Abstract
The establishment of cell polarity is a prerequisite for many developmental processes. However, how it is achieved during tip growth in plants remains elusive. Here, we show that the RHO OF PLANTs (ROPs), ROP GUANINE NUCLEOTIDE EXCHANGE FACTORs (RopGEFs), and ROP GTPASE-ACTIVATING PROTEINs (RopGAPs) assemble into membrane domains in tip-growing cells of the moss Physcomitrium patens. The confinement of membrane domains requires redundant global inactivation of ROPs by PpRopGAPs and the PLECKSTRIN HOMOLOGY (PH) domain-containing RenGAP PpREN. Unexpectedly, PpRopGAPs and PpREN exert opposing effects on domain size and cell width upon overexpression. Biochemical and functional analyses indicate that PpRopGAPs are recruited to the membrane by active ROPs to restrict domain size through clustering, whereas PpREN rapidly inactivates ROPs and inhibits PpRopGAP-induced clustering. We propose that the activity- and clustering-based domain organization by RopGAPs and RenGAPs is a general mechanism for coordinating polarized cell growth and cell size regulation in plants.
Similar content being viewed by others
Introduction
Tip growth is essential for nutrient uptake, fertilization, and signal perception in various cells such as filamentous fungi1, pollen tubes and root hairs in seed plants2,3, and protonemata and rhizoids in mosses4. The specification of the growing tip is regulated by the conserved Cdc42/Rho/Rac small GTPases, which, in plants, involve the unique Rho-type family RHO OF PLANTS (ROPs)5. ROPs polarize tip growth by promoting calcium signaling6, cytoskeleton remodeling6,7, vesicle fusion8, and exocytosis9. To date, many ROP effectors have been identified. However, how active ROPs are confined to the growing tip is poorly understood.
Like Cdc42/Rho/Rac GTPases, plant ROPs cycle between an active GTP-bound form and an inactive GDP-bound form5. ROP GUANINE NUCLEOTIDE EXCHANGE FACTORS (RopGEFs) and ROP GTPASE-ACTIVATING PROTEINS (RopGAPs) convert ROP-GDP to ROP-GTP and ROP-GTP to ROP-GDP, respectively10. In many cells, ROPs are polarly localized and their localization depends on lipid modification, membrane lipid association, GTP/GDP-binding status, and membrane trafficking11,12,13,14,15. Among these, dynamic activation and inactivation by RopGEFs and RopGAPs, respectively, play a central role and have been proposed to involve a self-organization mechanism13,15. ROPs have been shown to form membrane clusters in root epidermal cells16,17, pollen tubes18, and root hairs19, suggesting the involvement of self-organization in tip-growing cells.
The self-organization model is based on the fact that ROPs are activated by RopGEFs, and inactivated and restricted by RopGAPs. Although much progress has been made in the regulation of RopGEFs recently11,12,13,14,15, the roles of RopGAPs in polar domain formation are less understood. There are two structurally distinct subtypes of ROP-related GAPs in land plants10,20: one contains a CDC42/RAC-INTERACTIVE BINDING (CRIB) motif before the GAP domain and the other bears a PLECKSTRIN HOMOLOGY (PH) domain, a GAP domain, and a coiled-coil (CC) motif-carrying tail, referred to as RopGAP and RenGAP, respectively5. RopGAPs and RenGAPs both stimulate GTP hydrolysis by ROPs21,22,23, but appear to have differential functions. RopGAPs in Arabidopsis and tobacco localize at the apical membrane of pollen tubes to inactivate ROPs locally23,24,25, while Arabidopsis ROP ENHANCER 1 (AtREN1), one of the RenGAPs, globally inhibits ROP signaling in the cytoplasm22. A recent study reports that AtREN1 also localizes at the shank membrane of root hairs and interacts with the tip-growth regulator ARMADILLO REPEAT ONLY PROTEINs (AROs)26,27, implying its potential function as a local inactivator. In Arabidopsis, only two RopGAPs are detectable in pollen tubes. However, their loss of function does not cause discernable growth defects24. These findings imply an intricate regulation of ROPs by GAPs and necessitate a comprehensive loss-of-function study of all RenGAPs and RopGAPs. However, the strong male fertility defects in Atren1 mutants22 preclude the generation of high-order mutants26.
The moss Physcomitrium patens (P. patens) represents one of the extant species closest to the land plant ancestor28,29. During the lifecycle of P. patens, spores grow into filamentous tissues (protonemata) and develop leafy gametophores at later stages29. The protonemata comprise two types of cells termed caulonema cells and chloronema cells, of which the former has fewer and smaller chloroplasts and grows faster30. Due to the high regenerative capacity, the protonemata can be indefinitely maintained in the laboratory31. This feature, together with the high rate of homologous recombination32, makes P. patens an excellent model for studying tip growth33. Previously, ROPs and ROP regulators have been reported to control protonema growth34,35,36,37,38. In this study, we show that PpRopGAPs and PpREN redundantly regulate tip growth by globally inhibiting ROP activity. Unexpectedly, they also oppositely regulate cell width. The diverged functions of GAPs suggest that ROP activity and spatial organization are coordinated to establish a polar domain, which may represent a common mechanism for tip growth and cell size regulation in plants.
Results
PpRopGAPs localize to the apical membrane and delimit the polar membrane domain
The genome of P. patens encodes one RenGAP (PpREN) and six RopGAPs (PpRopGAP1 to PpRopGAP6)20. To investigate how PpRopGAPs regulate cell growth, we examined the localization of endogenous PpRopGAPs by fusing the bright green fluorescent protein mNeonGreen (mNG) at the amino- (N-) terminus through homologous recombination. In protonemata, only PpRopGAP1 and PpRopGAP5 were detectable and PpRopGAP1 exhibited a relatively higher expression level. Both proteins were found at the tip of caulonema cells and branch initiation sites (Supplementary Fig. 1a, b). PpRopGAP1 fused with a carboxyl- (C-) terminal Citrine or mNG showed identical localization patterns (Supplementary Fig. 1c–e). Interestingly, PpRopGAP1 signals at the apex were relatively lower than those at the flanking regions (Supplementary Fig. 1d). Upon growth direction change, PpRopGAP1 at one side of the flanking regions extended toward the tip. The actin foci (labeled by Lifeact-mCherry) were disassembled and reassembled at a neighboring cytoplasmic region (Fig. 1a, Supplementary Movie 1). Meanwhile, the redistributed PpRopGAP1 generated a new low-intensity zone at the tip. During branch initiation and growth, a low-intensity zone was not observed at the expanding apex, although the fluctuation of actin foci was observed as seen in tip cells (Supplementary Fig. 1d, f).
As PpROPs and PpRopGEF4 are enriched at the apical membrane35,36,39, we labeled the endogenous PpRopGEF4 and PpROP4 with a C-terminal mCherry (PpRopGEF4-mCherry) and an internal mNG (PpROP4-mNG), respectively, and compared their localization patterns with PpRopGAP1. As shown in Fig. 1b–d, PpRopGEF4 was exclusively found at the apical dome. PpROP4 displayed a similar enrichment but labeled a slightly broader region and showed weak fluorescence on the remaining membrane. Noticeably, PpRopGAP1 marked a much wider region than PpRopGEF4. Similar to PpRopGAP1, PpRopGEF4 and PpROP4 displayed a small low-intensity zone at the apex. Time-lapse imaging revealed signal oscillation at the low-intensity zone (Supplementary Movies 2 and 3), implying that the apex is an actively remodeling region. On the membrane, all proteins were present in mobile particles of nanoscale size (< 1 µm) (Fig. 2a) and the PpRopGAP1 particles were more visible than those formed by PpROP4 or PpRopGEF4 (Supplementary Movie 4). In addition, 79% (n = 42 particles from 9 cells) and 80% (n = 44 particles from 9 cells) of PpROP4 and PpRopGAP1 particles, respectively, exhibited colocalization with PpRopGEF4 particles (Fig. 2g), implying that they are largely present in the same clusters. Taken together, these data indicate that endogenous ROPs, RopGEFs, and RopGAPs are organized into membrane domains with the boundary marked by RopGAPs. They rapidly diffuse and potentially form protein clusters within the domain.
PpRopGAP1 is recruited to transient membrane domains following PpRopGEF4 and PpROP4
To investigate the ability of PpROP4, PpRopGEF4, and PpRopGAP1 in forming domains, we next investigated their spatiotemporal relationship during polarity initiation in pre-branching subapical cells. As shown in Supplementary Fig. 1g, h, all proteins labeled the future branching site before bulge initiation. As in tip cells, PpRopGAP1 marked a broader region than PpRopGEF4. Interestingly, using time-lapse imaging we detected ectopic transient domains marked by PpROP4, PpRopGEF4, and PpRopGAP1 (Fig. 2b–f). These domains formed either at the apicolateral membrane of the subapical cell or at the basolateral membrane of the tip cell and persisted for several to tens of minutes. In some cases, the transient domain moved along the membrane and underwent disassembly and reassembly, suggesting that they were unstable. Nevertheless, PpROP4 and PpRopGEF4 were almost simultaneously recruited to these domains with PpROP4 occurring in a slightly wider time window (Fig. 2b, c, Supplementary Movie 5). By contrast, PpRopGAP1 was recruited much later than PpRopGEF4 (Fig. 2d–f, Supplementary Movie 6). These observations suggest that ROP-mediated domain assembly is initiated by unstable clustering of ROPs and RopGEFs and is presumably reinforced by subsequently recruited RopGAPs.
The N-terminal CRIB motif and a conserved pre-CRIB motif are necessary and sufficient for the membrane targeting of PpRopGAP1
We next investigated how PpRopGAP1 was targeted to membrane domains. Plant RopGAPs comprise a CRIB motif-containing amino terminus (Nter) and a GAP domain-containing carboxyl terminus (Cter)10,20. When tagged with mNG, the Nter portion of PpRopGAP1 (amino acid 1-193) was localized to the apical membrane and displayed an enrichment similar to the full-length PpRopGAP1 (Fig. 3a). Introducing a histidine-to-alanine mutation in the core CRIB motif (H165A)21 or removing the Nter portion (Δ1-193) completely abolished membrane localization, suggesting that polar localization of PpRopGAP1 is primarily mediated by the CRIB domain. Interestingly, when GAP activity was inhibited by an arginine-to-leucine mutation (R239L)23, the membrane localization of PpRopGAP1 was also lost (Fig. 3a). It is plausible that the mutated GAP domain may adopt a conformational change that hinders the binding of the CRIB domain with ROPs. As CRIB domains preferentially bind active ROPs21,23, these results suggest that RopGAPs are recruited to the membrane by active ROPs.
We next investigated the minimal portion of the CRIB domain for membrane targeting. The CRIB motif has a 16-amino-acid core peptide in animals40 and is present in plant RICs and RopGAPs21,23,41. The fully functional CRIB domain in plants appears to require additional sequences23,41,42, which have not been unequivocally identified. To characterize the functional region of a CRIB domain in RopGAPs, we obtained protein sequences carrying a CRIB motif and a RhoGAP domain from the InterPro database and performed alignment analysis. RopGAPs are present in land plants (Embryophyta), green algae Klebsormidium nitens, and protists of the Sar supergroup (stramenopiles, alveolates, and rhizaria) (Supplementary Data 1, Supplementary Fig. 2). In addition to the CRIB motif, we revealed a highly conserved region comprising ~20 amino acids before the CRIB motif (hereafter referred to as the PCRIB motif). The PCRIB motif contains abundant hydrophobic residues and a conserved (R/K)(R/K)SX4C module (Fig. 3b) and is predicted to adopt an α-helix fold (Supplementary Fig. 3)43. Interestingly, RICs also contain a conserved region before the CRIB motif (Supplementary Fig. 4a)42. Although the corresponding sequences are not similar to the PCRIB motifs in RopGAPs, they also tend to form α-helical structures (Supplementary Fig. 4b).
We mutated the two conserved arginine residues (R134 and R135) to leucine and assessed their effects on PpRopGAP1-PpROP4 binding in yeast-two-hybrid assays. Although R134L or R135L alone did not abolish the interactions between PpRopGAP1 and constitutively active PpROP4 (ROP4CA), the interaction was lost when both mutations were introduced (Fig. 3c). Truncated forms of the CRIB domain containing either the CRIB motif alone (S150-S193) or the PCRIB plus CRIB motif (Q122-S193) showed interactions with PpROP4CA. However, the interaction between the CRIB motif-only fragment and PpROP4 was relatively weak (Fig. 3c). When the H165A mutation was introduced, the binding ability of the CRIB motif was completely blocked. As the PCRIB motif is rich in hydrophobic residues, we speculate that the R134/135 L mutation may disrupt intramolecular interactions and influence protein stability. Indeed, the R134/135 L double-mutant protein was poorly expressed in contrast to those carrying one mutation (Fig. 3d), although the expression at the transcription level was not affected (Supplementary Fig. 5). Moreover, we failed to express R134/135L mutant proteins using bacterial expression systems in vitro. These data suggest that the PCRIB motif plays a critical role in protein folding. We next expressed the mutated forms of the CRIB domain in planta and examined their localization. Interestingly, the CRIB motif alone was only found in the cytoplasm (Fig. 3e). The addition of the PCRIB motif led to a membrane localization pattern similar to wild-type (WT) PpRopGAP1, albeit the overall expression was lower. Importantly, the R134L or R135L mutation severely impaired membrane localization, indicating that the PCRIB motif is part of a functional CRIB domain in vivo. Based on these results, we concluded that the membrane targeting of PpRopGAP1 depends on the CRIB core motif and is facilitated by a conserved PCRIB motif that stabilizes protein conformation.
PpRopGAPs and PpREN redundantly regulate tip growth and development
To investigate how PpRopGAPs regulate tip growth, we generated mutants of all PpRopGAPs by introducing mutations in the CRIB motif or the GAP domain using the clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9-based genome editing technology44. We first obtained a quintuple mutant carrying frameshift mutations in PpRopGAP1/3/6 and in-frame deletions in PpRopGAP2/4. This quintuple mutant appeared superficially wild-type, suggesting a functional redundancy between PpRopGAPs. Further editing of PpRopGAP2/4/5 with oligonucleotide templates45 allowed us to obtain two types of sextuple mutants: one carrying frameshift mutations in all PpRopGAPs (hereafter referred to as ropgap mutant), and the other containing in-frame deletions in PpRopGAP2/4 but frameshift mutations in other PpRopGAPs (referred to as ropgap hypomorphic mutant or ropgap-HM) (Supplementary Fig. 6). The presence of frameshift mutations suggests that the ropgap mutant is potentially null. To verify this, we inserted an mNG or a Citrine tag to the C-terminus of endogenously mutated PpRopGAP1. As expected, we did not detect any expression (Supplementary Fig. 7a, b). Furthermore, overexpression of the mutated PpRopGAP1 fused with mNG did not produce detectable fluorescence either (Supplementary Fig. 7c). These results strongly indicate that PpRopGAP expression is substantially blocked at the protein level by frameshift mutations and the ropgap mutant is null. Interestingly, the growth of protonemata and gametophores was only marginally affected in the ropgap mutant (Fig. 4a; Supplementary Fig. 8a). Therefore, PpRopGAPs are not essential for P. patens viability and may play a minor role in development.
Because the knockout of PpREN in P. patens (hereafter referred to as ren mutant) via homologous recombination revealed no discernable growth defects (Fig. 4a), we suspected that PpREN and PpRopGAPs have redundant roles, and thus knocked out ren in the ropgap or ropgap-HM background. As expected, the ropgap, ren and ropgap-HM, ren septuple mutants exhibited strong defects in protonema growth and gametophore development when cultured on standard BCDAT medium (Fig. 4a–f; Supplementary Fig. 8a–c). The protonema growth defects were due to a significant loss of fast-growing caulonema cells (Fig. 4c) and could be rescued by overexpressing PpREN or PpRopGAP1 fused with an N-terminal Cerulean, demonstrating the functional redundancy between PpREN and PpRopGAPs (Fig. 4b, c, e; Supplementary Fig. 8d). When cultured on the nutrient-limiting BCD medium, which stimulates caulonema cell development, mutant plants developed wavy filaments. The caulonema cells were shorter and grew slower than WT cells (Fig. 4d–f). Similarly, the ropgap-HM, ren mutants exhibited a decrease in cell length and growth and could not regenerate efficiently after protoplasting (Supplementary Fig. 8e–g). These findings indicate that PpREN and PpRopGAPs play overlapping roles in regulating tip growth and development in P. patens.
We have previously shown that overexpressed PpREN is mostly localized in the cytoplasm36. This fact contradicts the functional redundancy between PpREN and PpRopGAPs. We examined the endogenous PpREN with mNG fusion and confirmed its cytoplasmic localization (Fig. 4g). Although we occasionally observed dot-like structures in the cytoplasm or associated with membranes, we did not detect clear enrichment or colocalization of PpREN1 with vesicles, as has been reported for the Arabidopsis AtREN122,26. We next examined the function of membrane localization of PpRopGAP1 by performing rescue experiments. The Nter portion and the inactive form of PpRopGAP1 did not rescue cell growth phenotypes (Fig. 4h). Interestingly, PpRopGAP1 mutants carrying the H165A mutation or lacking the Nter portion could partially rescue growth phenotypes although they completely lost membrane localization (Fig. 3a). These results indicate that RopGAPs can globally inactivate ROPs as RenGAPs do. More importantly, local inactivation of ROPs by RopGAPs is not essential for polarity formation and tip growth (Fig. 4h).
PpRopGAPs and PpREN restrict branch initiation
The wavy filaments in ropgap, ren mutants imply defects in cell morphology regulation. We thus examined the development of side branches. In WT cells, side branches sequentially develop on subapical cells with the latest one formed commonly on the second subapical cell. We quantified the percentages of subapical cells that had developed a side branch. In the ropgap sextuple mutant or ren mutant alone, branch formation was normal. However, there was a clear increase of branches on the second subapical cell in ropgap, ren septuple mutants (Fig. 5a). Complementation with PpREN fully rescued the branching phenotype, while the overexpression of PpRopGAP1 drastically inhibited branch formation (Fig. 5a). PpRopGAP1 overexpression in WT plants induced a similar effect, demonstrating that excess PpRopGAP1 can inhibit branching. These data indicate that branch initiation requires an optimal level of active ROPs and PpRopGAP1 has a stronger effect than PpREN in limiting branch formation. This notion is consistent with the localization of PpRopGAPs rather than PpREN at the branch initiation sites (Fig. 2d; Supplementary Fig. 1), where they may locally inactivate and cluster ROPs.
In addition to branch formation phenotypes, we found an increase in bulge width in ropgap, ren mutants (Fig. 5b, c). Since bulge size can affect the positioning of the division plane36,46, we performed time-lapse imaging to examine the division process. Interestingly, nuclear migration and cell division were largely normal. However, during cytokinesis, the phragmoplast frequently attached to one side of the basal region of bulges and expanded directionally toward the other side (Fig. 5d, Supplementary Movie 7). Although cell division was completed, cytokinesis was significantly delayed due to an increase in the duration of phragmoplast expansion (Fig. 5e). We concluded that PpREN and PpRopGAPs are not essential for phragmoplast guidance but could indirectly influence cell division by altering bulge morphology. In support of this notion, PpRopGAP1 overexpression strongly delayed bulge emergence (Fig. 5a) and could lead to division plane mispositioning when the cells failed to initiate bulges (Fig. 5f). Such phenotypes are also observed in rop mutants or cells treated with actin inhibitors36,46, supporting that PpRopGAPs, PpROPs, and actin may function in a common pathway. Interestingly, the width and height of bulges in PpRopGAP1 overexpression lines were not significantly different from those of WT cells (Fig. 5c). Therefore branching defects are likely caused by reduced polarity initiation rather than by global inhibition of polar cell growth. This notion is in line with the facts that PpRopGAP1 overexpression does not markedly affect cell length (Fig. 4f) and the apical growth of bulges (i.e. height in Fig. 5c) is normal in the ropgap, ren septuple mutant. Taken together, these results indicate that PpREN and PpRopGAPs function to restrict branch initiation and suggest that polarity initiation and polar cell growth have different sensitivities to ROP activity, with the former more sensitive to ROP inactivation and the latter more sensitive to ROP hyperactivation.
Membrane distribution of PpROP4 is altered in the ropgap, ren mutants
The overall phenotypes in ropgap, ren mutants suggest a dysregulation of cell polarity. Therefore we examined the localization of PpROP4 in tip cells. Surprisingly, the polar localization of PpROP4 was only mildly affected (Fig. 6a). The boundary between the apical dome and basal membrane became less evident (Fig. 6b). Moreover, we observed a decrease in PpROP4 enrichment at the tip (Fig. 6c). These results suggest that ROPs are less concentrated at the apical region in the absence of PpRopGAPs and PpREN. Interestingly, before tip cell division, we observed early accumulation of PpROP4 in ropgap, ren mutants (11/11 cells) at the lateral membrane around the interphase nucleus (Fig. 6d). If the cells were bent, stronger accumulation was always found at the protruding side. After cytokinesis, a polar localization was established at the apical end of the basal daughter cell (i.e. subapical cell), reminiscent of the branch initiation site. Such a phenomenon was never observed in WT cells. These data are consistent with the premature emergence of side branches in ropgap, ren mutants (Fig. 5a). Taken together, our results indicate that PpRopGAPs and PpREN facilitate ROP concentration at the polarizing domain but inhibit domain initiation.
RopGAPs and REN play antagonistic roles in cell width regulation
Although loss-of-function analyses reveal redundant roles of PpRopGAPs and PpREN in tip growth and branch formation, the dramatic effects of ectopically expressed PpRopGAP1 rather than PpREN on branch reduction suggest that their functions are not identical (Fig. 5a, f). Indeed, when PpREN or PpRopGAP1 was overexpressed in ropgap, ren septuple mutants, they fully rescued cell growth phenotypes (Fig. 4c, e) but oppositely increased and decreased cell width, respectively (Fig. 7a, b). To verify such a striking difference, we overexpressed PpREN and PpRopGAP1 in WT plants and confirmed that they could oppositely influence cell width (Fig. 7c). Interestingly, we did not detect a significant change in cell width in the ren or ropgap mutant alone. Cell width was only marginally increased in ropgap, ren septuple mutants (Fig. 7b). Because loss-of-function and gain-of-function of ROPs can both lead to short and round cells34,35,36,37,39, we reasoned that cell width regulation requires an optimal amount of active ROPs and this process exhibits distinct sensitivities to changes in the levels of active or inactive ROPs. The increase of active ROPs in ropgap or ren mutants was not sufficient to induce a change in cell width, while this effect could be enhanced in ropgap, ren septuple mutants likely due to their redundant functions in globally inactivating ROPs. By contrast, the inactivation of ROPs by excess PpREN or PpRopGAP1 could effectively alter cell width and the opposite effects are due to their distinct binding capacities and GAP activities (see below and Discussion).
PpRopGAP1 exhibits higher binding capacity to active PpROP4 than PpREN
To explore how PpRopGAP1 and PpREN interact with ROPs, we performed yeast-two-hybrid analyses and pull-down experiments. As PpROPs are functionally redundant20,34,36, we only focused on PpROP4. As expected, the full-length PpRopGAP1 interacted with PpROP4CA (represented by a G15V mutation), but not the WT or dominant-negative form (PpROP4DN, represented by a D20N mutation) in yeast-two-hybrid assays47,48 (Fig. 8a). The Nter portion (1-193) and the Cter portion (194-469) showed similar binding ability and specificity (Fig. 8b). Due to this dual-binding capacity, the H165A or R239L mutation of PpRopGAP1 alone did not abolish the interaction. However, when R239L was introduced into the Cter, the binding with PpROP4CA was abolished. In the pull-down experiments, PpRopGAP1 fused with maltose binding protein (MBP) displayed preferential interactions with Glutathione S-transferase (GST)-tagged PpROP4CA and only weakly interacted with PpROP4 or PpROP4DN (Fig. 8c, Supplementary Fig. 9a). The H165A or R239L mutants exhibited a strong and a mild decrease in binding capacity, respectively (Fig. 8d). Interestingly, we failed to detect interactions between PpROP4CA and PpREN or between PpROP4CA and the PpREN GAP domain in yeast-two-hybrid analyses (Supplementary Fig. 9b). PpROP4CA-PpREN interactions could only be detected via western blotting after pull-down (Fig. 8e). These data suggest that PpREN has a much weaker binding ability than PpRopGAP1 toward active PpROP4. However, our western blotting analyses did not reveal differences in the binding ability of PpREN and PpRopGAP1 toward PpROP4DN (Supplementary Fig. 9c). Additionally, PpREN-PpROP4CA binding was slightly decreased and increased by deleting the PH domain (Δ1-162) and the CC motif-containing tail (Δ375-921), respectively. However, an arginine-to-leucine mutation (R204L) that blocks the GAP activity did not affect PpREN-PpROP4CA interaction. Therefore, the PpREN-PpROP4 binding appears to be regulated by intramolecular interactions and does not require GAP activity. Together, our data suggest that PpRopGAP1 has a stronger binding capacity to active PpROP4 than PpREN, and the binding ability of PpRopGAP1 and PpREN with active PpROP4 is differentially regulated by their non-catalytic domains.
PpREN displays ten-fold higher GAP activity than PpRopGAP1
We next examined the GAP activity of PpRopGAP1 and PpREN in vitro. To our knowledge, the GTPase activity of PpROPs has not been tested to date. Therefore, we purified PpROP4 fused with polyhistidine (His) tags and examined its GTPase activity. Similar to Arabidopsis AtROP147, PpROP4 exhibited a dose-dependent intrinsic GTPase activity (Supplementary Fig. 10a). The addition of either MBP-PpRopGAP1 or MBP-AtRopGAP1 efficiently stimulated His-PpROP4- or His-AtROP1-catalyzed GTP hydrolysis (Fig. 8f; Supplementary Fig. 10b). Interestingly, although PpROP4 was more active than AtROP1 (Supplementary Fig. 10a), AtRopGAP1 exhibited relatively higher GAP activity than PpRopGAP1. Previously, the CRIB domain has been shown to enhance the GAP activity of RopGAPs in flowering plants21,23. We generated truncated forms of PpRopGAP1 and confirmed that the CRIB domain was also necessary for the full activity of PpRopGAP1 (Supplementary Fig. 10c). However, in contrast to the strong inhibition of GAP activity by the R239L mutation, the H165A mutation exhibited a mild effect on GTP hydrolysis (Supplementary Fig. 10d), demonstrating that the CRIB domain is not essential for but functions to promote GAP-stimulated GTP hydrolysis. Compared to PpRopGAP1, PpREN stimulated GTP hydrolysis by His-PpROP4 at a concentration about ten times lower (Fig. 8g). The GAP activity was severely reduced when the R204L mutation was introduced. Truncated proteins lacking the PH domain, CC motifs, or both did not significantly influence PpREN activity (Supplementary Fig. 11a). However, when the truncated proteins, as well as the R204L mutant protein, were expressed in P. patens, they failed to rescue cell growth defects (Supplementary Fig. 11b). Therefore, the PH domain, CC motifs, and GAP activity are required for PpREN function in vivo. Taken together, our results demonstrate that PpREN and PpRopGAP1 are bona fide GAPs but exhibit distinct biochemical activities.
Overexpression of PpREN disrupts the membrane clustering of PpRopGAP1
We next asked how PpREN and PpRopGAP1 differently alter cell width. Since PpRopGAP1 delimits the apical membrane domain and the CRIB domain is necessary and sufficient for its membrane targeting, we reasoned that PpRopGAP1 was recruited to the membrane by active ROPs and formed clusters to limit ROP diffusion. Indeed, RopGAPs have been shown to form dimers in vitro49 and are present in dispersive particles in vivo (Fig. 2a). Because PpRopGAP1 but not PpREN or its GAP domain was efficiently pulled down by GST-PpROP4CA (Fig. 9a), it appears unlikely that PpREN antagonizes PpRopGAP1 through competitive binding with PpROP4. However, when the Nter portion of PpRopGAP1 was fused with the GAP domain of PpREN, its membrane targeting ability was severely blocked (Fig. 9b, c). When PpREN was overexpressed, the endogenous PpRopGAP1 exhibited a similar reduction but to a lesser extent in membrane association (Fig. 9d, e). Noticeably, PpRopGAP1 decorated a broader region along the subapical membrane, although the mobility of PpRopGAP1-labeled particles was not affected (Fig. 9d–g). These data suggest that the expansion of the PpRopGAP1-labeled domain is not caused by changes in the diffusion ability or the assembly of PpRopGAP1 particles. Given the high activity of PpREN and the dependence of active ROPs in RopGAP1 recruitment, we concluded that the coalescence of PpRopGAP1 particles into large domains requires a substantial amount of active ROPs and relatively stable interactions between PpRopGAPs and PpROPs, which could be impeded by rapid inactivation of ROPs.
Discussion
In this study, we report that PpRopGAPs and PpREN play overlapping roles in globally inactivating PpROPs and facilitating ROP accumulation in membrane domains during tip growth. Meanwhile, they have differential biochemical activities and act oppositely to regulate ROP organization and cell width. We propose that an optimal amount of active ROPs and proper clustering are necessary for tip growth and cell width regulation. However, cell width is more sensitive to the size of polar domains and is balanced by the antagonistic action of clustering-promoting RopGAPs and potent ROP inactivator RenGAPs (Fig. 10).
The roles of global and local inhibitions of ROPs in polarity establishment
ROP-dependent polarity establishment has been well known to involve spatiotemporal control of ROP activity. In particular, local restriction of ROPs by RopGAPs is considered essential23,50,51,52. In this study, using complete knockout mutants, we unequivocally demonstrate that local inactivation of ROPs by RopGAPs is not essential for polarity formation (Fig. 4). By contrast, ROPs must be globally inactivated by PpRopGAPs and PpREN for efficient membrane enrichment (Fig. 6). Interestingly, even in the ropgap, ren mutants, polar localization of ROPs and tip growth are not completely abolished (Fig. 6a). As loss-of-function or overexpression of ROPs causes drastic growth and developmental defects34,35,36,37,47,48,53,54, other regulators such as RopGDIs may also have overlapping roles with RopGAPs and RenGAPs. Indeed, RopGDIs have been reported to restrict polarity formation55,56,57 and are partly redundant with RopGAPs in flowering plants24. If the local restriction of ROPs is not essential for polarity initiation, how could ROPs segregate into polar membrane domains? One explanation is that globally inhibited ROPs could be locally activated by RopGEFs19,58 and membrane lipids18,48,59 and the activated ROPs are sufficient to assemble a polar membrane domain. This possibility is consistent with membrane cluster formation of ectopically expressed RopGEFs and ROPs in the absence of RopGAPs51. More importantly, we observed a much earlier recruitment of PpROP4 and PpRopGEF4 in transient domains than PpRopGAP1 (Fig. 2), implying that RopGAPs are not required for polar domain initiation. Then how do ROP-related GAPs contribute to polarity establishment? One major function of GAPs is to maintain an optimal level of active ROPs, which is important for restricting polarity initiation and preventing overactivation-induced depolarization. This is accomplished by the synergistic action of RopGAPs and RenGAPs to globally inactivate ROPs (Figs. 4, 5). Another function of GAPs is to set up an optimally restricted polar membrane domain for cell width regulation. This is triggered by the recruitment of RopGAPs to cluster ROPs into a confined region, a process that depends on relatively stable interactions between RopGAPs and active ROPs. On the other hand, RenGAPs rapidly inactivate ROPs and counteract with RopGAPs in generating confined membrane domains, thus providing an elaborate mechanism to finetune cell size.
Functional conservation and diversification of RopGAPs and RenGAPs
RopGAPs and RenGAPs are exclusively found in land plants20 and green algae K. nitens, suggesting that they emerged around the time of the colonization of land by plants or earlier. However, the presence of distinct regulatory domains as well as biochemical characteristics indicates that they may be regulated in different ways. Indeed, the CRIB domain in RopGAPs enhances the GAP activity21,23 and targets RopGAPs to the membrane by interacting with active ROPs (Fig. 3). By contrast, the activity of RenGAPs appears to be not influenced by the PH domain or the CC tail domain (Supplementary Fig. 11a)22. An early study reveals a possible role of the CC tail of AtREN1 in association with exocytic vesicles22. Recently, AtPHGAPs have been shown to be stabilized by phosphorylation in the CC tail domain, which consequently inactivates AtROP2 at the indenting regions in pavement cells60,61. RenGAPs could also potentially associate with the membrane and vesicles via the lipid-binding PH domain26. However, whether this mechanism is widely conserved requires further investigation because the membrane association ability of RenGAPs in moss protonema cells (Fig. 4g), pollen tubes22, and root hairs26 seems to be different.
Even within the same subfamilies, sequence and functional analyses suggest diversifications. For instance, the PCRIB motif is absent in two subclades including AtRopGAP5 (Supplementary Figs. 2, 3), implying that some members of RopGAPs may employ different mechanisms for regulation. In barley Hordeum vulgare, HvMAGAP1 localizes to microtubules through its C-terminal tail and cooperates with HvRACB, a ROP GTPase family member, to regulate microtubule organization during fungal infection62,63. Recent studies have also revealed diverse functions of RenGAPs in Arabidopsis. AtREN2/AtPHGAP1 and AtREN3/AtPHGAP2 are located at the CDZ and regulate division plane orientation in root cells64. In pavement cells, they negatively regulate AtROP2 on the anticlinal surface of indenting regions but are degraded by brassinosteroid signaling in the lobe to promote interdigitation60,61. In contrast, AtREN1 associates with secretory vesicles and controls pollen tube growth by globally inactivating ROPs22. P. patens has only one RenGAP PpREN. Although it plays a similar role in tip growth to AtREN1, it also labels the CDZ and this localization requires the CC motif-containing tail but not the PH domain or GAP activity (Supplementary Fig. 11c). However, we did not reveal obvious defects in division plane orientation in the ren or ropgap, ren mutants. In addition, none of the PpRopGAPs was found at the CDZ, thus excluding possible genetic redundancy between PpREN and PpRopGAPs. Whether PpRopGAPs and PpREN play a direct role in cell division remains an open question.
Methods
Moss strains and culture conditions
All strains used in this study were derived from the Gransden ecotype of Physcomitrium patens. Mosses were routinely grown on standard solid BCDAT medium under continuous white light at 23 ~ 25 °C. For colony growth analyses, approximately equal amounts of protonema tissues were inoculated on BCDAT plates and cultured for two weeks. Colonies were photographed every two days with a digital camera. Mosses used for high-resolution imaging were cultured for 5–7 days on a thin layer of BCD medium in 35 mm imaging dishes.
Sequence alignment and phylogenetic analyses
P. patens genes encoding RopGAPs, RenGAPs, and RopGEFs were characterized by BLASTing the Arabidopsis homologs against the P. patens genome v3.3 in the Phytozome database (https://phytozome-next.jgi.doe.gov). A total of six RopGAPs, one RenGAP, and six RopGEFs were characterized, which was consistent with previous reports20,38. To isolate RopGAPs in other species, protein isoforms carrying a CRIB domain (IPR000095) and a RhoGAP domain (IPR000198) in the InterPro database (https://www.ebi.ac.uk/interpro/search/ida) were retrieved through domain architecture search. A total of 1 276 sequences were isolated. 1 244 of them were successfully mapped in the UniProt database (http://www.uniprot.org) and used for alignment and phylogenic analyses. Alignment was performed with the Clustal Omega service in Jalview (https://www.jalview.org). Phylogenetic trees were constructed by MEGA11 (https://www.megasoftware.net) using the neighbor-joining method with a 1 000 bootstrap value. Trees were rendered with the iTOL tool (https://itol.embl.de). Domain characterization was performed with the Conserved-Domain Search service from National Center for Biotechnology Information (https://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi). Among the 1 244 sequences, 1 063 were characterized to contain a conserved pre-CRIB motif preceding the core CRIB motif. Homologs of RenGAPs were isolated similarly by searching for sequences carrying a pleckstrin homology domain (IPR001849), a RhoGAP domain (IPR000198), and a ternary complex factor MIP1-leucine-zipper domain (IPR025757). A total of 985 sequences were isolated. Homologs of RopGAPs and RenGAPs were found in land plants and the green algae Klebsormidium nitens but not in other available genomes of plants including red algae and other green algae in the InterPro or Uniprot databases.
Plasmid construction
For homologous recombination-mediated knockout of PpREN and knock-in of PpROP4, PpREN, PpRopGAPs, and PpRopGEF4, genomic sequences at the length of 700 ~ 2000 bp were amplified from genomic DNAs with PrimeStar Max DNA polymerase (Takara, cat. R045A). The resulting fragments were cloned into vector backbones derived from pPY22 (mNeonGreen-Hygromycin B) or pPY23 (mNeonGreen-Nourseothricin) using the In-Fusion Snap Assembly Kit (Takara, cat. 638948). For PpREN knockout, the entire coding region was replaced by a Nourseothricin-resistance gene expression cassette. For N-terminal knock-in, the codon-optimized mNeonGreen sequence36 was inserted to the 3’ of the start codon of PpRopGAPs. For C-terminal knock-in, the mNeonGreen (or Citrine)-Hygromycin B or mCherry-Hygromycin B cassette was inserted to the 3’ end of the coding region of PpRopGAP1 or PpRopGEF4, respectively. For internal tagging of PpROP4, the mNeonGreen sequence was inserted between the Gly134 and Ala13535. Overexpression plasmids were constructed as follows: First, the coding sequences were cloned into a pENTR plasmid in-frame with an N-terminal Cerulean (for rescue experiments) or mNG (for localization analysis) by In-Fusion Snap Assembly. The resulting fusion protein-coding fragment was then inserted into the vector pPY138 via LR reactions (Gateway LR Clonase Enzyme Kit, Thermo, cat. 11791019) for constitutive overexpression under the control of EF1α promoter. Intron-free coding sequences used in this study were amplified from total cDNAs reverse transcribed using the Hifair III 1st Strand cDNA Synthesis Kit (Yeason, cat. 11139ES10). The sequences were inserted into pGBKT7/pGADT7 vectors for yeast-two-hybrid analyses or into pGEX-GST/pET28a-MBP vectors for protein expression by In-Fusion Snap Assembly. Mutations were introduced by PCR amplification of plasmids with mutation-carrying primers and subsequent transformation into DH5α bacteria. In the Phytozome database, there are three isoforms of PpREN. Our cDNA amplification isolated the shortest one that comprises 921 amino acids and lacks several unconserved residues in the PH domain.
For CRISPR/Cas9 knockout of PpRopGAPs, single-guide RNA (sgRNA) target sequences (Supplementary Fig. 6) were selected at conserved regions in the CRIB domain or GAP domain. They were synthesized as oligonucleotides and annealed to form double-stranded DNAs carrying overhangs. Subsequently, the DNAs were inserted into the plasmid pPY156, which carried a Hygromycin B-resistance gene expression cassette, between the pU6 promoter and sgRNA scaffold via Bsa I digestion and T4 DNA ligase-mediated ligation (Takara, cat. 6023). For oligonucleotide-dependent mutation knock-in, the templates (shown in Supplementary Fig. 6) were synthesized as oligonucleotides as described before45. Detailed information about the plasmids used in this study is available in Supplementary Table 1.
Moss transgenesis
Our moss transformation followed the previously published protocol65. In brief, the moss protonemata were collected and digested with driselase to obtain protoplasts. The protoplasts were transformed with 30 µg linearized plasmids for homologous recombination. If the homologous templates did not carry an antibiotics-selection marker, an appropriate antibiotics-resistance plasmid was cotransformed. In the CRISPR/Cas9 knockout experiments, 10 µg Cas9-expressing plasmids and 10 µg sgRNA plasmids were cotransformed. For oligonucleotide-dependent mutation knock-in, 10 µL of 50 µM single-stranded oligonucleotides were cotransformed together with the Cas9 and sgRNA plasmids. Transformed protoplasts were plated on cellophane-laid PRM plates and cultured for 10-14 days during which the cellophane disks were transferred to selection plates once for transient selection of CRISPR/Cas9 knockout lines or N-terminal tagging lines and twice for selection of C-terminal tagging lines, overexpression lines, or PpREN knockout lines, respectively. Positive colonies were propagated and verified by PCR-amplifying the inserted fragments with KOD-ONE DNA polymerase (Toyobo, cat. KMM-101).
Image acquisition, processing, and analysis
Protonema tissues cultured in BCD medium were imaged using an Axio Observer Z1 spinning-disk confocal microscope (Zeiss) equipped with a Yokogawa confocal spinning disk unit, a 63 × 1.40 NA or 100 × 1.40 NA oil-immersion objective lens, and a Hamamatsu camera. The excitation/emission wavelengths for green and red fluorescence imaging were 488/517 nm and 561/603 nm, respectively. Time-lapse imaging of cell growth or division was performed at an interval of two or three minutes. An interval of 500 ms was used for imaging of PpRopGAP1 dynamics on the membrane. FM4-64 staining was carried out by adding 10 µM of FM4-64 diluted in water to imaging samples and keeping the samples in the dark for 30 min. The dye solution was subsequently removed before confocal imaging. Images were acquired using Zeiss Zen software (Version 2.3, Blue Edition) and processed and analyzed with Fiji software (Version 2.14.0). As tip cells can vary in length, the quantification of cell length was performed using the subapical cells. Similarly, cell width was measured at the subapical region to ensure consistency. To visualize fluorescence distribution at the apical membrane, a segmented line with spline fit was manually drawn along the apical membrane to cover the entire dome region. The corresponding intensity profile was subsequently generated. The enrichment of PpROP4 at the apical membrane was quantified by calculating the ratio of peak intensity to base intensity in the profile. The membrane association of PpRopGAP1 or its Nter portion was measured by normalizing fluorescence intensity at a manually selected flanking region of the cell apex to a nearby area in the cytoplasm of the same size.
Yeast-two-hybrid analyses
Approximately 0.1 µg of plasmids that express the AD- or BD-fusion proteins were cotransformed with 100 µg carrier DNAs into competent cells of the AH109 strain. Transformed cells were cultured on a solid YPDA medium without Trp and Leu supplements for selection. Positive colonies were propagated and subcultured on a solid YPDA medium without Trp, Leu, His, and Ade supplements to test protein interactions.
Protein expression, purification, and pull-down assays
Proteins were expressed in 40 ml of the BL21(DE3) bacteria and harvested in lysis buffers containing 50 mM Tris-HCl (pH 7.5), 300 mM NaCl, 1% NP-40, 1 mM PMSF, and 1 mM DTT. The TALON Metal Affinity Resin (Takara, cat. 635501), Pierce Glutathione Agarose (Thermo, cat. 16100), and Amylose Resin (New England Biolabs, cat. E9021V) were used for purifying proteins fused with polyhistidine (His), Glutathione S-transferase (GST), and maltose binding protein (MBP) tags, respectively, following the manufacturer’s instructions. His-, GST-, and MBP-tagged Proteins were eluted with 300 mM imidazole, 10 mM glutathione, and 10 mM maltose, respectively, in 20–50 mM Tris-HCl pH 7.5 solutions. In the pull-down experiments, equal amounts of purified constitutively active (G15V) or dominant-negative (D20N) forms of PpROP4 were incubated with Amylose resins loaded with wild-type or mutated MBP-PpRopGAP1/MBP-PpREN for four hours. MBP-Citrine was used as a negative control. Resins were washed at least three times and then boiled. Denatured proteins were separated on polyacrylamide gels. For the PpRopGAP1 pull-down, GST-PpROP4 was directly stained with Coomassie blue for analysis. For PpREN pull-down, GST-PpROP4 was detected with an anti-GST primary antibody (26H1) of mouse origin (Cell Signaling Technology, cat. 2624-100 µl, 5000 × dilution) and an anti-mouse-HRP (goat) secondary antibody (Sigma, cat. A4416-1 mL, 2000 × dilution) by western blotting. Hemagglutinin (HA)-tagged mutant PpRopGAP1 in yeasts was detected with an anti-HA primary antibody (C29F4) of rabbit origin (Cell Signaling Technology, cat. 3724 T, 5000× dilution) and an anti-rabbit-HRP (goat) secondary antibody (Cell Signaling Technology, cat. 7074P2, 2000 × dilution).
In vitro GTPase assays
Before GTPase reactions, proteins were concentrated by ultracentrifugation in Ultrafiltration Spin Columns (Beyotime, cat. FUF051). The concentration of each protein was determined by running a parallel BSA gradient with 1–5 µL of purified protein on the polyacrylamide gel. GTPase assays were performed using the GTPase-Glo Assay Kit (Promega, cat. V7681) basically following the manufacturer’s instructions. In brief, each reaction was carried out in a 16 µL solution, which contained 5 µM GTP and appropriate amounts of ROP GTPases and/or GAPs. To measure the intrinsic GTPase activity of His-PpROP4 and His-AtROP1, 0.5 µg (1.3 µM), 1.0 µg (2.5 µM), and 5.0 µg (12.5 µM) of each protein were added to the reaction mixture, and the GTPase reaction was performed at room temperature for 1 h. After that, 16 µL of reconstituted GTPase-Glo reagent was added and mixed. The mixture was kept at room temperature for 30 min. Subsequently, the detection reagent was added and incubated for 5–10 min. Then fluorescence signals were detected with a microplate reader (BioTek Synergy H1). To measure GAP-stimulated GTPase activity, 0.1 µg (0.25 µM) of His-PpROP4 or His-AtROP1 was used for each reaction. 0.2 µg (0.13 µM), 0.4 µg (0.26 µM), and 0.8 µg (0.52 µM) of MBP-PpRopGAP1 or MBP-AtRopGAP1 were tested for GAP activity. 30 ng (0.013 µM), 60 ng (0.026 µM), and 120 ng (0.052 µM) of MBP-PpREN were tested for GAP activity. The GAP-stimulated GTPase reaction was performed at room temperature for 30 min. Each reaction was carried out with three replicates. The intensity was normalized to controls tested with purified MBP alone. To compare the GAP activity of wild-type and mutant forms of MBP-PpRopGAP1 or MBP-PpREN, an equal molar ratio of each protein was used as shown in the figure legends.
Statistical analyses
All sample sizes for analysis were indicated in the figure or figure legends. Data were presented as mean ± SD or box-and-whisker plots. Statistical analyses were performed using two-tailed student’s t-tests for comparing two groups or adjusted one-way ANOVA for comparing multiple groups. A significant difference between groups was determined when the p-value was less than 0.05.
Accession numbers
PpRopGAP1 (Pp3c4_16800); PpRopGAP2 (Pp3c3_5940); PpRopGAP3 (Pp3c13_4010); PpRopGAP4 (Pp3c4_24980); PpRopGAP5 (Pp3c26_4490); PpRopGAP6 (Pp3c26_5960); PpREN (Pp3c9_17460); PpRopGEF4 (Pp3c2_28420); PpROP4 (Pp3c10_4950); AtROP1 (AT3G51300); AtRopGAP1 (AT5G22400).
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Data availability
All relevant data are available within the manuscript and its supplementary materials. Source data are provided with this paper.
References
Riquelme, M. Tip growth in filamentous fungi: a road trip to the apex. Annu. Rev. Microbiol. 67, 587–609 (2013).
Scholz, P., Anstatt, J., Krawczyk, H. E. & Ischebeck, T. Signalling Pinpointed to the Tip: The Complex Regulatory Network That Allows Pollen Tube Growth. Plants (Basel) 9, 1098 (2020).
Mendrinna, A. & Persson, S. Root hair growth: it’s a one way street. F1000Prime Rep. 7, 23 (2015).
Rounds, C. M. & Bezanilla, M. Growth mechanisms in tip-growing plant cells. Annu. Rev. Plant Biol. 64, 243–265 (2013).
Feiguelman, G., Fu, Y. & Yalovsky, S. ROP GTPases structure-function and signaling pathways. Plant Physiol. 176, 57–79 (2018).
Gu, Y. et al. A Rho family GTPase controls actin dynamics and tip growth via two counteracting downstream pathways in pollen tubes. J. Cell Biol. 169, 127–138 (2005).
Zhou, Z. et al. Arabidopsis RIC1 severs actin filaments at the apex to regulate pollen tube growth. Plant Cell 27, 1140–1161 (2015).
Cui, X. et al. Arabidopsis SYP121 acts as an ROP2 effector in the regulation of root hair tip growth. Mol. Plant 15, 1008–1023 (2022).
Lee, Y. J., Szumlanski, A., Nielsen, E. & Yang, Z. Rho-GTPase-dependent filamentous actin dynamics coordinate vesicle targeting and exocytosis during tip growth. J. Cell Biol. 181, 1155–1168 (2008).
Brembu, T., Winge, P., Bones, A. M. & Yang, Z. A RHOse by any other name: a comparative analysis of animal and plant Rho GTPases. Cell Res. 16, 435–445 (2006).
Li, E. et al. Signaling network controlling ROP-mediated tip growth in Arabidopsis and beyond. Plant Commun. 4, 100451 (2023).
Ou, H. & Yi, P. ROP GTPase-dependent polarity establishment during tip growth in plants. N. Phytol. 236, 49–57 (2022).
Pan, X., Perez-Henriquez, P., Van Norman, J. M. & Yang, Z. Membrane nanodomains: Dynamic nanobuilding blocks of polarized cell growth. Plant Physiol. 193, 83–97 (2023).
Muller, S. Update: on selected ROP cell polarity mechanisms in plant cell morphogenesis. Plant Physiol. 193, 26–41 (2023).
Smokvarska, M., Jaillais, Y. & Martiniere, A. Function of membrane domains in rho-of-plant signaling. Plant Physiol. 185, 663–681 (2021).
Platre, M. P. et al. Developmental control of plant Rho GTPase nano-organization by the lipid phosphatidylserine. Science 364, 57–62 (2019).
Smokvarska, M. et al. A Plasma Membrane Nanodomain Ensures Signal Specificity during Osmotic Signaling in Plants. Curr. Biol. 30, 4654–4664 (2020).
Fratini, M. et al. Plasma membrane nano-organization specifies phosphoinositide effects on Rho-GTPases and actin dynamics in tobacco pollen tubes. Plant Cell 33, 642–670 (2021).
Fiona Fuchs V. A., et al. Nanodomain-mediated lateral sorting drives polarization of the small GTPase ROP2 in the plasma membrane of root hair cells. bioRxiv, https://doi.org/10.1101/2021.09.10.459822 (2021).
Eklund, D. M., Svensson, E. M. & Kost, B. Physcomitrella patens: a model to investigate the role of RAC/ROP GTPase signalling in tip growth. J. Exp. Bot. 61, 1917–1937 (2010).
Wu, G., Li, H. & Yang, Z. Arabidopsis RopGAPs are a novel family of rho GTPase-activating proteins that require the Cdc42/Rac-interactive binding motif for rop-specific GTPase stimulation. Plant Physiol. 124, 1625–1636 (2000).
Hwang, J. U., Vernoud, V., Szumlanski, A., Nielsen, E. & Yang, Z. A tip-localized RhoGAP controls cell polarity by globally inhibiting Rho GTPase at the cell apex. Curr. Biol. 18, 1907–1916 (2008).
Klahre, U. & Kost, B. Tobacco RhoGTPase ACTIVATING PROTEIN1 spatially restricts signaling of RAC/Rop to the apex of pollen tubes. Plant Cell 18, 3033–3046 (2006).
Hwang, J. U. et al. Pollen-tube tip growth requires a balance of lateral propagation and global inhibition of Rho-family GTPase activity. J. Cell Sci. 123, 340–350 (2010).
Fu, Y., Wu, G. & Yang, Z. Rop GTPase-dependent dynamics of tip-localized F-actin controls tip growth in pollen tubes. J. Cell Biol. 152, 1019–1032 (2001).
Kulich, I. et al. ARMADILLO REPEAT ONLY proteins confine Rho GTPase signalling to polar growth sites. Nat. Plants 6, 1275–1288 (2020).
Gebert, M., Dresselhaus, T. & Sprunck, S. F-actin organization and pollen tube tip growth in Arabidopsis are dependent on the gametophyte-specific Armadillo repeat protein ARO1. Plant Cell 20, 2798–2814 (2008).
Naramoto, S., Hata, Y., Fujita, T. & Kyozuka, J. The bryophytes Physcomitrium patens and Marchantia polymorpha as model systems for studying evolutionary cell and developmental biology in plants. Plant Cell 34, 228–246 (2022).
Rensing, S. A., Goffinet, B., Meyberg, R., Wu, S. Z. & Bezanilla, M. The Moss Physcomitrium (Physcomitrella) patens: A Model Organism for Non-Seed Plants. Plant Cell 32, 1361–1376 (2020).
Menand, B., Calder, G. & Dolan, L. Both chloronemal and caulonemal cells expand by tip growth in the moss Physcomitrella patens. J. Exp. Bot. 58, 1843–1849 (2007).
Cove, D. The moss Physcomitrella patens. Annu. Rev. Genet. 39, 339–358 (2005).
Schaefer, D. G. & Zryd, J. P. Efficient gene targeting in the moss Physcomitrella patens. Plant J. 11, 1195–1206 (1997).
Vidali, L. & Bezanilla, M. Physcomitrella patens: a model for tip cell growth and differentiation. Curr. Opin. Plant Biol. 15, 625–631 (2012).
Burkart, G. M., Baskin, T. I. & Bezanilla, M. A family of ROP proteins that suppresses actin dynamics, and is essential for polarized growth and cell adhesion. J. Cell Sci. 128, 2553–2564 (2015).
Cheng, X., Mwaura, B. W., Chang Stauffer, S. R. & Bezanilla, M. A fully functional ROP fluorescent fusion protein reveals roles for this GTPase in subcellular and tissue-level patterning. Plant Cell 32, 3436–3451 (2020).
Yi, P. & Goshima, G. Rho of Plants GTPases and cytoskeletal elements control nuclear positioning and asymmetric cell division during Physcomitrella patens branching. Curr. Biol. 30, 2860–2868 (2020).
Ito, K., Ren, J. & Fujita, T. Conserved function of Rho-related Rop/RAC GTPase signaling in regulation of cell polarity in Physcomitrella patens. Gene 544, 241–247 (2014).
Bascom, C. et al. Systematic survey of the function of ROP regulators and effectors during tip growth in the moss Physcomitrella patens. J. Exp. Bot. 70, 447–457 (2019).
Le Bail, A. et al. Analysis of the Localization of Fluorescent PpROP1 and PpROP-GEF4 Fusion Proteins in Moss Protonemata Based on Genomic “Knock-In” and Estradiol-Titratable Expression. Front. Plant Sci. 10, 456 (2019).
Burbelo, P. D., Drechsel, D. & Hall, A. A conserved binding motif defines numerous candidate target proteins for both Cdc42 and Rac GTPases. J. Biol. Chem. 270, 29071–29074 (1995).
Wu, G., Gu, Y., Li, S. & Yang, Z. A genome-wide analysis of Arabidopsis Rop-interactive CRIB motif-containing proteins that act as Rop GTPase targets. Plant Cell 13, 2841–2856 (2001).
Ntefidou, M. et al. Physcomitrium patens PpRIC, an ancestral CRIB-domain ROP effector, inhibits auxin-induced differentiation of apical initial cells. Cell Rep. 42, 112130 (2023).
Jumper, J. et al. Highly accurate protein structure prediction with AlphaFold. Nature 596, 583–589 (2021).
Collonnier, C. et al. CRISPR-Cas9-mediated efficient directed mutagenesis and RAD51-dependent and RAD51-independent gene targeting in the moss Physcomitrella patens. Plant Biotechnol. J. 15, 122–131 (2017).
Yi, P. & Goshima, G. Transient cotransformation of CRISPR/Cas9 and oligonucleotide templates enables efficient editing of target loci in Physcomitrella patens. Plant Biotechnol. J. 18, 599–601 (2020).
Ruan, J. & Yi, P. Exogenous 6-benzylaminopurine inhibits tip growth and cytokinesis via regulating actin dynamics in the moss Physcomitrium patens. Planta 256, 1 (2022).
Li, H., Lin, Y., Heath, R. M., Zhu, M. X. & Yang, Z. Control of pollen tube tip growth by a Rop GTPase-dependent pathway that leads to tip-localized calcium influx. Plant Cell 11, 1731–1742 (1999).
Kost, B. et al. Rac homologues and compartmentalized phosphatidylinositol 4, 5-bisphosphate act in a common pathway to regulate polar pollen tube growth. J. Cell Biol. 145, 317–330 (1999).
Schaefer, A., Miertzschke, M., Berken, A. & Wittinghofer, A. Dimeric plant RhoGAPs are regulated by its CRIB effector motif to stimulate a sequential GTP hydrolysis. J. Mol. Biol. 411, 808–822 (2011).
Oda, Y. & Fukuda, H. Initiation of cell wall pattern by a Rho- and microtubule-driven symmetry breaking. Science 337, 1333–1336 (2012).
Sternberg, H. et al. Formation of self-organizing functionally distinct Rho of plants domains involves a reduced mobile population. Plant Physiol. 187, 2485–2508 (2021).
Nagashima, Y. et al. A Rho-based reaction-diffusion system governs cell wall patterning in metaxylem vessels. Sci. Rep. 8, 11542 (2018).
Molendijk, A. J. et al. Arabidopsis thaliana Rop GTPases are localized to tips of root hairs and control polar growth. EMBO J. 20, 2779–2788 (2001).
Jones, M. A. et al. The Arabidopsis Rop2 GTPase is a positive regulator of both root hair initiation and tip growth. Plant Cell 14, 763–776 (2002).
Carol, R. J. et al. A RhoGDP dissociation inhibitor spatially regulates growth in root hair cells. Nature 438, 1013–1016 (2005).
Klahre, U., Becker, C., Schmitt, A. C. & Kost, B. Nt-RhoGDI2 regulates Rac/Rop signaling and polar cell growth in tobacco pollen tubes. Plant J. 46, 1018–1031 (2006).
Feng, Q. N. et al. Arabidopsis RhoGDIs Are Critical for Cellular Homeostasis of Pollen Tubes. Plant Physiol. 170, 841–856 (2016).
Denninger, P. et al. Distinct RopGEFs successively drive polarization and outgrowth of root hairs. Curr. Biol. 29, 1854–1865 (2019).
Stanislas, T. et al. Arabidopsis D6PK is a lipid domain-dependent mediator of root epidermal planar polarity. Nat. Plants 1, 15162 (2015).
Zhang, C. et al. ROPGAP-dependent interaction between brassinosteroid and ROP2-GTPase signaling controls pavement cell shape in Arabidopsis. Curr. Biol. 32, 518–531 (2022).
Lauster, T. et al. Arabidopsis pavement cell shape formation involves spatially confined ROPGAP regulators. Curr. Biol. 32, 532–544 (2022).
Hoefle, C. et al. A barley ROP GTPase ACTIVATING PROTEIN associates with microtubules and regulates entry of the barley powdery mildew fungus into leaf epidermal cells. Plant Cell 23, 2422–2439 (2011).
Hoefle, C., McCollum, C. & Huckelhoven, R. Barley ROP-Interactive Partner-a organizes into RAC1- and MICROTUBULE-ASSOCIATED ROP-GTPASE ACTIVATING PROTEIN 1-dependent membrane domains. BMC Plant Biol. 20, 94 (2020).
Stockle, D. et al. Putative RopGAPs impact division plane selection and interact with kinesin-12 POK1. Nat. Plants 2, 16120 (2016).
Yamada, M., Miki, T. & Goshima, G. Imaging Mitosis in the Moss Physcomitrella patens. Methods Mol. Biol. 1413, 263–282 (2016).
Acknowledgements
We thank Dr. Gohta Goshima for providing some of the plasmids, reagents, and moss strains. The original plasmids used for yeast-two-hybrid analyses and protein expression and some of our primary antibodies were gifts from Dr. Xingguang Deng, Dr. Dawei Zhang, and Dr. Kefeng Lu. We are grateful to Dr. Honghui Lin, Dr. Huapeng Zhou, Dr. Zhonghan Li, Jingjing Sha, and Xiaohui Liu for their kind technical support. This project was supported by the National Natural Science Foundation of China (32270776), Sichuan University (1082204112609, 2022SCUH0010, 0082604151432), and the People’s Government of Sichuan Province to P.Y.
Author information
Authors and Affiliations
Contributions
P.Y. conceptualized this project. J.R. and P.Y. conducted moss experiments. L.L. performed biochemistry experiments. L.L. and H.O. conducted yeast-two-hybrid assays. J.R., L.L., and P.Y. analyzed the data. P.Y. wrote the manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature Communications thanks Philipp Denninger and the other anonymous reviewer(s) for their contribution to the peer review of this work. A peer review file is available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Source data
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Ruan, J., Lai, L., Ou, H. et al. Two subtypes of GTPase-activating proteins coordinate tip growth and cell size regulation in Physcomitrium patens. Nat Commun 14, 7084 (2023). https://doi.org/10.1038/s41467-023-42879-y
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41467-023-42879-y
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.