatment and mutation of the CTR1 gene on AIP1-2 expression prompted us to further explore the genetic relationship between ethylene signalling and the actin cytoskeleton in polar initiation of root hairs. Mutation of the two vegetative actin genes ACT2 and ACT7 completely suppressed the hyperpolar phenotype of the ctr1btk mutant (Paper II, Fig. 1df,i), indicating that ACT2 and ACT7 are epistatic to CTR1 in root hair positioning. However, when combining the ctr1btk mutation with an aip1-2 mutant, the ctr1 phenotype was epistatic to aip1-2 and double mutant seedlings were indistinguishable from ctr1btk single mutants (Paper II, Fig. 7a-e). These findings are consistent with ACT7 being epistatic to AIP1-2 (Paper II, Fig. 4e and Fig. 5c,d,g and Fig. S4c), and could outline a scenario where ethylene can control F-actin stability and bundling in hair cells by affecting the expression of AIP1-2, while at the same time proper formation of actin filaments is necessary downstream of ethylene and actin organizing proteins to direct the correct placement of root hairs. Although ethylene signalling has long been known to affect organization of the microtubule cytoskeleton (Steen and Chadwick, 1981), our observations point out an additional genetic link with actin. No direct connection between ethylene signalling and actin is however demonstrated, and thus this effect could be indirect or mediated for instance by auxin synthesis and perception. Patterned expression of AIP1-2 depends on WER function Upon finding that ethylene and auxin affected the patterned expression of AIP1-2 in the root epidermis, we set out to further determine factors restricting AIP1-2 accumulation to root hair files. In the root epidermis, ethylene and auxin modulate gene expression and cell differentiation downstream of a network of transcription factors that translate positional signals into hair and non-hair cell patterning (Schiefelbein, 2000; Dolan, 2001). One of the central components of this patterning gene network is the Myb transcription factor WER (Lee and Schiefelbein, 1999). We thus investigated whether AIP1-2 expression was under the control of WER by analyzing AIP1-2 localization in the roots of wer mutant seedlings, which are defective in non-hair fate specification and thus form ectopic hairs on most of their non-hair cells. Strikingly, in wer mutants all epidermal cell files evenly expressed AIP1-2, indicating that preferential accumulation of this protein in hair files depends on WER function (Paper II, Fig. 6a-c). The reason for enrichment of AIP1-2 in hair cells is elusive, as no role has been 51 described so far for AIP1-2 downstream of WER. However, controlling the expression of an actin modulating protein like AIP1-2 could be a strategy to restrict the energy-consuming reorganization of the actin cytoskeleton to cells that are going to initiate outgrowth of a root hair. In Drosophila, expression of the core components of the frizzled planar polarity pathway and of the effectors that control reorganization of the actin cytoskeleton is homogeneous in all cells of the tissues whose polarity they control (Goodrich and Strutt, 2011). Together with RHD6 (Menand et al., 2007), AIP1-2 is therefore one of the few reported examples of patterned expression of a gene affecting planar polarity. Although no role has been described for AIP1-2 in patterning of the root epidermis, the finding that restriction of its expression depends on WER suggested interplay between planar polarity establishment and epidermal cell fate specification. We thus decided to further investigate common features between planar polarity formation and epidermal patterning. To this end, we started analyzing mutants of the planar polarity gene SAB for defects in patterning of the root epidermis. SAB has a role in root epidermal patterning sab mutants differentiate excessive ectopic root hairs The root epidermis of wild-type Arabidopsis seedlings is patterned by the alternation of longitudinal files of hair-forming and non-hair-forming cells, whose fate is determined by their position with respect to the underlying cortex: epidermal cells on top of two cortical cell files produce a hair and are considered in hair (H) position, while cells over a single cortical cell file do not differentiate a hair and are thus referred to as non-hair (N) cells (Dolan et al., 1994). In two mutant alleles of the SAB gene we observed a significantly higher number of hairs emerging from cells in the N position, causing the seedlings to display about four times more ectopic hairs than wild type and strongly suggesting that SAB has a role in root epidermal patterning (Paper III, Fig. 1a-c and Table 1). Ectopic hairs are characteristic of mutants defective in genes that promote non-hair fate within the patterning pathway, but are frequent also in mutant seedlings of the ECTOPIC ROOT HAIR 2/POM POM 1 (ERH2/POM1) and ECTOPIC ROOT HAIR 3 (ERH3) genes (Schneider et al., 1997). Roots of erh2/pom1 seedlings have radially swollen cortical cells similar to sab (Hauser et al., 1995; Benfey et al., 1993), and erh3 mutants are also deficient in anisotropic cell growth (Schneider et al., 1997; Uyttewaal et al., 2012), indicating that control of cell expansion could be involved in patterning of the root epidermis and may be one of the mechanisms through which SAB affects this process. Furthermore, ERH3 is the KTN1 katanin, which severs and reorganizes 52 microtubule arrays, contributing to formation of mitotic structures and orientation of cell division (Bichet et al., 2001; Panteris et al., 2011; Webb et al., 2002); the similarity with the role of SAB in microtubule rearrangement suggests that manipulation of microtubule dynamics and orientation may also have a function in determining cell fate of epidermal cells. Longitudinal cell divisions are not responsible for sab patterning defects Once cell fate has been established in the root meristem, hair and non-hair cells generally divide transversally to the root axis, and daughter cells assume the same fate as their parent. However, in rare instances cells divide longitudinally and generate daughter cells that initiate two new cell files that do not necessarily have the same fate (Berger et al., 1998b). Considering the role of SAB in orientation of mitotic structures and cell division planes, we investigated whether the increased amount of ectopic hairs in sab mutants could be caused by a greater number of longitudinal cell divisions and subsequent H-fate duplications. This was however not the case, as the number of cell file divisions in sab mutants was comparable to wild-type seedlings, indicating that SAB function in cell division orientation is not likely to play a role in patterning of the root epidermis (Paper III, Fig. 2). Moreover, in sab seedlings about 40% of the longitudinal divisions took place in N cells, whereas in wild type the file duplications initiated in most of the cases from H cells (Paper III, Fig. 2). It is difficult to point out a functional reason for the preferential occurrence of aberrant longitudinal divisions in cells with a specific fate. Nevertheless, selection of the files undergoing duplication seems to be under the control of a mechanism that requires SAB function. Interestingly, similar defects have been reported for mutants of genes belonging to the core cell fate patterning pathway (Berger et al., 1998b), and this result therefore supports the involvement of SAB in specification of epidermal cell fate. Destabilized expression of fate markers in the sab root epidermis We employed cell fate and differentiation markers to visualize the stability of fate selection within root epidermal cell files. We expected homogeneous expression of the markers among cells of the same lineage; however, we noticed that cells would sometimes switch fate with respect to the file they belonged to, determining a disparity between their position over the cortex and the marker they expressed. We observed sporadic fate switches in epidermal cell files of wild-type roots carrying both hair differentiation and non-hair fate markers. Strikingly though, all of the analyzed markers revealed substantially more frequent fate alternations in sab-mutant seedlings than in wild type, suggesting that SAB function is required for 53 uniform fate specification within files of root epidermal cells (Paper III, Fig. 3 and Fig. S1). One of the employed reporter constructs, pWER:GFP, indicates expression of a core component of the patterning pathway and positive regulator of non-hair cell fate (Lee and Schiefelbein, 1999). Considering the increase of ectopic root hairs in sab mutants (Paper III, Fig. 1a-c and Table 2), SAB may be expected to negatively affect WER expression in epidermal cells. However, we could not find any preference in the cell fate of epidermal files undergoing switches of pWER:GFP expression, indicating that the sab mutation apparently has no effect on the overall number of cells expressing WER and adopting the N fate. Further studies will thus be needed to clarify how SAB acts on WER expression and to explain how lack of SAB function translates into root hair patterning defects, since we currently cannot exclude an additional effect of the core patterning pathway on SAB. The unstable expression of fate markers suggests that SAB controls fate maintenance within cell files rather than being involved in the specification of one or the other fate, and that the role of SAB is therefore in pattern stability within the root and not in selection of single cell fate. SAB affects epidermal cell fate upstream of the core patterning pathway We further explored the function of SAB in epidermal fate specification by analyzing genetic interaction between SAB and genes of the core patterning pathway. Hair patterning defects of mutants lacking function of either the WER or CPC genes, which promote non-hair and hair fate respectively, were epistatic to the phenotype of sab mutants, and led to either very hairy or hairless double mutant seedlings (Paper III, Fig. 4 and Table 3). These results suggested that SAB acts in the same pathway as WER and CPC and may function upstream of them. The latter hypothesis is supported by the observation that mutation of SAB can affect the expression of WER and of genes further downstream in the cell fate specification and differentiation pathway (Paper III, Fig. 3 and Fig. S1). Additionally, the finding that mutation of SAB in the cpc background determines hairless seedlings indicates that activity of the non-hair-fate-promoting complex is not impaired in sab mutants and hence SAB is not likely to be one of its components. The function of SAB upstream of the core epidermal patterning pathway further rules out that the excessive amount of ectopic root hairs in sab mutants may be a consequence of enhanced sensitivity to phosphorus (P) starvation or an effect of ethylene signalling. Recent studies have demonstrated that sab mutants are highly sensitive to P starvation (Yu et al., 2012), and this physiological stress is known to promote the formation of root hairs (Schmidt and Schikora, 2001). Moreover, SAB seems to counteract 54 ethylene signalling, which is known to stimulate root hair development (Aeschbacher et al., 1995; Yu et al., 2012; Tanimoto et al., 1995). However, both P and ethylene have been shown to affect differentiation of epidermal cells downstream of the core patterning gene network (Müller and Schmidt, 2004; Masucci and Schiefelbein, 1996). Placing SAB upstream of WER and CPC action excludes that the patterning defects of sab mutants depend on the effect of SAB on P and ethylene signalling, and further supports a role for SAB upstream of the pathway that controls patterning of epidermal cell fate. SAB does not control patterning of leaf trichomes After discovering that SAB is involved in root epidermal patterning, we investigated whether this gene does also have a role in specification of leaf trichomes, whose regular spacing on the leaf surface is controlled by a gene network that employs components very similar to the ones determining root epidermal fate (Hülskamp, 2004; Ishida et al., 2008). To calculate the proportion of leaf epidermal cells that differentiate into trichomes, we quantified the number of trichomes on leaves of wild-type and sab-5 plants and compensated for differences in leaf size and cell density between the two genotypes. Surprisingly, values were very similar for wild-type and mutant leaves, suggesting that SAB does not affect trichome density and that its role in epidermal cell fate specification is likely to be restricted to the root (Paper III, Fig. 1d-g and Table 2). This finding may be explained in several ways: first, SAB could be a rootspecific component of the cell fate specification pathway, similar to the MYB transcription factor WER that is only present in the root and hypocotyl (Lee and Schiefelbein, 1999). SAB function might not be required for patterning of the leaf epidermis, or may be executed by another shoot-specific component. However, considering the role that SAB has in planar polarity and branching of leaf trichomes (Paper I, Fig. 1h-m and Fig. 5a and Fig. S2e,f) and the strong defects in leaf size observed in sab mutants (Paper I, Fig. 8a,b and Paper III, Fig. 1d,f), it is unlikely that the SAB protein is absent from leaf tissues. It could however be that SAB function in the shoot is restricted to other roles and that SAB in the leaf is not part of the patterning pathway. Alternatively, there could be a shoot-specific redundant factor with functions similar to SAB in leaf epidermal patterning. This factor would be able to compensate for SAB loss-of-function in leaves but not in roots, determining the absence of a visible shoot patterning phenotype in sab mutants. Finally, the SAB gene might play a role that is independent of the core network of patterning transcription factors and is required only for epidermal patterning of roots. The main difference within epidermal fate specification is that root cells receive positional cues from the underlying cortical layer, while patterning of leaf cells is only driven by stochastic 55 fluctuations and cell-to-cell interactions (Balkunde et al., 2010; Schiefelbein et al., 2009). If indeed SAB were implicated in generating, perceiving, and/or transmitting the positional signals from the cortex, there would be no role for it in patterning of the leaf epidermis, regardless of its expression in aerial tissues. SAB may act on transduction of patterning cues from the cortex Strikingly, sab mutants display various similarities to mutants of genes that are involved in radial communication between the root cortex and epidermis. For instance, mutants defective in either the SCM or JKD gene show an increase in the number of ectopic root hairs similar to sab mutants (Kwak et al., 2005; Hassan et al., 2010; Paper III, Fig. 1a-c and Table 1). Additionally, scm and jkd mutants also have a higher amount of non-hair cells in the H position (Kwak et al., 2005; Hassan et al., 2010). Although in sab mutants the number of H cells not producing a hair was not significantly different from wild type, thorough observation revealed that both analyzed sab alleles also had a subtle defect in specification of the H fate (Paper III, Table 1). Furthermore, expression of two non-hair fate markers in the root epidermis is unstable and randomized in scm and jkd, similar to what we observed in sab seedlings (Kwak et al., 2005; Hassan et al., 2010; Paper III, Fig. 3). Finally, the effect of the wer and cpc mutations on scm seedlings strongly resembles what we observed for sab;wer and sab;cpc in that the epidermal patterning of scm;wer and scm;cpc double mutants is indistinguishable from the single wer and cpc mutants (Kwak and Schiefelbein, 2007; Paper III, Table 3). These considerations support the idea that, similar to SCM and JKD, SAB functions in stabilizing fate acquisition upstream of the core patterning pathway. In the hypothetical scenario that, like SCM and JKD, SAB may be involved in the interaction between root epidermis and cortex, interesting considerations could come from comparison of the localization of the three proteins. JKD, which modulates root hair specification upstream of the patterning pathway, is not present in epidermal cells and operates non-cellautonomously from the cortex (Hassan et al., 2010). In contrast, both SCM and SAB are expressed in all root cell types (Kwak et al., 2005; Paper I, Fig. 8i), suggesting that the role of SAB might be more similar to that of SCM. However, neither scm nor jkd mutants have defects in the density of root hairs on their epidermis (Kwak et al., 2005; Hassan et al., 2010), while sab seedlings display an overall increased proportion of hair-bearing cells (Paper III, Fig. 1a-c and Table 1). Therefore, even on the assumption that SAB may have a position analogous to either SCM or JKD in the pathway, its role and mechanism of action are likely to differ markedly. 56 Conclusions and Future Perspectives Similar mechanisms for establishment of polarity are present in all eukaryotic systems. They involve an initial polarity cue, polar accumulation of signalling complexes and, as a last step, organization of the cytoskeleton to orient cells and their structures. Here we have identified SABRE as a novel player in the Arabidopsis planar polarity pathway and have shown that this gene also functions in organization of cortical microtubules, orientation of mitotic structures, and patterning of the root epidermis. Genetic interaction with CLASP and ACT7 supports an involvement of SAB in arrangement of microtubule arrays and in actin function. Accordingly, SAB acts on planar polarity upstream of ROPs, which have been shown to control organization of both microtubules and F-actin in other processes (Fu et al., 2005; Fu et al., 2009; Lin et al., 2013). Actin and its organization have been suggested to play a role in planar polarity (Ringli et al., 2002; Ren et al., 2007); we further characterize their contribution, identify new interactions and reveal a connection with the pathway for epidermal cell patterning. On the other hand, our findings of specific cortical microtubule arrays at root hair initiation sites, of planar polarity defects in clasp mutants, and of roles of the SAB planar polarity gene in microtubule organization strongly encourage further studies on the involvement of the microtubule cytoskeleton during coordination of planar polarity. Treatment with microtubule inhibitors and mutations in MAP genes or actin have been shown to induce misorientation of mitotic structures or cell division planes resembling those we observed for sab mutants (Baskin et al., 2004; Ambrose et al., 2007; Panteris et al., 2011; Gilliland et al., 2003). This resemblance is further indicative of a likely role for SAB in connection with both cytoskeletal arrays. In addition, the phenotypic similarities between sab and act7, which in contrast to most published planar polarity mutants display both an apical and a basal shift of root hair distribution, suggest an analogous mechanism of SAB and ACT7 action on root hair positioning. Crosstalk between microtubule and actin dynamics in Arabidopsis has already been demonstrated (Sampathkumar et al., 2011), thus it is reasonable to assume that SAB could have an effect on both arrays and to imagine a strong interplay between microtubules and F-actin in both planar polarity and epidermal patterning. Additional studies on the functional interaction between SAB and components or modulators of the microtubule and actin cytoskeleton may help understanding the exact role of SAB in organization of these arrays. The large size of the SAB protein makes it a good candidate for acting as an organizing hub or scaffold for cytoskeletal structures. This hypothesis is intriguing also in light of the localization of SAB at the cell cortex and in 57 proximity to mitotic arrays. A structural role for SAB might also explain the partial conservation of its sequence, as its function and interactors would be similar among distantly related species. Mapping of functional protein domains could help to elucidate roles and mechanisms of action of the SAB protein. These would additionally be clarified by uncovering potential SAB interactors; so far the yeast two-hybrid screens we conducted with five different SAB bait proteins including the full-length sequence have not identified any candidate for direct interaction with SAB (Pietra and Grebe, unpublished results), but biochemical methods such as coimmunoprecipitation might prove more successful in the future. Mechanisms of SAB action as well as its effect on microtubule and actin organization at the molecular level are still unknown, and future studies may reveal whether the SAB protein can directly bind to microtubules or F-actin. An effect of SAB on microtubule organization exclusively through CLASP would not explain the synergistic interaction between SAB and CLASP and the stronger phenotypes of sab compared to clasp mutants. Therefore, SAB may either affect microtubule arrays directly or through genetic interaction with other modulators of their dynamics. A direct role for SAB on microtubule dynamics would be intriguing, considering the analogies between SAB action and the microtubule severing protein katanin, which similarly to SAB is widely conserved and affects cell division orientation, anisotropic cell expansion and epidermal patterning (Panteris et al., 2011; Schneider et al., 1997; Webb et al., 2002). Strikingly, despite its local effect on individual microtubules, katanin is able to coordinate global organization of cortical arrays within an entire cell, in events such as response to mechanical stress or rapid reorientation of cortical microtubule arrays upon light perception (Uyttewaal et al., 2012; Lindeboom et al., 2013). Analogously, SAB could act on microtubule dynamics to direct cell-wide cytoskeletal organization. Taken together, our findings identify SAB as a novel planar polarity and epidermal patterning player involved in organization of microtubule arrays and actin function. This study opens the way for understanding the function of SAB-related proteins in eukaryotic organisms and stimulates further analyses on the roles of the microtubule and actin cytoskeleton in planar polarity and epidermal cell fate specification. 58 Acknowledgements Thank you to Prof. Dr. Markus Grebe for conceiving this project, supervising me during my doctoral studies, and for critically reading and commenting on this thesis. I am grateful to all the past and present group members for fruitful discussions, suggestions, and nice times during and after work. Thank you to Yoshi who initiated the work on kuq and guided me in the lab when I got started, and a big thank you to Anna for all her work on the SAB project, and for swedish translations. Thank you to the three incredibly talented master students I had the pleasure to work with, who allowed me to achieve results that would not have been possible had I been working on my own. A big thank you to Jose and Anna and all the members of the AlonsoStepanova lab, who gave me an amazingly warm welcome in Raleigh and made my stay there so enjoyable. Thank you to all my office mates for their tolerance during the long unexpected meeting, and to Kriszta and Carole in particular for the good mood and laughters. Thanks a lot to all those helping UPSC to run smoothly and efficiently and to those that keep it such a lively and stimulating environment. Thanks to Benni for the help with practical issues concerning the thesis and for all the motivating messages, and to Paola for drawing Figure 8. A great thank you to my family and friends, for their constant support and affection, even from a distance. Thanks to Matilda for her patience when I couldn’t dedicate to her the time she deserved, and for inspiring discussions during the preparation of this thesis. And the biggest thank you to Anna, for always helping me to put things in perspective and see the real priorities, for the butt kicking to help me writing, and for always being convinced that I would make it, even when I didn’t believe it myself. 59 References Adler, P.N. (2002). Planar Signaling and Morphogenesis in Drosophila. Dev. Cell 2, 525–535. Adler, P.N., Krasnow, R.E., and Liu, J. (1997). Tissue polarity points from cells that have higher Frizzled levels towards cells that have lower Frizzled levels. Curr. Biol. 7, 940–949. Aeschbacher, R.A., Hauser, M.-T., Feldmann, K.A., and Benfey, P.N. (1995). The SABRE gene is required for normal cell expansion in Arabidopsis. Genes Dev. 9, 330–340. Akhmanova, A., Hoogenraad, C.C., Drabek, K., Stepanova, T., Dortland, B., Verkerk, T., Vermeulen, W., Burgering, B.M., De Zeeuw, C.I., Grosveld, F., et al. (2001). CLASPs Are CLIP-115 and -170 Associating Proteins Involved in the Regional Regulation of Microtubule Dynamics in Motile Fibroblasts. Cell 104, 923–935. Al-Bassam, J., Kim, H., Brouhard, G., van Oijen, A., Harrison, S.C., and Chang, F. (2010). CLASP Promotes Microtubule Rescue by Recruiting Tubulin Dimers to the Microtubule. Dev. Cell 19, 245–258. Allwood, E.G., Anthony, R.G., Smertenko, A.P., Reichelt, S., Drobak, B.K., Doonan, J.H., Weeds, A.G., and Hussey, P.J. (2002). Regulation of the Pollen-Specific Actin-Depolymerizing Factor LlADF1. Plant Cell 14, 2915–2927. Amberg, D.C., Basart, E., and Botstein, D. (1995). Defining protein interactions with yeast actin in vivo. Nat. Struct. Biol. 2, 28–35. Ambrose, C., and Wasteneys, G.O. (2008). CLASP Modulates Microtubule-Cortex Interaction during Self-Organization of Acentrosomal Microtubules. Mol. Biol. Cell 19, 4730–4737. Ambrose, C., Shoji, T., Kotzer, A.M., Pighin, J.A., and Wasteneys, G.O. (2007). The Arabidopsis CLASP Gene Encodes a Microtubule-Associated Protein Involved in Cell Expansion and Division. Plant Cell 19, 2763–2775. Ambrose, C., Allard, J.F., Cytrynbaum, E.N., and Wasteneys, G.O. (2011). A CLASP-modulated cell edge barrier mechanism drives cell-wide cortical microtubule organization in Arabidopsis. Nat. Commun. 2. Ambrose, C., Ruan, Y., Gardiner, J.C., Tamblyn, L.M., Catching, A., Kirik, V., Marc, J., Overall, R., and Wasteneys, G.O. (2013). CLASP Interacts with Sorting Nexin 1 to Link Microtubules and Auxin Transport via PIN2 Recycling in Arabidopsis thaliana. Dev. Cell 24, 649–659. An, L., Zhou, Z., Yan, A., and Gan, Y. (2011). Progress on trichome development regulated by phytohormone signaling. Plant Signal. Behav. 6, 1959–1962. Anderhag, P., Hepler, P.K., and Lazzaro, M.D. (2000). Microtubules and microfilaments are both responsible for pollen tube elongation in the conifer Picea abies (Norway spruce). Protoplasma 214, 141–157. Augustine, R.C., Pattavina, K.A., Tüzel, E., Vidali, L., and Bezanilla, M. (2011). Actin Interacting Protein1 and Actin Depolymerizing Factor Drive Rapid Actin Dynamics in Physcomitrella patens. Plant Cell 23, 3696–3710. Azimzadeh, J., Nacry, P., Christodoulidou, A., Drevensek, S., Camilleri, C., Amiour, N., Parcy, F., Pastuglia, M., and Bouchez, D. (2008). Arabidopsis TONNEAU1 Proteins Are Essential for Preprophase Band Formation and Interact with Centrin. Plant Cell 20, 2146–2159. Balkunde, R., Pesch, M., and Hülskamp, M. (2010). Trichome patterning in Arabidopsis thaliana from genetic to molecular models. Curr. Top. Dev. Biol. 91, 299–321. Balkunde, R., Bouyer, D., and Hülskamp, M. (2011). Nuclear trapping by GL3 controls intercellular transport and redistribution of TTG1 protein in Arabidopsis. Development 138, 5039–5048. Baluška, F., Salaj, J., Mathur, J., Braun, M., Jasper, F., Samaj, J., Chua, N.-H., Barlow, P.W., and Volkmann, D. (2000). Root Hair Formation: F-Actin-Dependent Tip Growth Is Initiated by Local Assembly of Profilin-Supported F-Actin Meshworks Accumulated within Expansin-Enriched Bulges. Dev. Biol. 227, 618–632. Bao, Y., Kost, B., and Chua, N.-H. (2001). Reduced expression of alpha-tubulin genes in Arabidopsis thaliana specifically affects root growth and morphology, root hair development and root gravitropism. Plant J. 28, 145–157. Barr, F.A., and Gruneberg, U. (2007). Cytokinesis: Placing and Making the Final Cut. Cell 131, 847–860. Baskin, T.I. (2001). On the alignment of cellulose microfibrils by cortical microtubules: a review and a model. Protoplasma 215, 150–171. Baskin, T.I., Beemster, G.T.S., Judy-march, J.E., and Marga, F. (2004). Disorganization of Cortical Microtubules Stimulates Tangential Expansion and Reduces the Uniformity of 60 Cellulose Microfibril Alignment among Cells in the Root of Arabidopsis. Plant Physiol. 135, 2279–2290. Basu, D., Le, J., El-Essal, S.E.-D., Huang, S., Zhang, C., Mallery, E.L., Koliantz, G., Staiger, C.J., and Szymanski, D.B. (2005). DISTORTED3/SCAR2 Is a Putative Arabidopsis WAVE Complex Subunit That Activates the Arp2/3 Complex and Is Required for Epidermal Morphogenesis. Plant Cell 17, 502–524. Basu, D., Le, J., Zakharova, T., Mallery, E.L., and Szymanski, D.B. (2008). A SPIKE1 signaling complex controls actin-dependent cell morphogenesis through the heteromeric WAVE and ARP2/3 complexes. Proc. Natl. Acad. Sci. USA 105, 4044–4049. Benfey, P.N., Linstead, P.J., Roberts, K., Schiefelbein, J.W., Hauser, M.-T., and Aeschbacher, R.A. (1993). Root development in Arabidopsis: four mutants with dramatically altered root morphogenesis. Development 119, 57–70. Bennett, M.J., Marchant, A., Green, H.G., May, S.T., Ward, S.P., Millner, P.A., Walker, A.R., Schulz, B., and Feldmann, K.A. (1996). Arabidopsis AUX1 Gene: A Permease-Like Regulator of Root Gravitropism. Science 273, 948–950. Benschop, J.J., Mohammed, S., O’Flaherty, M., Heck, A.J.R., Slijper, M., and Menke, F.L.H. (2007). Quantitative Phosphoproteomics of Early Elicitor Signaling in Arabidopsis. Mol. Cell. Proteomics 6, 1198–1214. Berger, F., Hung, C.-Y., Dolan, L., and Schiefelbein, J.W. (1998a). Control of Cell Division in the Root Epidermis of Arabidopsis thaliana. Dev. Biol. 194, 235–245. Berger, F., Haseloff, J., Schiefelbein, J.W., and Dolan, L. (1998b). Positional information in root epidermis is defined during embryogenesis and acts in domains with strict boundaries. Curr. Biol. 8, 421–430. Bernhardt, C., Lee, M.M., Gonzalez, A., Zhang, F., Lloyd, A.M., and Schiefelbein, J.W. (2003). The bHLH genes GLABRA3 (GL3) and ENHANCER OF GLABRA3 (EGL3) specify epidermal cell fate in the Arabidopsis root. Development 130, 6431–6439. Bernhardt, C., Zhao, M., Gonzalez, A., Lloyd, A.M., and Schiefelbein, J.W. (2005). The bHLH genes GL3 and EGL3 participate in an intercellular regulatory circuit that controls cell patterning in the Arabidopsis root epidermis. Development 132, 291–298. Bibikova, T.N., Jacob, T., Dahse, I., and Gilroy, S. (1998). Localized changes in apoplastic and cytoplasmic pH are associated with root hair development in Arabidopsis thaliana. Development 125, 2925–2934. Bibikova, T.N., Blancaflor, E.B., and Gilroy, S. (1999). Microtubules regulate tip growth and orientation in root hairs of Arabidopsis thaliana. Plant J. 17, 657–665. Bichet, A., Desnos, T., Turner, S., Grandjean, O., and Höfte, H. (2001). BOTERO1 is required for normal orientation of cortical microtubules and anisotropic cell expansion in Arabidopsis. Plant J. 25, 137–148. Blair, A., Tomlinson, A., Pham, H., Gunsalus, K.C., Goldberg, M.L., and Laski, F.A. (2006). Twinstar, the Drosophila homolog of cofilin/ADF, is required for planar cell polarity patterning. Development 133, 1789–1797. Blilou, I., Xu, J., Wildwater, M., Willemsen, V., Friml, J., Paponov, I., Heidstra, R., Aida, M., Palme, K., and Scheres, B. (2005). The PIN auxin efflux facilitator network controls growth and patterning in Arabidopsis roots. Nature 433, 39–44. Bouyer, D., Geier, F., Kragler, F., Schnittger, A., Pesch, M., Wester, K., Balkunde, R., Timmer, J., Fleck, C., and Hülskamp, M. (2008). Two-Dimensional Patterning by a Trapping/Depletion Mechanism: The Role of TTG1 and GL3 in Arabidopsis Trichome Formation. PLoS Biol. 6, e141. Bramsiepe, J., Wester, K., Weinl, C., Roodbarkelari, F., Kasili, R., Larkin, J.C., Hülskamp, M., and Schnittger, A. (2010). Endoreplication Controls Cell Fate Maintenance. PLoS Genet. 6. Brandizzi, F., and Wasteneys, G.O. (2013). Cytoskeleton-dependent endomembrane organization in plant cells: an emerging role for microtubules. Plant J. 75, 339–349. Van Bruaene, N., Joss, G., and Van Oostveldt, P. (2004). Reorganization and in Vivo Dynamics of Microtubules during Arabidopsis Root Hair Development. Plant Physiol. 136, 3905–3919. Burk, D.H., Liu, B., Zhong, R., Morrison, W.H., and Ye, Z.-H. (2001). A Katanin-like Protein Regulates Normal Cell Wall Biosynthesis and Cell Elongation. Plant Cell 13, 807–827. Buschmann, H., Chan, J., Sanchez-Pulido, L., Andrade-Navarro, M.A., Doonan, J.H., and Lloyd, C.W. (2006). Microtubule-Associated AIR9 Recognizes the Cortical Division Site at Preprophase and Cell-Plate Insertion. Curr. Biol. 16, 1938–1943. Camilleri, C., Azimzadeh, J., Pastuglia, M., Bellini, C., Grandjean, O., and Bouchez, D. (2002). The Arabidopsis TONNEAU2 Gene Encodes a Putative Novel Protein Phosphatase 2A 61 Regulatory Subunit Essential for the Control of the Cortical Cytoskeleton. Plant Cell 14, 833–845. Cannon, M.C., Terneus, K., Hall, Q., Tan, L., Wang, Y., Wegenhart, B.L., Chen, L., Lamport, D.T.A., Chen, Y., and Kieliszewski, M.J. (2008). Self-assembly of the plant cell wall requires an extensin scaffold. Proc. Natl. Acad. Sci. USA 105, 2226–2231. Cárdenas, L. (2009). New findings in the mechanisms regulating polar growth in root hair cells. Plant Signal. Behav. 4, 4–8. Carol, R.J., Takeda, S., Linstead, P.J., Durrant, M.C., Kakesova, H., Derbyshire, P., Drea, S., Zarsky, V., and Dolan, L. (2005). A RhoGDP dissociation inhibitor spatially regulates growth in root hair cells. Nature 438. Chan, J., Calder, G.M., Doonan, J.H., and Lloyd, C.W. (2003). EB1 reveals mobile microtubule nucleation sites in Arabidopsis. Nat. Cell Biol. 5, 967–971. Churchman, M.L., Brown, M.L., Kato, N., Kirik, V., Hülskamp, M., Inzé, D., De Veylder, L., Walker, J.D., Zheng, Z., Oppenheimer, D.G., et al. (2006). SIAMESE, a Plant-Specific Cell Cycle Regulator, Controls Endoreplication Onset in Arabidopsis thaliana. Plant Cell 18, 3145–3157. Cleary, A.L., and Smith, L.G. (1998). The Tangled1 Gene Is Required for Spatial Control of Cytoskeletal Arrays Associated with Cell Division during Maize Leaf Development. Plant Cell 10, 1875–1888. Costa, S., and Shaw, P. (2006). Chromatin organization and cell fate switch respond to positional information in Arabidopsis. Nature 439, 493–496. Cove, D.J. (2000). The generation and modification of cell polarity. J. Exp. Bot. 51, 831–838. Di Cristina, M., Sessa, G., Dolan, L., Linstead, P.J., Baima, S., Ruberti, I., and Morelli, G. (1996). The Arabidopsis Athb-10 (GLABRA2) is an HD-Zip protein required for regulation of root hair development. Plant J. 10, 393–402. Van Damme, D., Coutuer, S., De Rycke, R., Bouget, F.-Y., Inzé, D., and Geelen, D. (2006). Somatic Cytokinesis and Pollen Maturation in Arabidopsis Depend on TPLATE, Which Has Domains Similar to Coat Proteins. Plant Cell 18, 3502–3518. Van Damme, D., Gadeyne, A., Vanstraelen, M., Inzé, D., Van Montagu, M.C.E., De Jaeger, G., Russinova, E., and Geelen, D. (2011). Adaptin-like protein TPLATE and clathrin recruitment during plant somatic cytokinesis occurs via two distinct pathways. Proc. Natl. Acad. Sci. USA 108, 615–620. Desai, A., and Mitchison, T.J. (1997). Microtubule Polymerization Dynamics. Annu. Rev. Cell Dev. Biol. 13, 83–117. Dhonukshe, P., and Gadella, T.W.J. (2003). Alteration of Microtubule Dynamic Instability during Preprophase Band Formation Revealed by Yellow Fluorescent Protein–CLIP170 Microtubule Plus-End Labeling. Plant Cell 15, 597–611. Dhonukshe, P., Laxalt, A.M., Goedhart, J., Gadella, T.W.J., and Munnik, T. (2003). Phospholipase D Activation Correlates with Microtubule Reorganization in Living Plant Cells. Plant Cell 15, 2666–2679. Dixit, R., and Cyr, R.J. (2004). Encounters between Dynamic Cortical Microtubules Promote Ordering of the Cortical Array through Angle-Dependent Modifications of Microtubule Behavior. Plant Cell 16, 3274–3284. Dolan, L. (2001). The role of ethylene in root hair growth in Arabidopsis. J. Plant Nutr. Soil Sci. 164, 141–145. Dolan, L., and Costa, S. (2001). Evolution and genetics of root hair stripes in the root epidermis. J. Exp. Bot. 52, 413–417. Dolan, L., Janmaat, K., Willemsen, V., Linstead, P.J., Poethig, S., Roberts, K., and Scheres, B. (1993). Cellular organisation of the Arabidopsis thaliana root. Development 119, 71–84. Dolan, L., Duckett, C.M., Grierson, C.S., Linstead, P.J., Schneider, K., Lawson, E., Dean, C., Poethig, S., and Roberts, K. (1994). Clonal relationships and cell patterning in the root epidermis of Arabidopsis. Development 120, 2465–2474. Doonan, J.H., Cove, D.J., and Lloyd, C.W. (1985). Immunofluorescence Microscopy of Microtubules in Intact Cell Lineages of the Moss, Physcomitrella patens. J. Cell Sci. 75, 131– 147. Eaton, S., Wepf, R., and Simons, K. (1996). Roles for Rac1 and Cdc42 in Planar Polarization and Hair Outgrowth in the Wing of Drosophila. J. Cell Biol. 135, 1277–1289. Ehrhardt, D.W. (2008). Straighten up and fly right - microtubule dynamics and organization of non-centrosomal arrays in higher plants. Curr. Opin. Cell Biol. 20, 107–116. Ehrhardt, D.W., and Shaw, S.L. (2006). Microtubule Dynamics and Organization in the Plant Cortical Array. Annu. Rev. Plant Biol. 57, 859–875. 62 Fischer, U., Ikeda, Y., Ljung, K., Serralbo, O., Singh, M., Heidstra, R., Palme, K., Scheres, B., and Grebe, M. (2006). Vectorial Information for Arabidopsis Planar Polarity Is Mediated by Combined AUX1, EIN2, and GNOM Activity. Curr. Biol. 16, 2143–2149. Folkers, U., Berger, J., and Hülskamp, M. (1997). Cell morphogenesis of trichomes in Arabidopsis: differential control of primary and secondary branching by branch initiation regulators and cell growth. Development 124, 3779–3786. Folkers, U., Kirik, V., Schöbinger, U., Falk, S., Krishnakumar, S., Pollock, M.A., Oppenheimer, D.G., Day, I., Reddy, A.R., Jürgens, G., et al. (2002). The cell morphogenesis gene ANGUSTIFOLIA encodes a CtBP/BARS-like protein and is involved in the control of the microtubule cytoskeleton. EMBO J. 21, 1280–1288. Foreman, J., Demidchik, V., Bothwell, J.H.F., Mylona, P., Miedema, H., Torres, M.A., Linstead, P.J., Costa, S., Brownlee, C., Jones, J.D.G., et al. (2003). Reactive oxygen species produced by NADPH oxidase regulate plant cell growth. Nature 422, 442–446. Fu, Y., Gu, Y., Zheng, Z., Wasteneys, G.O., and Yang, Z. (2005). Arabidopsis Interdigitating Cell Growth Requires Two Antagonistic Pathways with Opposing Action on Cell Morphogenesis. Cell 120, 687–700. Fu, Y., Xu, T., Zhu, L., Wen, M., and Yang, Z. (2009). A ROP GTPase Signaling Pathway Controls Cortical Microtubule Ordering and Cell Expansion in Arabidopsis. Curr. Biol. 19, 1827–1832. Gadeyne, A., Sánchez-Rodríguez, C., Vanneste, S., Di Rubbo, S., Zauber, H., Vanneste, K., Van Leene, J., De Winne, N., Eeckhout, D., Persiau, G., et al. (2014). The TPLATE Adaptor Complex Drives Clathrin-Mediated Endocytosis in Plants. Cell 156, 691–704. Galjart, N. (2005). CLIPs and CLASPs and Cellular Dynamics. Nat. Rev. Mol. Cell Biol. 6, 487– 498. Galway, M.E., Masucci, J.D., Lloyd, A.M., Walbot, V., Davis, R.W., and Schiefelbein, J.W. (1994). The TTG Gene Is Required to Specify Epidermal Cell Fate and Cell Patterning in the Arabidopsis Root. Dev. Biol. 166, 740–754. Gälweiler, L., Guan, C., Müller, A., Wisman, E., Mendgen, K., Yephremov, A., and Palme, K. (1998). Regulation of Polar Auxin Transport by AtPIN1 in Arabidopsis Vascular Tissue. Science 282, 2226–2230. Gardiner, J.C. (2013). The evolution and diversification of plant microtubule-associated proteins. Plant J. 75, 219–229. Gardiner, J.C., Harper, J.D.I., Weerakoon, N.D., Collings, D.A., Ritchie, S., Gilroy, S., Cyr, R.J., and Marc, J. (2001). A 90-kD Phospholipase D from Tobacco Binds to Microtubules and the Plasma Membrane. Plant Cell 13, 2143–2158. Geldner, N. (2009). Cell polarity in plants - a PARspective on PINs. Curr. Opin. Plant Biol. 12, 42–48. Geldner, N., Anders, N., Wolters, H., Keicher, J., Kornberger, W., Muller, P., Delbarre, A., Ueda, T., Nakano, A., and Jürgens, G. (2003). The Arabidopsis GNOM ARF-GEF Mediates Endosomal Recycling, Auxin Transport, and Auxin-Dependent Plant Growth. Cell 112, 219– 230. Gilliland, L.U., Pawloski, L.C., Kandasamy, M.K., and Meagher, R.B. (2003). Arabidopsis actin gene ACT7 plays an essential role in germination and root growth. Plant J. 33, 319–328. Goodrich, L. V., and Strutt, D. (2011). Principles of planar polarity in animal development. Development 138, 1877–1892. Granger, C.L., and Cyr, R.J. (2001). Spatiotemporal relationships between growth and microtubule orientation as revealed in living root cells of Arabidopsis thaliana transformed with green-fluorescent-protein gene construct GFP-MBD. Protoplasma 216, 201–214. Grebe, M. (2004). Ups and downs of tissue and planar polarity in plants. BioEssays 26, 719– 729. Grebe, M., Friml, J., Swarup, R., Ljung, K., Sandberg, G., Terlou, M., Palme, K., Bennett, M.J., and Scheres, B. (2002). Cell Polarity Signaling in Arabidopsis Involves a BFA-Sensitive Auxin Influx Pathway. Curr. Biol. 12, 329–334. Grierson, C.S., Parker, J.S., and Kemp, A.C. (2001). Arabidopsis genes with roles in root hair development. J. Plant Nutr. Soil Sci. 164, 131–140. Hall, Q., and Cannon, M.C. (2002). The Cell Wall Hydroxyproline-Rich Glycoprotein RSH Is Essential for Normal Embryo Development in Arabidopsis. Plant Cell 14, 1161–1172. Hassan, H., Scheres, B., and Blilou, I. (2010). JACKDAW controls epidermal patterning in the Arabidopsis root meristem through a non-cell-autonomous mechanism. Development 137, 1523–1529. 63 Hauser, M.-T., Morikami, A., and Benfey, P.N. (1995). Conditional root expansion mutants of Arabidopsis. Development 121, 1237–1252. Henty-Ridilla, J.L., Li, J., Blanchoin, L., and Staiger, C.J. (2013). Actin dynamics in the cortical array of plant cells. Curr. Opin. Plant Biol. 16, 678–687. Ten Hove, C.A., Willemsen, V., de Vries, W.J., van Dijken, A., Scheres, B., and Heidstra, R. (2010). SCHIZORIZA Encodes a Nuclear Factor Regulating Asymmetry of Stem Cell Divisions in the Arabidopsis Root. Curr. Biol. 20, 452–457. Hülskamp, M. (2004). Plant trichomes: a model for cell differentiation. Nat. Rev. Mol. Cell Biol. 5, 471–480. Hülskamp, M., Miséra, S., and Jürgens, G. (1994). Genetic Dissection of Trichome Cell Development in Arabidopsis. Cell 76, 555–566. Hussey, P.J., Ketelaar, T., and Deeks, M.J. (2006). Control of the Actin Cytoskeleton in Plant Cell Growth. Annu. Rev. Plant Biol. 57, 109–125. Ikeda, Y., Men, S., Fischer, U., Alonso, J.M., Stepanova, A.N., Ljung, K., and Grebe, M. (2009). Local auxin biosynthesis modulates gradient-directed planar polarity in Arabidopsis. Nat. Cell Biol. 11. Ishida, T., Kaneko, Y., Iwano, M., and Hashimoto, T. (2007). Helical microtubule arrays in a collection of twisting tubulin mutants of Arabidopsis thaliana. Proc. Natl. Acad. Sci. USA 104, 8544–8549. Ishida, T., Kurata, T., Okada, K., and Wada, T. (2008). A Genetic Regulatory Network in the Development of Trichomes and Root Hairs. Annu. Rev. Plant Biol. 59, 365–386. Jaillais, Y., Fobis-Loisy, I., Miège, C., Rollin, C., and Gaude, T. (2006). AtSNX1 defines an endosome for auxin-carrier trafficking in Arabidopsis. Nature 443, 106–109. Jones, A.R., Kramer, E.M., Knox, K., Swarup, R., Bennett, M.J., Lazarus, C.M., Leyser, H.M.O., and Grierson, C.S. (2009). Auxin transport through non-hair cells sustains root-hair development. Nat. Cell Biol. 11, 78–84. Jones, M.A., Shen, J.-J., Fu, Y., Li, H., Yang, Z., and Grierson, C.S. (2002). The Arabidopsis Rop2 GTPase Is a Positive Regulator of Both Root Hair Initiation and Tip Growth. Plant Cell 14, 763–776. Jones, M.A., Raymond, M.J., Yang, Z., and Smirnoff, N. (2007). NADPH oxidase-dependent reactive oxygen species formation required for root hair growth depends on ROP GTPase. J. Exp. Bot. 58, 1261–1270. Jürgens, G. (2005). Cytokinesis in Higher Plants. Annu. Rev. Plant Biol. 56, 281–299. Jürgens, G., Mayer, U., Torres-Ruiz, R.A., Berleth, T., and Miséra, S. (1991). Genetic analysis of pattern formation in the Arabidopsis embryo. Development 27–38. Kandasamy, M.K., McKinney, E.C., and Meagher, R.B. (2009). A Single Vegetative Actin Isovariant Overexpressed under the Control of Multiple Regulatory Sequences Is Sufficient for Normal Arabidopsis Development. Plant Cell 21, 701–718. Kang, Y.H., Kirik, V., Hülskamp, M., Nam, K.H., Hagely, K., Lee, M.M., and Schiefelbein, J.W. (2009). The MYB23 Gene Provides a Positive Feedback Loop for Cell Fate Specification in the Arabidopsis Root Epidermis. Plant Cell 21, 1080–1094. Kang, Y.H., Song, S.-K., Schiefelbein, J.W., and Lee, M.M. (2013). Nuclear Trapping Controls the Position-Dependent Localization of CAPRICE in the Root Epidermis of Arabidopsis. Plant Physiol. 163, 193–204. Karahara, I., Suda, J., Tahara, H., Yokota, E., Shimmen, T., Misaki, K., Yonemura, S., Staehelin, L.A., and Mineyuki, Y. (2009). The preprophase band is a localized center of clathrinmediated endocytosis in late prophase cells of the onion cotyledon epidermis. Plant J. 57, 819–831. Ketelaar, T., Faivre-Moskalenko, C., Esseling, J.J., de Ruijter, N.C.A., Grierson, C.S., Dogterom, M., and Emons, A.M.C. (2002). Positioning of Nuclei in Arabidopsis Root Hairs: An ActinRegulated Process of Tip Growth. Plant Cell 14, 2941–2955. Ketelaar, T., Allwood, E.G., Anthony, R.G., Voigt, B., Menzel, D., and Hussey, P.J. (2004). The Actin-Interacting Protein AIP1 Is Essential for Actin Organization and Plant Development. Curr. Biol. 14, 145–149. Kirik, V., Grini, P.E., Mathur, J., Klinkhammer, I., Adler, K., Bechtold, N., Herzog, M., Bonneville, J.-M., and Hülskamp, M. (2002a). The Arabidopsis TUBULIN-FOLDING COFACTOR A Gene Is Involved in the Control of the alfa/beta-Tubulin Monomer Balance. EMBO J. 14, 2265–2276. Kirik, V., Mathur, J., Grini, P.E., Klinkhammer, I., Adler, K., Bechtold, N., Herzog, M., Bonneville, J.-M., and Hülskamp, M. (2002b). Functional Analysis of the Tubulin-Folding Cofactor C in Arabidopsis thaliana. Curr. Biol. 12, 1519–1523. 64 Kirik, V., Simon, M., Hülskamp, M., and Schiefelbein, J.W. (2004). The ENHANCER OF TRY AND CPC1 gene acts redundantly with TRIPTYCHON and CAPRICE in trichome and root hair cell patterning in Arabidopsis. Dev. Biol. 268, 506–513. Kirik, V., Herrmann, U., Parupalli, C., Sedbrook, J.C., Ehrhardt, D.W., and Hülskamp, M. (2007). CLASP localizes in two discrete patterns on cortical microtubules and is required for cell morphogenesis and cell division in Arabidopsis. J. Cell Sci. 120, 4416–4425. Kleine-Vehn, J., and Friml, J. (2008). Polar Targeting and Endocytic Recycling in AuxinDependent Plant Development. Annu. Rev. Cell Dev. Biol. 24, 447–473. Kleine-Vehn, J., Dhonukshe, P., Sauer, M., Brewer, P.B., Wiśniewska, J., Paciorek, T., Benková, E., and Friml, J. (2008). ARF GEF-Dependent Transcytosis and Polar Delivery of PIN Auxin Carriers in Arabidopsis. Curr. Biol. 18, 526–531. Koshino-Kimura, Y., Wada, T., Tachibana, T., Tsugeki, R., Ishiguro, S., and Okada, K. (2005). Regulation of CAPRICE Transcription by MYB Proteins for Root Epidermis Differentiation in Arabidopsis. Plant Cell Physiol. 46, 817–826. Kramer, E.M., and Bennett, M.J. (2006). Auxin transport: a field in flux. Trends Plant Sci. 11, 382–386. Kurata, T., Ishida, T., Kawabata-Awai, C., Noguchi, M., Hattori, S., Sano, R., Nagasaka, R., Tominaga, R., Koshino-Kimura, Y., Kato, T., et al. (2005). Cell-to-cell movement of the CAPRICE protein in Arabidopsis root epidermal cell differentiation. Development 132, 5387–5398. Kusano, H., Testerink, C., Vermeer, J.E.M., Tsuge, T., Shimada, H., Oka, A., Munnik, T., and Aoyama, T. (2008). The Arabidopsis Phosphatidylinositol Phosphate 5-Kinase PIP5K3 is a Key Regulator of Root Hair Tip Growth. Plant Cell 20, 367–380. Kwak, S.-H., and Schiefelbein, J.W. (2007). The role of the SCRAMBLED receptor-like kinase in patterning the Arabidopsis root epidermis. Dev. Biol. 302, 118–131. Kwak, S.-H., and Schiefelbein, J.W. (2008). A Feedback Mechanism Controlling SCRAMBLED Receptor Accumulation and Cell-Type Pattern in Arabidopsis. Curr. Biol. 18, 1949–1954. Kwak, S.-H., Shen, R., and Schiefelbein, J.W. (2005). Positional Signaling Mediated by a Receptor-like Kinase in Arabidopsis. Science 307, 1111–1113. Łangowski, L., Růzicka, K., Naramoto, S., Kleine-Vehn, J., and Friml, J. (2010). Trafficking to the Outer Polar Domain Defines the Root-Soil Interface. Curr. Biol. 20, 904–908. Lansbergen, G., Grigoriev, I., Mimori-Kiyosue, Y., Ohtsuka, T., Higa, S., Kitajima, I., Demmers, J., Galjart, N., Houtsmuller, A.B., Grosveld, F., et al. (2006). CLASPs Attach Microtubule Plus Ends to the Cell Cortex through a Complex with LL5beta. Dev. Cell 11, 21–32. Lawrence, P.A. (1966). Gradients in the insect segment: the orientation of hairs in the milkweed bug Oncopeltus Fasciatus. J. Exp. Bot. 44, 607–620. Lee, M.M., and Schiefelbein, J.W. (1999). WEREWOLF, a MYB-Related Protein in Arabidopsis, Is a Position-Dependent Regulator of Epidermal Cell Patterning. Cell 99, 473–483. Leyser, O. (2005). Auxin Distribution and Plant Pattern Formation: How Many Angels Can Dance on the Point of PIN? Cell 121, 819–822. Li, H., Xu, T., Lin, D., Wen, M., Xie, M., Duclercq, J., Bielach, A., Kim, J., Reddy, G.V., Zuo, J., et al. (2013). Cytokinin signaling regulates pavement cell morphogenesis in Arabidopsis. Cell Res. 23, 290–299. Lin, D., Nagawa, S., Chen, J., Cao, L., Chen, X., Xu, T., Li, H., Dhonukshe, P., Yamamuro, C., Friml, J., et al. (2012). A ROP GTPase-Dependent Auxin Signaling Pathway Regulates the Subcellular Distribution of PIN2 in Arabidopsis Roots. Curr. Biol. 22, 1319–1325. Lin, D., Cao, L., Zhou, Z., Zhu, L., Ehrhardt, D.W., Yang, Z., and Fu, Y. (2013). Rho GTPase Signaling Activates Microtubule Severing to Promote Microtubule Ordering in Arabidopsis. Curr. Biol. 23, 290–297. Lindeboom, J.J., Nakamura, M., Hibbel, A., Shundyak, K., Gutierrez, R., Ketelaar, T., Emons, A.M.C., Mulder, B.M., Kirik, V., and Ehrhardt, D.W. (2013). A Mechanism for Reorientation of Cortical Microtubule Arrays Driven by Microtubule Severing. Science 342. Lu, L., Lee, Y.-R.J., Pan, R., Maloof, J.N., and Liu, B. (2005). An Internal Motor Kinesin is Associated with the Golgi Apparatus and Plays a Role in Trichome Morphogenesis in Arabidopsis. Mol. Biol. Cell 16, 811–823. Maiato, H., Khodjakov, A., and Rieder, C.L. (2005). Drosophila CLASP is required for the incorporation of microtubule subunits into fluxing kinetochore fibres. Nat. Cell Biol. 7, 42– 47. Marco, E., Wedlich-Soldner, R., Li, R., Altschuler, S.J., and Wu, L.F. (2007). Endocytosis Optimizes the Dynamic Localization of Membrane Proteins that Regulate Cortical Polarity. Cell 129, 411–422. 65 Martin, S.G., McDonald, W.H., Yates, J.R., and Chang, F. (2005). Tea4p Links Microtubule Plus Ends with the Formin For3p in the Establishment of Cell Polarity. Dev. Cell 8, 479–491. Martin, S.G., Rincón, S.A., Basu, R., Pérez, P., and Chang, F. (2007). Regulation of the Formin for3p by cdc42p and bud6p. Mol. Biol. Cell 18, 4155–4167. Masucci, J.D., and Schiefelbein, J.W. (1994). The rhd6 Mutation of Arabidopsis thaliana Alters Root-Hair Initiation through an Auxin- and Ethylene-Associated Process. Plant Physiol. 106, 1335–1346. Masucci, J.D., and Schiefelbein, J.W. (1996). Hormones Act Downstream of TTG and GL2 to Promote Root Hair Outgrowth during Epidermis Development in the Arabidopsis Root. Plant Cell 8, 1505–1517. Masucci, J.D., Rerie, W.G., Foreman, D.R., Zhang, M., Galway, M.E., Marks, M.D., and Schiefelbein, J.W. (1996). The homeobox gene GLABRA 2 is required for positiondependent cell differentiation in the root epidermis of Arabidopsis thaliana. Development 122, 1253–1260. Mata, J., and Nurse, P. (1997). tea1 and the Microtubular Cytoskeleton Are Important for Generating Global Spatial Order within the Fission Yeast Cell. Cell 89, 939–949. Mathur, J., and Chua, N.-H. (2000). Microtubule Stabilization Leads to Growth Reorientation in Arabidopsis Trichomes. Plant Cell 12, 465–477. Mathur, J., Spielhofer, P., Kost, B., and Chua, N.-H. (1999). The actin cytoskeleton is required to elaborate and maintain spatial patterning during trichome cell morphogenesis in Arabidopsis thaliana. Development 126, 5559–5568. Mathur, J., Mathur, N., Kirik, V., Kernebeck, B., Srinivas, B.P., and Hülskamp, M. (2003a). Arabidopsis CROOKED encodes for the smallest subunit of the ARP2/3 complex and controls cell shape by region specific fine F-actin formation. Development 130, 3137–3146. Mathur, J., Mathur, N., Kernebeck, B., and Hülskamp, M. (2003b). Mutations in Actin-Related Proteins 2 and 3 Affect Cell Shape Development in Arabidopsis. Plant Cell 15, 1632–1645. Matis, M., and Axelrod, J.D. (2013). Regulation of PCP by the Fat signaling pathway. Genes Dev. 27, 2207–2220. Mattsson, J., Ckurshumova, W., and Berleth, T. (2003). Auxin Signaling in Arabidopsis Leaf Vascular Development. Plant Physiol. 131, 1327–1339. Men, S., Boutté, Y., Ikeda, Y., Li, X., Palme, K., Stierhof, Y.-D., Hartmann, M.-A., Moritz, T., and Grebe, M. (2008). Sterol-dependent endocytosis mediates post-cytokinetic acquisition of PIN2 auxin efflux carrier polarity. Nat. Cell Biol. 10, 237–244. Menand, B., Yi, K., Jouannic, S., Hoffmann, L., Ryan, E., Linstead, P.J., Schaefer, D.G., and Dolan, L. (2007). An Ancient Mechanism Controls the Development of Cells with a Rooting Function in Land Plants. Science 316, 1477–1480. Molendijk, A.J., Bischoff, F., Friml, J., Rajendrakumar, C.S. V., Braun, M., Gilroy, S., and Palme, K. (2001). Arabidopsis thaliana Rop GTPases are localized to tips of root hairs and control polar growth. EMBO J. 20, 2779–2788. Monshausen, G.B., Bibikova, T.N., Messerli, M.A., Shi, C., and Gilroy, S. (2007). Oscillations in extracellular pH and reactive oxygen species modulate tip growth of Arabidopsis root hairs. Proc. Natl. Acad. Sci. USA 104, 20996–21001. Monshausen, G.B., Messerli, M.A., and Gilroy, S. (2008). Imaging of the Yellow Cameleon 3.6 Indicator Reveals That Elevations in Cytosolic Ca2+ Follow Oscillating Increases in Growth in Root Hairs of Arabidopsis. Plant Physiol. 147, 1690–1698. Muday, G.K., Rahman, A., and Binder, B.M. (2012). Auxin and ethylene: collaborators or competitors? Trends Plant Sci. 17, 181–195. Müller, M., and Schmidt, W. (2004). Environmentally Induced Plasticity of Root Hair Development in Arabidopsis. Plant Physiol. 134, 409–419. Müller, A., Guan, C., Gälweiler, L., Tänzler, P., Huijser, P., Marchant, A., Parry, G., Bennett, M.J., Wisman, E., and Palme, K. (1998). AtPIN2 defines a locus of Arabidopsis for root gravitropism control. EMBO J. 17, 6903–6911. Müller, S., Han, S., and Smith, L.G. (2006). Two Kinesins Are Involved in the Spatial Control of Cytokinesis in Arabidopsis thaliana. Curr. Biol. 16, 888–894. Müller, S., Wright, A.J., and Smith, L.G. (2009). Division plane control in plants: new players in the band. Trends Cell Biol. 19, 180–188. Murata, T., Sonobe, S., Baskin, T.I., Hyodo, S., Hasezawa, S., Nagata, T., Horio, T., and Hasebe, M. (2005). Microtubule-dependent microtubule nucleation based on recruitment of gamma-tubulin in higher plants. Nat. Cell Biol. 7, 961–968. 66 Mylona, P., Linstead, P.J., Martienssen, R., and Dolan, L. (2002). SCHIZORIZA controls an asymmetric cell division and restricts epidermal identity in the Arabidopsis root. Development 129, 4327–4334. Nakajima, K., Furutani, I., Tachimoto, H., Matsubara, H., and Hashimoto, T. (2004). SPIRAL1 Encodes a Plant-Specific Microtubule-Localized Protein Required for Directional Control of Rapidly Expanding Arabidopsis Cells. Plant Cell 16, 1178–1190. Nakamura, M., and Hashimoto, T. (2009). A mutation in the Arabidopsis gamma-tubulincontaining complex causes helical growth and abnormal microtubule branching. J. Cell Sci. 122, 2208–2217. Nakamura, M., Ehrhardt, D.W., and Hashimoto, T. (2010). Microtubule and katanindependent dynamics of microtubule nucleation complexes in the acentrosomal Arabidopsis cortical array. Nat. Cell Biol. 12, 1064–1070. Nakamura, M., Kiefer, C.S., and Grebe, M. (2012). Planar polarity, tissue polarity and planar morphogenesis in plants. Curr. Opin. Plant Biol. 15, 593–600. Nelson, W.J. (2003). Adaptation of core mechanisms to generate cell polarity. Nature 422, 766–774. Nübler-Jung, K., Bonitz, R., and Sonnenschein, M. (1987). Cell polarity during wound healing in an insect epidermis. Development 100, 163–170. Ohashi, Y., Oka, A., Rodrigues-Pousada, R., Possenti, M., Ruberti, I., Morelli, G., and Aoyama, T. (2003). Modulation of Phospholipid Signaling by GLABRA2 in Root-Hair Pattern Formation. Science 300, 1427–1430. Oppenheimer, D.G., Herman, P.L., Sivakumaran, S., Esch, J., and Marks, M.D. (1991). A myb gene required for leaf trichome differentiation in Arabidopsis is expressed in stipules. Cell 67, 483–493. Oppenheimer, D.G., Pollock, M.A., Vacik, J., Szymanski, D.B., Ericson, B., Feldmann, K.A., and Marks, M.D. (1997). Essential role of a kinesin-like protein in Arabidopsis trichome morphogenesis. Proc. Natl. Acad. Sci. USA 94, 6261–6266. Ovecka, M., Berson, T., Beck, M., Derksen, J., Samaj, J., Baluška, F., and Lichtscheidl, I.K. (2010). Structural Sterols Are Involved in Both the Initiation and Tip Growth of Root Hairs in Arabidopsis thaliana. Plant Cell 22, 2999–3019. Panteris, E., Adamakis, I.-D.S., Voulgari, G., and Papadopoulou, G. (2011). A Role for Katanin in Plant Cell Division: Microtubule Organization in Dividing Root Cells of fra2 and lue1 Arabidopsis thaliana Mutants. Cytoskeleton 68, 401–413. Paredez, A.R., Somerville, C.R., and Ehrhardt, D.W. (2006). Visualization of Cellulose Synthase Demonstrates Functional Association with Microtubules. Science 312, 1491– 1495. Parker, J.S., Cavell, A.C., Dolan, L., Roberts, K., and Grierson, C.S. (2000). Genetic Interactions during Root Hair Morphogenesis in Arabidopsis. Plant Cell 12, 1961–1974. Payne, R.J.H., and Grierson, C.S. (2009). A Theoretical Model for ROP Localisation by Auxin in Arabidopsis Root Hair Cells. PLoS One 4. Pérez, P., and Rincón, S.A. (2010). Rho GTPases: regulation of cell polarity and growth in yeasts. Biochem. J. 426, 243–253. Pesch, M., and Hülskamp, M. (2004). Creating a two-dimensional pattern de novo during Arabidopsis trichome and root hair initiation. Curr. Opin. Genet. Dev. 14, 422–427. Pesch, M., and Hülskamp, M. (2009). One, two, three...models for trichome patterning in Arabidopsis? Curr. Opin. Plant Biol. 12, 587–592. Petrášek, J., Mravec, J., Bouchard, R., Blakeslee, J.J., Abas, M., Wiśniewska, J., Seifertova, D., Tadele, Z., Kubes, M., Covanová, M., et al. (2006). PIN Proteins Perform a Rate-Limiting Function in Cellular Auxin Efflux. Science 312, 914–918. Piel, M., and Tran, P.T. (2009). Cell Shape and Cell Division in Fission Yeast. Curr. Biol. 19, 823–827. Pietra, S., and Grebe, M. (2010). Auxin Paves the Way for Planar Morphogenesis. Cell 143, 29– 31. Pitts, R.J., Cernac, A., and Estelle, M. (1998). Auxin and ethylene promote root hair elongation in Arabidopsis. Plant J. 16, 553–560. Procissi, A., Guyon, A., Pierson, E.S., Giritch, A., Knuiman, B., Grandjean, O., Derksen, J., Tonelli, C., Pelletier, G., and Bonhomme, S. (2003). KINKY POLLEN encodes a SABRE-like protein required for tip growth in Arabidopsis and conserved among eukaryotes. Plant J. 36, 894–904. Qiu, J.-L., Jilk, R.A., Marks, M.D., and Szymanski, D.B. (2002). The Arabidopsis SPIKE1 Gene Is Required for Normal Cell Shape Control and Tissue Development. Plant Cell 14, 101–118. 67 Rasmussen, C.G., Humphries, J.A., and Smith, L.G. (2011). Determination of Symmetric and Asymmetric Division Planes in Plant Cells. Annu. Rev. Plant Biol. 62, 387–409. Rasmussen, C.G., Wright, A.J., and Müller, S. (2013). The role of the cytoskeleton and associated proteins in determination of the plant cell division plane. Plant J. 75, 258–269. Ren, N., Charlton, J., and Adler, P.N. (2007). The flare Gene, Which Encodes the AIP1 Protein of Drosophila, Functions to Regulate F-Actin Disassembly in Pupal Epidermal Cells. Genetics 176, 2223–2234. Rerie, W.G., Feldmann, K.A., and Marks, M.D. (1994). The GLABRA2 gene encodes a homeo domain protein required for normal trichome development in Arabidopsis. Genes Dev. 8, 1388–1399. Ringli, C., Baumberger, N., Diet, A., Frey, B., and Keller, B. (2002). ACTIN2 Is Essential for Bulge Site Selection and Tip Growth during Root Hair Development of Arabidopsis. Plant Physiol. 129, 1464–1472. Rodal, A.A., Tetreault, J.W., Lappalainen, P., Drubin, D.G., and Amberg, D.C. (1999). Aip1p Interacts with Cofilin to Disassemble Actin Filaments. J. Cell Biol. 145, 1251–1264. Rounds, C.M., and Bezanilla, M. (2013). Growth Mechanisms in Tip-Growing Plant Cells. Annu. Rev. Plant Biol. 64, 243–265. Ryu, K.H., Kang, Y.H., Park, Y., Hwang, I., Schiefelbein, J.W., and Lee, M.M. (2005). The WEREWOLF MYB protein directly regulates CAPRICE transcription during cell fate specification in the Arabidopsis root epidermis. Development 132, 4765–4775. Sabatini, S., Beis, D., Wolkenfelt, H., Murfett, J., Guilfoyle, T., Malamy, J., Benfey, P.N., Leyser, O., Bechtold, N., Weisbeek, P., et al. (1999). An Auxin-Dependent Distal Organizer of Pattern and Polarity in the Arabidopsis Root. Cell 99, 463–472. Sambade, A., Pratap, A., Buschmann, H., Morris, R.J., and Lloyd, C.W. (2012). The Influence of Light on Microtubule Dynamics and Alignment in the Arabidopsis Hypocotyl. Plant Cell 24, 192–201. Sampathkumar, A., Lindeboom, J.J., Debolt, S., Gutierrez, R., Ehrhardt, D.W., Ketelaar, T., and Persson, S. (2011). Live Cell Imaging Reveals Structural Associations between the Actin and Microtubule Cytoskeleton in Arabidopsis. Plant Cell 23, 2302–2313. Samuels, A.L., Giddings, T.H., and Staehelin, A. (1995). Cytokinesis in Tobacco BY-2 and Root Tip Cells: A New Model of Cell Plate Formation in Higher Plants. J. Cell Biol. 130, 1345– 1357. Schellmann, S., and Hülskamp, M. (2005). Epidermal differentiation: trichomes in Arabidopsis as a model system. Int. J. Dev. Biol. 49, 579–584. Schellmann, S., Schnittger, A., Kirik, V., Wada, T., Okada, K., Beermann, A., Thumfahrt, J., Jürgens, G., and Hülskamp, M. (2002). TRIPTYCHON and CAPRICE mediate lateral inhibition during trichome and root hair patterning in Arabidopsis. EMBO J. 21, 5036– 5046. Schiefelbein, J.W. (2000). Constructing a Plant Cell. The Genetic Control of Root Hair Development. Plant Physiol. 124, 1525–1531. Schiefelbein, J.W., and Somerville, C.R. (1990). Genetic Control of Root Hair Development in Arabidopsis thaliana. Plant Cell 2, 235–243. Schiefelbein, J.W., Kwak, S.-H., Wieckowski, Y., Barron, C., and Bruex, A. (2009). The gene regulatory network for root epidermal cell-type pattern formation in Arabidopsis. J. Exp. Bot. 60, 1515–1521. Schiefelbein, J.W., Huang, L., and Zheng, X. (2014). Regulation of epidermal cell fate in Arabidopsis roots: the importance of multiple feedback loops. Front. Plant Sci. 5, 47. Schmidt, W., and Schikora, A. (2001). Different Pathways Are Involved in Phosphate and Iron Stress-Induced Alterations of Root Epidermal Cell Development. Plant Physiol. 125, 2078– 2084. Schneider, K., Wells, B., Dolan, L., and Roberts, K. (1997). Structural and genetic analysis of epidermal cell differentiation in Arabidopsis primary roots. Development 124, 1789–1798. Sedbrook, J.C., and Kaloriti, D. (2008). Microtubules, MAPs and plant directional cell expansion. Trends Plant Sci. 13, 303–310. Sedbrook, J.C., Ehrhardt, D.W., Fisher, S.E., Scheible, W.-R., and Somerville, C.R. (2004). The Arabidopsis SKU6/SPIRAL1 Gene Encodes a Plus End–Localized Microtubule-Interacting Protein Involved in Directional Cell Expansion. Plant Cell 16, 1506–1520. Seguí-Simarro, J., Austin, J.R., White, E.A., and Staehelin, L.A. (2004). Electron Tomographic Analysis of Somatic Cell Plate Formation in Meristematic Cells of Arabidopsis Preserved by High-Pressure Freezing. Plant Cell 16, 836–856. 68 Seltzer, V., Janski, N., Canaday, J., Herzog, E., Erhardt, M., Evrard, J.-L., and Schmit, A.-C. (2007). Arabidopsis GCP2 and GCP3 are part of a soluble gamma-tubulin complex and have nuclear envelope targeting domains. Plant J. 52, 322–331. Shaw, S.L., Kamyar, R., and Ehrhardt, D.W. (2003). Sustained Microtubule Treadmilling in Arabidopsis Cortical Arrays. Science 300, 1715–1718. Simon, M., Lee, M.M., Lin, Y., Gish, L., and Schiefelbein, J.W. (2007). Distinct and overlapping roles of single-repeat MYB genes in root epidermal patterning. Dev. Biol. 311, 566–578. Simons, M., and Mlodzik, M. (2008). Planar Cell Polarity Signaling: From Fly Development to Human Disease. Annu. Rev. Genet. 42, 517–540. Singh, S.K., Fischer, U., Singh, M., Grebe, M., and Marchant, A. (2008). Insight into the early steps of root hair formation revealed by the procusteI cellulose synthase mutant of Arabidopsis thaliana. BMC Plant Biol. 8, 1–12. Sinnott, E.W., and Bloch, R. (1939). Cell polarity and the Differentiation of Root Hairs. Proc. Natl. Acad. Sci. USA 25, 248–252. Smertenko, A.P., Saleh, N., Igarashi, H., Mori, H., Hauser-Hahn, I., Jiang, C.-J., Sonobe, S., Lloyd, C.W., and Hussey, P.J. (2000). A new class of microtubule-associated proteins in plants. Nat. Cell Biol. 2, 750–753. Smertenko, A.P., Chang, H.-Y., Wagner, V., Kaloriti, D., Fenyk, S., Sonobe, S., Lloyd, C.W., Hauser, M.-T., and Hussey, P.J. (2004). The Arabidopsis Microtubule-Associated Protein AtMAP65-1: Molecular Analysis of Its Microtubule Bundling Activity. Plant Cell 16, 2035– 2047. Smith, L.G. (2003). Cytoskeletal control of plant cell shape: getting the fine points. Curr. Opin. Plant Biol. 6, 63–73. Smith, L.G., and Oppenheimer, D.G. (2005). Spatial Control of Cell Expansion by the Plant Cytoskeleton. Annu. Rev. Cell Dev. Biol. 21, 271–295. Snaith, H.A., and Sawin, K.E. (2003). Fission yeast mod5p regulates polarized growth through anchoring of tea1p at cell tips. Nature 423, 647–651. Song, J., Yang, W., Shih, I.-M., Zhang, Z., and Bai, J. (2006). Identification of BCOX1, a novel gene overexpressed in breast cancer. Biochim. Biophys. Acta 1760, 62–69. Sousa, A., Reis, R., Sampaio, P., and Sunkel, C.E. (2007). The Drosophila CLASP Homologue, Mast/Orbit Regulates the Dynamic Behaviour of Interphase Microtubules by Promoting the Pause State. Cell Motil. Cytoskeleton 64, 605–620. Staiger, C.J., and Blanchoin, L. (2006). Actin dynamics: old friends with new stories. Curr. Opin. Plant Biol. 9, 554–562. Staiger, C.J., Sheahan, M.B., Khurana, P., Wang, X., McCurdy, D.W., and Blanchoin, L. (2009). Actin filament dynamics are dominated by rapid growth and severing activity in the Arabidopsis cortical array. J. Cell Biol. 184, 269–280. Steen, D.A., and Chadwick, A. V. (1981). Ethylene Effects in Pea Stem Tissue. Plant Physiol. 67, 460–466. Stein, M., Dittgen, J., Sánchez-Rodríguez, C., Hou, B.-H., Molina, A., Schulze-Lefert, P., Lipka, V., and Somerville, S. (2006). Arabidopsis PEN3/PDR8, an ATP Binding Cassette Transporter, Contributes to Nonhost Resistance to Inappropriate Pathogens That Enter by Direct Penetration. Plant Cell 18, 731–746. Steinmann, T., Geldner, N., Grebe, M., Mangold, S., Jackson, C.L., Paris, S., Gälweiler, L., Palme, K., and Jürgens, G. (1999). Coordinated Polar Localization of Auxin Efflux Carrier PIN1 by GNOM ARF GEF. Science 286, 316–318. Strader, L.C., and Bartel, B. (2009). The Arabidopsis PLEIOTROPIC DRUG RESISTANCE8/ABCG36 ATP Binding Cassette Transporter Modulates Sensitivity to the Auxin Precursor Indole-3-Butyric Acid. Plant Cell 21, 1992–2007. Struk, S., and Dhonukshe, P. (2014). MAPs: cellular navigators for microtubule array orientations in Arabidopsis. Plant Cell Rep. 33, 1–21. Strutt, D. (2009). The planar polarity pathway. Curr. Biol. 18, 898–902. Strutt, D., and Warrington, S.J. (2008). Planar polarity genes in the Drosophila wing regulate the localisation of the FH3-domain protein Multiple Wing Hairs to control the site of hair production. Development 135, 3103–3111. Strutt, H., and Strutt, D. (2005). Long-range coordination of planar polarity in Drosophila. BioEssays 27, 1218–1227. Strutt, H., and Strutt, D. (2009). Asymmetric localisation of planar polarity proteins: Mechanisms and consequences. Semin. Cell Dev. Biol. 20, 957–963. Strutt, D.I., Weber, U., and Mlodzik, M. (1997). The role of RhoA in tissue polarity and Frizzled signalling. Nature 387, 292–295. 69 Swarup, R., Friml, J., Marchant, A., Ljung, K., Sandberg, G., Palme, K., and Bennett, M.J. (2001). Localization of the auxin permease AUX1 suggests two functionally distinct hormone transport pathways operate in the Arabidopsis root apex. Genes Dev. 15, 2648– 2653. Swarup, R., Kargul, J., Marchant, A., Zadik, D., Rahman, A., Mills, R., Yemm, A., May, S.T., Williams, L., Millner, P., et al. (2004). Structure-Function Analysis of the Presumptive Arabidopsis Auxin Permease AUX1. Plant Cell 16, 3069–3083. Szymanski, D.B. (2005). Breaking the WAVE complex: the point of Arabidopsis trichomes. Curr. Opin. Plant Biol. 8, 103–112. Takano, J., Tanaka, M., Toyoda, A., Miwa, K., Kasai, K., Fuji, K., Onouchi, H., Naito, S., and Fujiwara, T. (2010). Polar localization and degradation of Arabidopsis boron transporters through distinct trafficking pathways. Proc. Natl. Acad. Sci. USA 107, 5220–5225. Takeda, S., Gapper, C., Kaya, H., Bell, E., Kuchitsu, K., and Dolan, L. (2008). Local Positive Feedback Regulation Determines Cell Shape in Root Hair Cells. Science 319, 1241–1244. Tanimoto, M., Roberts, K., and Dolan, L. (1995). Ethylene is a positive regulator of root hair development in Arabidopsis thaliana. Plant J. 8, 943–948. Thole, J.M., Vermeer, J.E.M., Zhang, Y., Gadella, T.W.J., and Nielsen, E. (2008). ROOT HAIR DEFECTIVE4 Encodes a Phosphatidylinositol-4-Phosphate Phosphatase Required for Proper Root Hair Development in Arabidopsis thaliana. Plant Cell 20, 381–395. Thomas, C., and Strutt, D. (2012). The Roles of the Cadherins Fat and Dachsous in Planar Polarity Specification in Drosophila. Dev. Dyn. 241, 27–39. Tree, D.R.P., Shulman, J.M., Rousset, R., Scott, M.P., Gubb, D., and Axelrod, J.D. (2002). Prickle Mediates Feedback Amplification to Generate Asymmetric Planar Cell Polarity Signaling. Cell 109, 371–381. Tsvetkov, A.S., Samsonov, A., Akhmanova, A., Galjart, N., and Popov, S. V. (2007). MicrotubuleBinding Proteins CLASP1 and CLASP2 Interact With Actin Filaments. Cell Motil. Cytoskeleton 64, 519–530. Uyttewaal, M., Burian, A., Alim, K., Landrein, B., Borowska-Wykręt, D., Dedieu, A., Peaucelle, A., Ludynia, M., Traas, J., Boudaoud, A., et al. (2012). Mechanical Stress Acts via Katanin to Amplify Differences in Growth Rate between Adjacent Cells in Arabidopsis. Cell 149, 439– 451. Valdez-Taubas, J., and Pelham, H.R.B. (2003). Slow Diffusion of Proteins in the Yeast Plasma Membrane Allows Polarity to Be Maintained by Endocytic Cycling. Curr. Biol. 13, 1636– 1640. Vanstraelen, M., Van Damme, D., De Rycke, R., Mylle, E., Inzé, D., and Geelen, D. (2006). Cell Cycle-Dependent Targeting of a Kinesin at the Plasma Membrane Demarcates the Division Site in Plant Cells. Curr. Biol. 16, 308–314. Vinson, C.R., and Adler, P.N. (1987). Directional non-cell autonomy and the transmission of polarity information by the frizzled gene of Drosophila. Nature 329, 549–551. Vos, J.W., Dogterom, M., and Emons, A.M.C. (2004). Microtubules Become More Dynamic But Not Shorter During Preprophase Band Formation: A Possible “Search-and-Capture” Mechanism for Microtubule Translocation. Cell Motil. Cytoskeleton 57, 246–258. Wada, T., Tachibana, T., Shimura, Y., and Okada, K. (1997). Epidermal Cell Differentiation in Arabidopsis Determined by a Myb Homolog, CPC. Science 277, 1113–1116. Wada, T., Kurata, T., Tominaga, R., Koshino-Kimura, Y., Tachibana, T., Goto, K., Marks, M.D., Shimura, Y., and Okada, K. (2002). Role of a positive regulator of root hair development, CAPRICE, in Arabidopsis root epidermal cell differentiation. Development 129, 5409–5419. Walker, K.L., Müller, S., Moss, D., Ehrhardt, D.W., and Smith, L.G. (2007). Arabidopsis TANGLED Identifies the Division Plane throughout Mitosis and Cytokinesis. Curr. Biol. 17, 1827–1836. Wang, C., Yan, X., Chen, Q., Jiang, N., Fu, W., Ma, B., Liu, J., Li, C., Bednarek, S.Y., and Pan, J. (2013). Clathrin Light Chains Regulate Clathrin-Mediated Trafficking, Auxin Signaling, and Development in Arabidopsis. Plant Cell 25, 499–516. Wasteneys, G.O. (2002). Microtubule organization in the green kingdom: chaos or self-order? J. Cell Sci. 115, 1345–1354. Wasteneys, G.O., and Galway, M.E. (2003). Remodeling the Cytoskeleton for Growth and Form: An Overview with Some New Views. Annu. Rev. Plant Biol. 54, 691–722. Webb, M., Jouannic, S., Foreman, J., Linstead, P.J., and Dolan, L. (2002). Cell specification in the Arabidopsis root epidermis requires the activity of ECTOPIC ROOT HAIR 3 - a kataninp60 protein. Development 129, 123–131. 70 Wester, K., Digiuni, S., Geier, F., Timmer, J., Fleck, C., and Hülskamp, M. (2009). Functional diversity of R3 single-repeat genes in trichome development. Development 136, 1487–1496. Whittington, A.T., Vugrek, O., Wei, K.J., Hasenbein, N.G., Sugimoto, K., Rashbrooke, M.C., and Wasteneys, G.O. (2001). MOR1 is essential for organizing cortical microtubules in plants. Nature 411, 610–613. Wickstead, B., and Gull, K. (2007). Dyneins Across Eukaryotes: A Comparative Genomic Analysis. Traffic 8, 1708–1721. Wightman, R., and Turner, S.R. (2007). Severing at sites of microtubule crossover contributes to microtubule alignment in cortical arrays. Plant J. 52, 742–751. Willemsen, V., Friml, J., Grebe, M., van den Toorn, A., Palme, K., and Scheres, B. (2003). Cell Polarity and PIN Protein Positioning in Arabidopsis Require STEROL METHYLTRANSFERASE1 Function. Plant Cell 15, 612–625. Winter, C.G., Wang, B., Ballew, A., Royou, A., Karess, R., Axelrod, J.D., and Luo, L. (2001). Drosophila Rho-Associated Kinase (Drok) Links Frizzled-Mediated Planar Cell Polarity Signaling to the Actin Cytoskeleton. Cell 105, 81–91. Wong, L.L., and Adler, P.N. (1993). Tissue Polarity Genes of Drosophila Regulate the Subcellular Location for Prehair Initiation in Pupal Wing Cells. J. Cell Biol. 123, 209–221. Wu, J., Roman, A.-C., Carvajal-Gonzalez, J.M., and Mlodzik, M. (2013). Wg and Wnt4 provide long-range directional input to planar cell polarity orientation in Drosophila. Nat. Cell Biol. 15, 1045–1055. Wymer, C.L., Bibikova, T.N., and Gilroy, S. (1997). Cytoplasmic free calcium distributions during the development of root hairs of Arabidopsis thaliana. Plant J. 12, 427–439. Xu, J., and Scheres, B. (2005). Dissection of Arabidopsis ADP-RIBOSYLATION FACTOR 1 Function in Epidermal Cell Polarity. Plant Cell 17, 525–536. Xu, Z., and Dooner, H.K. (2006). The Maize aberrant pollen transmission 1 Gene Is a SABRE/KIP Homolog Required for Pollen Tube Growth. Genetics 172, 1251–1261. Xu, T., Wen, M., Nagawa, S., Fu, Y., Chen, J.-G., Wu, M.-J., Perrot-Rechenmann, C., Friml, J., Jones, A.M., and Yang, Z. (2010). Cell Surface- and Rho GTPase-Based Auxin Signaling Controls Cellular Interdigitation in Arabidopsis. Cell 143, 99–110. Xu, T., Nagawa, S., and Yang, Z. (2011). Uniform auxin triggers the Rho GTPase-dependent formation of interdigitation patterns in pavement cells. Small GTPases 2, 227–232. Xu, T., Dai, N., Chen, J., Nagawa, S., Cao, M., Li, H., Zhou, Z., Chen, X., De Rycke, R., Rakusová, H., et al. (2014). Cell Surface ABP1-TMK Auxin-Sensing Complex Activates ROP GTPase Signaling. Science 343, 1025–1028. Xu, X.M., Zhao, Q., Rodrigo-Peiris, T., Brkljacic, J., He, C.S., Müller, S., and Meier, I. (2008). RanGAP1 is a continuous marker of the Arabidopsis cell division plane. Proc. Natl. Acad. Sci. USA 105, 18637–18642. Yang, Z. (2008). Cell Polarity Signaling in Arabidopsis. Annu. Rev. Cell Dev. Biol. 24, 551–575. Yang, C., Axelrod, J.D., and Simon, M.A. (2002). Regulation of Frizzled by Fat-like Cadherins during Planar Polarity Signaling in the Drosophila Compound Eye. Cell 108, 675–688. Yu, H., Luo, N., Sun, L., and Liu, D. (2012). HPS4/SABRE regulates plant responses to phosphate starvation through antagonistic interaction with ethylene signalling. J. Exp. Bot. 63, 4527–4538. Zallen, J.A. (2007). Planar Polarity and Tissue Morphogenesis. Cell 129, 1051–1063. Zhang, Y., Xiao, Y., Du, F., Cao, L., Dong, H., and Ren, H. (2011). Arabidopsis VILLIN4 is involved in root hair growth through regulating actin organization in a Ca2+-dependent manner. New Phytol. 190, 667–682. Zhou, F.L., Zhang, W.G., Meng, X., Chen, G., and Wang, J.L. (2008). Bioinformatic analysis and identification for a novel antigen MLAA-22 in acute monocytic leukemia. Zhongguo Shi Yan Xue Ye Xue Za Zhi 16, 466–471. Zhou, R., Benavente, L.M., Stepanova, A.N., and Alonso, J.M. (2011). A recombineering-based gene tagging system for Arabidopsis. Plant J. 66, 712–723. Zimmermann, P., Hirsch-hoffmann, M., Hennig, L., and Gruissem, W. (2004). GENEVESTIGATOR. Arabidopsis Microarray Database and Analysis Toolbox. Plant Physiol. 136, 2621–2632. Zourelidou, M., Müller, I., Willige, B.C., Nill, C., Jikumaru, Y., Li, H., and Schwechheimer, C. (2009). The polarly localized D6 PROTEIN KINASE is required for efficient auxin transport in Arabidopsis thaliana. Development 136, 627–636. 71
