These authors contributed equally to this work.
Arabidopsis COBRA-LIKE 10, a GPI-anchored protein, mediates directional growth of pollen tubes
Article first published online: 26 MAR 2013
© 2013 The Authors The Plant Journal © 2013 Blackwell Publishing Ltd
The Plant Journal
Volume 74, Issue 3, pages 486–497, May 2013
How to Cite
Li, S., Ge, F.-R., Xu, M., Zhao, X.-Y., Huang, G.-Q., Zhou, L.-Z., Wang, J.-G., Kombrink, A., McCormick, S., Zhang, X. S. and Zhang, Y. (2013), Arabidopsis COBRA-LIKE 10, a GPI-anchored protein, mediates directional growth of pollen tubes. The Plant Journal, 74: 486–497. doi: 10.1111/tpj.12139
- Issue published online: 23 APR 2013
- Article first published online: 26 MAR 2013
- Accepted manuscript online: 6 FEB 2013 03:08AM EST
- Manuscript Revised: 29 JAN 2013
- Manuscript Received: 15 OCT 2012
- Manuscript Accepted: 31 JAN 2012
- glycosylphosphatidylinositol-anchored protein;
- polar growth;
- asymmetric localization;
- wall organization;
- cell–cell communication
- Top of page
- Experimental procedures
- Supporting Information
Successful reproduction of flowering plants requires constant communication between female tissues and growing pollen tubes. Female cells secrete molecules and peptides as nutrients or guidance cues for fast and directional tube growth, which is executed by dynamic changes of intracellular activities within pollen tubes. Compared with the extensive interest in female cues and intracellular activities of pollen tubes, how female cues are sensed and interpreted intracellularly in pollen is poorly understood. We show here that COBL10, a glycosylphosphatidylinositol (GPI)-anchored protein, is one component of this pollen tube internal machinery. Mutations in COBL10 caused gametophytic male sterility due to reduced pollen tube growth and compromised directional sensing in the female transmitting tract. Deposition of the apical pectin cap and cellulose microfibrils was disrupted in cobl10 pollen tubes. Pollen tube localization of COBL10 at the apical plasma membrane is critical for its function and relies on proper GPI processing and its C-terminal hydrophobic residues. GPI-anchored proteins are widespread cell sensors in mammals, especially during egg-sperm communication. Our results that COBL10 is critical for directional growth of pollen tubes suggest that they play critical roles in cell-cell communications in plants.
- Top of page
- Experimental procedures
- Supporting Information
Double fertilization of flowering plants has been a fascinating research area for over a century. It is a process in which two immotile sperm cells carried within a pollen tube fertilize two female gametes, the egg and the central cell, to form the embryo and endosperm respectively (McCormick, 1993). After pollen lands on compatible stigmatic cells and germinates, pollen tubes grow deep inside pistils, penetrating different parts of the female tissues to arrive precisely at the micropyle of the ovule and deliver sperm (Cheung et al., 2010). Unlike most plant cells, a pollen tube grows in a polar, rather than in a diffuse way, forming a cylindrically shaped cell. Pollen tubes can grow in vitro when supplied with simple chemical ingredients but are able to achieve faster growth in the presence of female nutrients (Cheung et al., 1995; Wu et al., 2000; Tang et al., 2004). In addition, female tissues release molecules or proteins to guide the directional growth of pollen tubes (Higashiyama et al., 2001; Palanivelu et al., 2003; Marton et al., 2005; Palanivelu and Preuss, 2006; Chen et al., 2007; Okuda et al., 2009; Chae and Lord, 2011).
Polar growth is made possible through the dynamic maintenance of spatial asymmetry within pollen tubes. Elongating pollen tubes are structurally asymmetric. First, pollen tubes are composed of different intracellular zones defined by the presence of different organelles and distinct actin cytoskeletal structures (Cheung and Wu, 2007). Second, the cell wall of pollen tubes is organized differentially. That is, the pollen tube wall has a spatially distinct organization such that the apical wall consists mostly of a methylated pectic layer allowing cell expansion, while the shank wall is mostly made from demethylated pectin, whose rigidity maintains tube morphology and represses lateral growth (Geitmann, 2010). Third, the plasma membrane of pollen tubes is asymmetrical, such that signaling proteins or molecules are distributed in a spatially restricted manner (Benedikt, 2008), and in a way that resembles metazoan chemotactic cells in which signaling molecules such as Rac GTPases and phosphoinositides are distributed differentially either at the cell front or back for attractant sensing and directional movement (Kolsch et al., 2008).
To reach their precise targets, pollen tubes communicate with female cells along their path and adjust their growth axis in response to female cues (Johnson and Preuss, 2002; Cheung et al., 2010). However, compared with extensive studies on the identity of female cues as well as dynamic changes of intracellular activities within pollen tubes, little progress has been made in the identification of sensors in pollen tubes that perceive female cues and interpret them into dynamic changes of growth asymmetry.
Glycosylphosphatidylinositol (GPI)-anchored proteins (GPI-APs) are potential cell sensors in eukaryotes. GPI-APs generally consist of an N-terminal signal peptide (SP) for secretion, a hydrophilic middle region and a hydrophobic C-terminal SP for GPI processing and membrane anchoring (Chatterjee and Mayor, 2001; Ikezawa, 2002). GPI-APs are synthesized in the lumen of the endoplasmic reticulum (ER) via the sequential addition of monosaccharides, fatty acids, phosphoethanolamines and phosphatidylinositol (PI) (Ikezawa, 2002; Lalanne et al., 2004). Processed proteins are secreted from the ER and finally are anchored at the outer leaflet of the plasma membrane (PM). Some GPI-APs are further processed through proteolysis and released in the extracellular space (Ikezawa, 2002). Because the cell surface is the first region through which cells sense and respond to the environment, GPI-APs are likely candidates to perform such functions.
That GPI-APs play roles in pollen tube growth was suggested by genetic studies of GPI-AP-processing enzymes in Arabidopsis. The first processing step of GPI-APs at the ER was catalyzed by a GPI-N-acetylglucosaminyltransferase (GPI-GnT) complex that consists of at least six subunits in Arabidopsis, two of which regulate pollen tube growth. Mutations at SETH1 and SETH2, encoding the PIG-C and the PIG-A subunit of the GPI-GnT complex respectively, caused defects in pollen germination and tube growth, leading to male gametophytic lethality (Lalanne et al., 2004). Arabidopsis PEANUT1 (PNT1), encoding a mannosyltransferase (PIG-M) for GPI-AP-processing, was also involved in pollen tube growth in that its mutations substantially reduced male gametophytic transmission (Gillmor et al., 2005). Over 200 GPI-APs are encoded in the Arabidopsis genome (Borner et al., 2003; Elortza et al., 2003), dozens of which show pollen-specific or pollen-enriched expression patterns (Lalanne et al., 2004). Therefore, it was not known which GPI-APs contributed to the growth defects in SETH1, SETH2 and PNT1 loss-of-function mutants (Lalanne et al., 2004; Gillmor et al., 2005).
We show here that Arabidopsis COBL10 controls female-guided fast and directional growth of pollen tubes probably by regulating the deposition of the apical pectin cap and cellulose microfibrils. COBL10 belongs to the plant-specific COBRA family that encodes GPI-APs (Roudier et al., 2002; Brady et al., 2007). Members of the COBRA family mediate diverse cellular and developmental processes such as root growth (Schindelman et al., 2001; Brown et al., 2005; Roudier et al., 2005), pathogen resistance (Ko et al., 2006), stem strength (Li et al., 2003; Sindhu et al., 2007) and root-hair growth (Jones et al., 2006; Hochholdinger et al., 2008). Localization of COBL10 at the pollen tube plasma membrane was restricted to the most apical region and is dependent on its C-terminal hydrophobic residues as well as on an intact PIG complex. Our results show that COBL10 is an essential component of the internal machinery in pollen tubes that senses and interprets female signals.
- Top of page
- Experimental procedures
- Supporting Information
COBL10 loss of function caused gametophytic male sterility
Three T-DNA insertion lines within the COBL10 genomic region were available from Arabidopsis stock centers (Figure 1a). cobl10-1 (CS876299) and cobl10-2 (CS879411); both have insertions at the very end of the COBL10 coding sequence. Transcript analysis (Figure 1b) and DNA sequencing indicated that although expression levels of COBL10 in homozygous cobl10-1 and cobl10-2 mutants were equal to the expression level in wild type, the mutant transcripts potentially encode a COBL10 whose last nine amino acids were replaced by an asparagine (Asn) residue due to the T-DNA insertion (Figure 1c). cobl10-3 (CS875361) has a T-DNA inserted at the 3′ untranslated (UTR) region; this mutant showed no transcriptional changes (Figure 1a,b) and was not analyzed further.
Homozygous plants of cobl10 mutants were obtained at very low frequency from self-pollinated heterozygous plants (Table 1). No defects during vegetative growth were detected for homozygous cobl10 mutants, as expected given its restricted expression pattern (Brady et al., 2007). However, siliques of plants homozygous for either mutant allele were much shorter (Figure 1d–h), and contained on average four or five developing seeds in each silique (Figure 1l). Interestingly, these seeds were twice the size of the seeds from wild-type plants, mostly due to enlarged embryonic cells (Figure S1). Most ovules in siliques of homozygous cobl10 mutants were tiny, and still contained the egg cell and central cell at 48 h after pollination (HAP) (Figure 1k), a time at which embryo developmental processes had already started in wild type (Figure 1j) and suggesting failed fertilization rather than defective embryo development.
|Female X Male||Genotype||Expected||Observed ratio|
|cobl10-1+/− × cobl10-1+/−||COBL10+/+: +/−: −/−||1:2:1||100:126:7a|
|cobl10-1+/− × WT||COBL10+/+: +/−||1:1||121:108b|
|WT × cobl10-1+/||COBL10+/+: +/–||1:1||130:3c|
|cobl10-2+/− × cobl10-2+/−||COBL10+/+: +/−: −/−||1:2:1||76:92:4a|
|cobl10-2/– × WT||COBL10+/+: +/−||1:1||84:79b|
|WT × cobl10-2+/−||COBL10+/+: +/−||1:1||93:2c|
To find out which gametophyte was defective, we performed reciprocal crosses between cobl10 heterozygous mutants and wild type. The segregation ratios of progenies from these reciprocal crosses showed that mutant male gametophytes transmitted poorly while mutant female gametophytes were normal (Table 1), even though COBL10 was also expressed in ovules (Brady et al., 2007).
COBL10 is not essential for pollen development, hydration, or germination
There are different stages at which the male gametophyte could be defective, i.e. pollen development, germination, polar and directional tube growth, and sperm delivery (Johnson and Preuss, 2002). We therefore analyzed male gametophytes of cobl10 at different stages to determine the reason for poor transmission (Methods S1). Pollen viability staining with 4′,6-diamidino-2-phenylindole (DAPI) and Alexander's stain (Johnson-Brousseau and McCormick, 2004) showed no differences between wild type (Figure S2a) and cobl10 (Figure S2(b)). Scanning electron micrographs (SEM) showed that the pollen coat of cobl10 mutants (Figure S2(d)) was like that of wild type (Figure S2(c)). These results indicated that cobl10 pollen developed normally.
Next, we tested whether pollen hydration and germination was affected in cobl10 by analyzing pollen germination in vitro or on stigma. We introduced quartet1 (qrt) and ProLAT52:GUS into cobl10 to make scoring of germination more accurate. The qrt1 mutant contains pollen tetrads instead of separate pollen grains and ProLAT52:GUS allows histochemical staining of pollen tubes growing both in vitro and in vivo (Johnson-Brousseau and McCormick, 2004). Homozygous cobl10;qrt;ProLAT52:GUS pollen showed no differences in germination percentage and growth speed, compared with qrt;ProLAT52:GUS as wild type (Figure S3a,d). Germination and hydration on the stigma was not also affected (Figure S2(f)). These results indicated that COBL10 is not essential for pollen development, hydration, or germination.
Pollen tube growth in the transmitting tract is defective in cobl10
The distorted segregation ratios (Table 1) and near sterility of mutants (Figure 1) indicated that male transmission was disrupted in cobl10. Therefore, the lack of detectable defects for cobl10 pollen at developmental stages as well as during germination implied defects late in male gametophyte function, i.e. pollen tube growth. We therefore examined whether pollen tube growth was affected in cobl10. Surprisingly, no defects in germination, morphology and growth speed were detected for cobl10 pollen tubes growing in vitro (Figure S3).
We next examined pollen tube growth in pistils by histochemical analysis of emasculated wild-type pistils hand-pollinated with either cobl10;qrt;ProLAT52:GUS or ProLAT52:GUS pollen (WT). At 9 hours after pollination (HAP), WT pollen tubes reached the bottom of the pistil and most ovules were targeted by a pollen tube (Figure 2a). In comparison, cobl10 pollen tubes were slower (Figure 2b). We performed an additional time course analysis of pollen tube growth in vivo by analyzing pistils at progressive time points to determine in which female tissue(s) the growth of cobl10 pollen tubes slowed down. Pistils pollinated by cobl10 pollen at 0.5 HAP and 1.5 HAP showed a growth front of most of the tubes at the stigma and style, respectively (Figure 2b), similar to wild type (Figure 2a). However, cobl10 pollen tubes grew slower than those of wild type at time points from 2.5 HAP onwards, when the growth front of most WT tubes were in the transmitting tract, the so-called ‘pollen tube superhighway’ (Johnson and Preuss, 2002; Figure 2b).
Pollen tubes of cobl10 are hyposensitive to ovule guidance cues
In addition to reduced tube growth, pollen tubes of cobl10 were less likely to turn toward the micropyle (Figure 3a–d). Indeed, only a few ovules were targeted by cobl10 pollen tubes, even though the pollen tubes did reach the bottom of pistils at 48 HAP (Figure 3a–d), and suggesting a compromised ability to respond to ovular guidance signals.
To provide further evidence, we used a semi-in vivo pollen tube guidance assay (Palanivelu and Preuss, 2006), which assesses the ability of pollen tubes to sense ovular signals after emerging from the stigma and style, and thus is not influenced by female components in the transmitting tract. Our previous results showed that COBL10 was required for pollen tube in vivo growth primarily in the transmitting tract. Therefore, the semi-in vivo system would exclude the possibility that reduced growth of cobl10 pollen tubes indirectly affected its tube turning and would reflect the ability of cobl10 pollen tubes to sense guidance cues from the ovule. Both wild type (WTGUS) (Figure 3e) and cobl10GUS (Figure 3f) pollen tubes grew out from cut stylar ends on germination medium. However, pollen tubes of WTGUS grew toward all ovules located in front of the styles in a circle (Figure 3e,g), while pollen tubes that targeted cobl10GUS were compromised (Figure 3f,h). Over 60% of the ovules were targeted by a wild-type pollen tube, whereas <20% were targeted by a cobl10 tube (Figure 3i), a finding that indicated severely compromised signal sensing of cobl10 pollen tubes.
cobl10 pollen tubes growing in vivo had abnormal cell wall organization
Because mutations at other COBs were reported to affect dynamic cell wall organization (Roudier et al., 2002; Li et al., 2003; Brown et al., 2005; Sindhu et al., 2007; Hochholdinger et al., 2008), we wondered if cobl10 mutants would be similar. We first examined the cell wall in pollen tubes growing in vitro using histochemical stains. Aniline blue stains callose, calcofluor white stains both cellulose and callose, and ruthenium red stains pectin (Lalanne et al., 2004; Wang et al., 2011). In growing pollen tubes, calcofluor white stained uniformly along the pollen tube wall, with a slightly higher intensity along the shank than at the apical region, both in wild type (Figure S4a) and cobl10 (Figure S4(b)). Both wild-type and cobl10 tubes showed an apical pectin cap as well as a weak pectin signal along the tube shank (Figure S4(c,d)). Aniline blue stained the apical region, tube shank and callose plugs comparably in wild type and cobl10 (Figure S4e,f). That no staining differences for all three carbohydrate groups was detectable between wild type and cobl10 pollen tubes growing in vitro (Figure S3) correlated with the normal growth of cobl10 pollen tubes in vitro.
Because cobl10 pollen tubes grew more slowly than those of wild type only in the transmitting tract (Figure 2), its potential effects on wall organization would be seen only when growing in vivo. We first analyzed the distribution of pectins in pollen tubes growing out of the styles in the semi-in vivo experiments in which we did observe a slight reduction in tube growth in cobl10. As shown in Figure 5, we could clearly see the difference by JIM7 labeling between wild type and mutants. Instead of concentrating at the apical region in wild type (Figure 5a), in cobl10 JIM7 labeling was evenly spread to the whole tube wall (Figure 5b). JIM5 that labels de-esterified pectins (Chebli et al., 2012) did not show any marked difference between wild type and cobl10, such that JIM5 labeling was strongest along the shank wall but weakest at the apical wall in both genetic backgrounds. Considering that pollen tubes from both genotypes had been given the same treatments and captured using the same gain value in confocal imaging, the only difference between wild type and cobl10 by JIM5 labeling is that the signal intensity was stronger in cobl10 tubes (Figure 5d) than in the wild-type tubes (Figure 5c).
We therefore used transmission electron micrograph (TEM) to determine whether cobl10 pollen tubes had a defective wall structure in vivo, using wild-type pistils pollinated with either wild-type or cobl10 pollen for 4 h. Wild-type tubes growing in pistils had a very distinct spatial wall structure such that a thick layer of a porous cap was at the apical dome and dense fibrous cellulose was at the shank wall (Figure 4a–c). In contrast, in cobl10 pollen no apparent apical cap was detected (Figure 4d,e), and the shank walls contained no detectable fibrous cellulose (Figure 4f), such that spatial information regarding cell wall organization was lost. Considering the pollen tube growth defects of cobl10, that cobl10 pollen tubes had defective cell wall organization in vivo but not in vitro suggested that COBL10 functions only in the presence of female nutrients and regulatory cues.
Asymmetric localization of cobl10 at the plasma membrane
GPI-anchors often serve as a signal for polar protein sorting in metazoan cells such as neutrophils and epithelial cells (Chatterjee and Mayor, 2001), to which pollen tubes show a certain mechanistic resemblance. Therefore, we wondered about the subcellular localization of COBL10. Because GPI-APs contain both N- and C-terminal signal peptides and because protein fusions at either end would interfere with their processing and targeting, we generated ProCOBL10:SP-citrine-COBL10 in which the coding sequence of a pH-insensitive yellow fluorescent protein (YFP) variant (citrine) (Tian et al., 2004) was inserted in-frame behind the N-terminal SP coding sequence of the COBL10 genomic sequence. Introduction of this construct into cobl10 rescued the fertility defect fully, as shown by COBL10-citrine expression (Figure 6a) and fully elongated siliques (Figure 6e), suggesting that the citrine fusion did not affect the function of COBL10.
Using in vitro pollen germination, we detected the citrine signal at a restricted region of the apical plasma membrane as well as in motile punctate vesicles (Figure 6a). Although fairly weak, the apical plasma membrane localization of COBL10 was dynamically maintained during tube growth (Movie S1). To determine the identity of the punctate vesicles containing COBL10, we used pharmacological treatment on transgenic pollen tubes. Brefeldin A (BFA) inhibits trafficking of vesicles secreted from the trans-Golgi network (TGN)/early endosome (EE), while wortmannin (WM) inhibits the production of PI3P and therefore aggregates certain populations of endosomes in pollen tubes (Zhang et al., 2010). BFA did not cause aggregation of the punctate vesicles of COBL10 but abolished its apical PM localization (Figure 6b), suggesting that COBL10 at the apical plasma membrane was delivered by an ArfGEF-dependent post-Golgi trafficking route (Jürgens, 2004). The COBL10 signal was detected at the ‘BFA compartment’ together with the lipophilic dye FM4-64 (Figure 6b). WM treatment caused aggregation of most vesicles as well as loss of the apical PM signal (Figure 6c), suggesting that the COBL10-resident vesicles had an endosome identity.
Localization of COBL10 at the apical plasma membrane depended on its C-terminal hydrophobic residues and an intact PIG complex
Studies of animal GPI-APs indicated that the C-terminal hydrophobic residues were critical for their correct processing or stability (Berger et al., 1988; McDowell et al., 1998). That cobl10 potentially expressed a truncated COBL10 with the last nine amino acids missing, but still showed a mutant phenotype that was fully rescued by COBL10, suggested that the C-terminal hydrophobic residues were important. To determine whether the C-terminal hydrophobic residues were critical for the subcellular localization of COBL10, we generated ProCOBL10:SP-citrine-COBL10ΔC9, which expressed a COBL10 version whose last nine amino acids were replaced with an Asn, mimicking the protein potentially expressed in cobl10 mutants (Figure 1). Indeed, transgenic Arabidopsis pollen tubes that expressed SP-citrine-COBL10ΔC9 lost the apical plasma membrane signal (Figure 6d). Although punctate vesicles still were visible, a substantial signal was detected at filamentous structures (Figure 6d) similar to that of the ER (Cheung and Wu, 2007). To confirm the ER localization, we introduced calnexin–RFP, the ER marker (Gao et al., 2012), into pollen tubes expressing COBL10ΔC9. Indeed, most of COBL10ΔC9 was colocalized with calnexin, indicating its ER localization (Figure S5). ProCOBL10:SP-Citrine-COBL10ΔC9 could not rescue the sterility of cobl10 (Figure 6e,f).
Although COBL10 was predicted to be a GPI-AP (Roudier et al., 2002; Brady et al., 2007), it was not experimentally tested. We reasoned that if COBL10 were indeed a GPI-AP, it would show different subcellular localization when the Arabidopsis PIG complex was not functional. To find out whether this was the case, we characterized mutant alleles of SETH1 and SETH2 (Lalanne et al., 2004). The heterozygous seth2 mutant showed pollen growth defects such that germination percentage was reduced and mutant pollen tubes ceased growth early as had been reported previously (Lalanne et al., 2004). While a new mutant allele of seth1, seth1-4 (Methods S1) showed similar pollen growth defects (Figure S5) as reported for the other SETH1 alleles (Lalanne et al., 2004). ProCOBL10:SP-citrine-COBL10 was then introduced into seth1-4 or seth2 heterozygous mutants and the citrine signal was analyzed in transgenic pollen tubes. In most pollen tubes that showed the seth1-4 or seth2 phenotype (Lalanne et al., 2004), the citrine signal was detected at the filamentous ER structure as well as at punctate vesicles, but not at the apical plasma membrane (Figure S5), similar to the results shown by COBL10ΔC9 (Figure 6d). This result indicated that processing of COBL10 by the PIG complex was critical for its subcellular localization.
Pollen-specific COBL11 confers partial redundancy to COBL10
That cobl10 plants were still able to set a few seeds in each silique suggested functional redundancy. Indeed, COBL11, another member of the COBL family in Arabidopsis, is similar to COBL10 and also expressed specifically in pollen (Brady et al., 2007). We obtained two SAIL lines from ABRC (SAIL_326_A05 and SAIL_825_B08) that allegedly contained insertions in the COBL11 genomic region. However, no T-DNA insertions could be identified by genotype analysis. To find out whether COBL11 was functionally homologous to COBL10, we took a different approach by introducing a citrine-fused COBL11 into cobl10 mutants. The subcellular localization of COBL11 resembled that of COBL10 at the apical plasma membrane and punctate vesicles (Figure S6b). ProCOBL11:SP-Citrine-COBL11 in cobl10 mutants partially rescued the reduced seed set (Figure S6a,c,d), a finding that suggested that COBL11 plays a redundant, albeit less important function, than COBL10.
- Top of page
- Experimental procedures
- Supporting Information
Considering the increasingly rich information on both sides of the sexual game, the question as to how pollen tubes perceive female cues and translate them into coordinated cellular activities requires more attention. Genes whose mutations show defective male gametophytic function only in the presence of female signals are keys that link extracellular guidance cues to intracellular networks that define and reorient the growth of pollen tubes. Many mutants have defects in polarized growth of pollen tubes in vitro, without input from the female (Fu et al., 2001; Frietsch et al., 2007; Sousa et al., 2008; Szumlanski and Nielsen, 2009; Ye et al., 2009; Zhang et al., 2009; Wang et al., 2011), suggesting that they affect components of the pollen tube internal machinery. Although such components might also participate in cellular changes during female-guided growth, it is difficult to separate their functions in maintaining polar growth from responding to female cues. The expression of COBL10 is high in mature pollen grains and in tubes growing in vitro and in vivo (Honys and Twell, 2004; Brady et al., 2007; Qin et al., 2009). The semi-in vivo assays (Figure 3) showed that pollen tubes of cobl10 are severely compromised in female-guided growth. This finding is supported by the near sterility of cobl10. The increased seed size in cobl10 (Figure S1) might be because more resources were allocated to the few fertilized ovules in each silique.
The only other mutants reported to date that specifically showed guidance defects were mutated at genes that encode two potassium transporters (Lu et al., 2011) and POD1 (Li et al., 2011). However, the ER localization of the K+ transporters and POD1 (Li et al., 2011; Lu et al., 2011) suggests indirect signal sensing. In comparison, COBL10 functions at the apical plasma membrane of pollen tubes. In addition to previous predictions (Roudier et al., 2005; Brady et al., 2007) and results of proteomic studies (Borner et al., 2003; Elortza et al., 2003, 2006) suggesting that COBs are GPI-APs, we show that COBL10 lost its apical plasma membrane localization in seth1 and seth2 mutants (Figure S5), a result that strongly supports the idea that COBL10 is a GPI-AP. GPI-APs play critical roles in mammalian cell–cell communication (Chatterjee and Mayor, 2001; Ikezawa, 2002), especially important in fertilization, as mutations of GPI-APs expressed either in eggs (Alfieri et al., 2003; Primakoff and Myles, 2007) or in sperm (Kondoh et al., 2005) prevented fertilization. Comparable with the intense communication between egg and sperm in mammals, interactions between pollen tubes and female tissues are critical for fertilization in flowering plants. It was recently reported that LORELEI, a GPI-AP expressed in female gametophytes, is critical for communication with incoming pollen tubes during pollen tube reception (Capron et al., 2008; Tsukamoto et al., 2010). We showed that COBL10 at the apical plasma membrane of pollen tubes mediates perception or interpretation of female signals, suggesting that GPI-APs have been adopted as key determinants of fertilization during evolution, even though different phyla use divergent regulatory mechanisms.
The specific role of COBL10 during signal perception and interpretation probably involves dynamic organization of the pollen tube wall. Studies of other COB members have indicated that COBs affect cell wall organization (Li et al., 2003; Roudier et al., 2005; Ching et al., 2006; Sindhu et al., 2007; Hochholdinger et al., 2008). We show here that mutations at COBL10 disrupted the formation of the apical pectin cap as well as reduced the deposition of cellulose microfibrils along the tube shank (Figure 4). Such effects on carbohydrate deposition occur in vivo (Figure 4) but not in vitro (Figure S3), suggesting that COBL10 function relies on the presence of female signals. The reduced tube growth of cobl10 seen in vivo and the localization of COBL10 at punctate vesicles as well as the plasma membrane are consistent with the idea that COBL10 regulates cell wall synthesis or delivery of wall materials to the outside. The defects in tubes length in vivo also suggest that COBL10 might have a role in sensing and relaying female cues needed to stimulate tube growth in the transmitting tract. However, we could not exclude the possibility that the insensitivity of cobl10 pollen tubes to ovular guidance is an indirect outcome of compromised tip growth due to its inability to control wall modifications.
How COBL10 regulates the cell wall organization of pollen tubes upon sensing female signals remains to be understood. One hypothesis involves pectin modifications. Methyl-esterified pectins are deposited at the apical cap of pollen tubes by inhibiting PME activity, maintaining the elasticity of the apical cap and thereby facilitating polar growth (Geitmann, 2010). COBs contain putative carbohydrate binding domains (CBD). It was reported that CBDs released by bacteria regulate plant PMEs to facilitate their entry into host cells (Hewezi et al., 2008). Analogous to this mechanism during a plant–pathogen interaction, the CBD of COBL10 may play similar roles during pollen-pistil interaction. Indeed, an in silico co-expression analysis (Obayashi et al., 2011) showed that COBL10 is found with both PME inhibitors and pectin lyases. A likely scenario to be explored in the future is that COBL10 at the apical plasma membrane recruits or activates PME inhibitors and similar enzymes upon sensing female cues.
In animals, GPI-APs are processed at the ER by the PIG complex and transported via post-Golgi vesicles to the plasma membrane. Proper GPI processing relies on the C-terminal hydrophobic residues whose truncations caused either ER retention or protein degradation (Berger et al., 1988; Fujita and Kinoshita, 2012; Galian et al., 2012). We show here that localization of COBL10 at the apical plasma membrane also relies on the C-terminal hydrophobic motif and proper GPI processing, suggesting an evolutionary conservation of GPI-AP targeting. Results from pharmacological treatments indicate that COBL10 is present at post-Golgi vesicles (Figure 6), probably serving as trafficking vehicles for its final delivery.
Considering the dynamic feature of pollen tube growth, how the restricted localization of COBL10 at the plasma membrane is maintained remains to be understood. In animal cells, GPI-APs are enriched at lipid rafts, which aggregate at the growth front of polarized cells (Gomez-Mouton et al., 2004). Pollen tubes, such as animal chemotactic cells, may also contain lipid rafts at the very apical region. Indeed, an alternate hypothesis for the specific function of COBL10 is that it could regulate the clustering of lipid rafts, which serve as signaling platforms to integrate multiple proteins for sensing or interpreting female cues during pollen–pistil interactions.
- Top of page
- Experimental procedures
- Supporting Information
Plant growth and phenotype analysis
Arabidopsis plants were grown in a 4:1:1 mix of Fafard 4P:perlite:vermiculite under an 18-h light/6-h dark cycle at 21°C. T-DNA insertion lines including CS876299 (cobl10-1), CS879411 (cobl10-2), CS875361 (cobl10-3) and the alleged COBL11 mutant alleles (SAIL_326_A05 and SAIL_825_B08) were obtained from the Arabidopsis Biological Resource Center (ABRC, Ohio State University). The Col-0 quartet1-2 (qrt1) mutant or qrt1;ProLAT52:GUS was used as wild type. Stable Arabidopsis transformation was done with the floral dipping method (Clough and Bent, 1998). Transgenic plants were selected on Murashige and Skoog medium supplemented with 30 mg/mg Basta salt (Sigma, http://www.sigmaaldrich.com). Ovules from 48 HAP pistils were incubated in Hoyer's solution for 45 min at 4°C, then cleared at room temperature for 2 h before visualization.
The segregation ratio of progenies from cobl10 heterozygous plants or from reciprocal crosses was determined by polymerase chain reaction (PCR) analysis using the following primers: G1/G2 for COBL10, G2/G3 for cobl10-1, cobl10-2 and cobl10-3. RNA extraction was performed with Qiagen RNeasy Plant Mini Kit (Qiagen, www.qiagen.com) according to the manufacturer's instruction. OligodT-primed cDNA was synthesized using Superscript™ III Reverse Transcriptase with on-column DNase treatment, as recommended by the manufacturer (Invitrogen, www.invitrogen.com). Primers used to characterize three COBL10 mutants were F1, R1, R2 and R3. Primers used in reverse transcription (RT)-PCR for endogenous COBL10 were R4/R5; for exogenous COBL10 were R6/R7. Primers are listed in Table S1. Primers to amplify ACTIN2 (ACT2) are described in Zhang and McCormick (2007).
All constructs were generated using the Gateway™ technology (Invitrogen). The entry vector for ProCOBL10:SP-Citrine-COBL10 was generated according to a three-step PCR protocol described previously (Tian et al., 2004) with primers P1/P2/P3/P4 and the citrine-specific primer pair P5/P6. The entry vectors for ProCOBL10:SP-citrine-COBL10ΔC9 were generated with primers P1/P7 using the full length COBL10 entry vector as the template. The entry vector for ProCOBL11:SP-citrine-COBL11 was generated with primers P9/P10/P11/P12 and the citrine-specific primer pair P5/P6. The entry vector for calnexin was generated with primers P13/P14 using inflorescence cDNA as the template. Expression vectors were generated by LR reactions using LR Clonase II (Invitrogen, www.invitrogen.com) with a previously described Gateway destination vector (Zhang and McCormick, 2007). Primers are listed in Table S1.
All PCR amplifications used Phusion™ hot start high-fidelity DNA polymerase with the annealing temperature and extension times recommended by the manufacturer (Finnzyme). All entry vectors were sequenced using an ABI 3300 sequencer and sequences were analyzed using Vector NTI (Invitrogen). The Bioneer® PCR purification kit and Bioneer® Spin miniprep kit were used for PCR-product recovery and plasmid-DNA extraction, respectively.
Transmission electron micrograph
For sample fixation, four HAP wild-type pistils pollinated with either wild-type pollen or with cobl10 pollen were treated according to the protocol for SEM. Tissues were then treated twice with propylene oxide for 2 h, propylene oxide:EPO812(1:1) (SPI-Chem™) for 12 h, EPO812 for 12 h, EPO812 for 24 h, and EPO812 + DMP-30 for 12 h. Samples were then fixed at 36°C for 12 h, at 45°C for 12 h, 60°C for 48 h. Fixed samples were sectioned with RM2265 (Leica, www.leica.com) to locate pollen tubes, then sectioned with Super Nova (Leica) for super-thin sections. Slides were stained with 2% uranyl acetate for 15 min, then lead citrate for 2 min and observed with a JEM-1200EX transmission electron microscope.
Immunofluorescent labeling of semi-in vivo pollen tubes
Microscopic imaging was performed using a dissecting scope (Leica), or Axio Observer D1 microscope (Zeiss, www.zeiss.com) with either bright-field, differential interference contrast (DIC) or epifluorescence optics, or with LSM780 laser scanning confocal microscope (Zeiss). Images were captured using a Spot digital camera (Diagnostic Instruments, http://www.diaginc.com/), exported using axiovision software (Zeiss), and processed using Adobe Photoshop 7.0 (Adobe).
- Top of page
- Experimental procedures
- Supporting Information
We thank the ABRC (http://www.arabidopsis.org) for the T-DNA insertion lines, Dr Li-Qun Chen from the Chinese Agricultural University for help with immunofluorescent labeling of semi-in vivo pollen tubes and Dr Yi-Hua Zhou from the Institute of Genetics and Developmental Biology for the antibodies used in immunofluorescent labeling. This research was supported by the Major Research Plan (2013CB945100) from the Ministry of Science and Technology of China and by a grant from NSFC (31271578), and by the United States Department of Agriculture Current Research Information System to SM (grant no. 5335–21000–030–00D). YZ's laboratory and XZ's laboratory are supported by the Tai-Shan Scholar program from Shandong Provincial Government. The authors declare that there is no conflict of interest.
- Top of page
- Experimental procedures
- Supporting Information
- 2003) Infertility in female mice with an oocyte-specific knockout of GPI-anchored proteins. J. Cell Sci. 116, 2149–2155. , , , , and (
- 2008) Spatial control of Rho (Rac-Rop) signaling in tip-growing plant cells. Trends Cell Biol. 18, 119–127. (
- 1988) COOH-terminal requirements for the correct processing of a phosphatidylinositol-glycan anchored membrane protein. J. Biol. Chem. 263, 10016–10021. , , , , , and (
- 2003) Identification of glycosylphosphatidylinositol-anchored proteins in Arabidopsis. A proteomic and genomic analysis. Plant Physiol. 132, 568–577. , , and (
- 2007) Combining expression and comparative evolutionary analysis. The COBRA gene family. Plant Physiol. 143, 172–187. , , , and (
- 2005) Identification of novel genes in Arabidopsis involved in secondary cell wall formation using expression profiling and reverse genetics. Plant Cell, 17, 2281–2295. , , , and (
- 2008) Maternal control of male-gamete delivery in Arabidopsis involves a putative GPI-anchored protein encoded by the LORELEI gene. Plant Cell, 20, 3038–3049. , , et al. (
- 2011) Pollen tube growth and guidance: roles of small, secreted proteins. Ann. Bot. 108, 627–636. and (
- 2001) The GPI-anchor and protein sorting. Cell. Mol. Life Sci. 58, 1969–1987. and (
- 2012) The cell wall of the Arabidopsis thaliana pollen tube - spatial distribution, recycling and network formation of polysaccharides. Plant Physiol. 160, 1940–1955. , , and (
- 2007) The central cell plays a critical role in pollen tube guidance in Arabidopsis. Plant Cell, 19, 3563–3577. , , , , , , , and (
- 2007) Structural and functional compartmentalization in pollen tubes. J. Exp. Bot. 58, 75–82. and (
- 1995) A floral transmitting tissue-specific glycoprotein attracts pollen tubes and stimulates their growth. Cell, 82, 383–393. , and (
- 2010) The pollen tube journey in the pistil and imaging the in vivo process by two-photon microscopy. J. Exp. Bot. 61, 1907–1915. , , , and (
- 2006) Brittle stalk 2 encodes a putative glycosylphosphatidylinositol-anchored protein that affects mechanical strength of maize tissues by altering the composition and structure of secondary cell walls. Planta, 224, 1174–1184. , , , , , and (
- 1998) Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J. 16, 735–743. and (
- 2003) Proteomic analysis of glycosylphosphatidylinositol-anchored membrane proteins. Mol. Cell. Proteomics, 2, 1261–1270. , , , , and (
- 2006) Modification-specific proteomics of plasma membrane proteins: identification and characterization of glycosylphosphatidylinositol-anchored proteins released upon phospholipase D treatment. J. Proteome Res. 5, 935–943. , , , , , , and (
- 2007) A cyclic nucleotide-gated channel is essential for polarized tip growth of pollen. Proc. Natl Acad. Sci. USA, 104, 14531–14536. , , , , , and (
- 2001) Rop GTPase-dependent dynamics of tip-localized F-actin controls tip growth in pollen tubes. J. Cell Biol. 152, 1019–1032. , and (
- 2012) GPI-anchor remodeling: potential functions of GPI-anchors in intracellular trafficking and membrane dynamics. Biochim. Biophys. Acta 1821, 1050–1058. and (
- 2012) Efficient glycosylphosphatidylinositol (GPI) modification of membrane proteins requires a C-terminal anchoring signal of marginal hydrophobicity. J. Biol. Chem. 287, 16399–16409. , , and (
- 2012) The Golgi-localized Arabidopsis endomembrane protein12 contains both endoplasmic reticulum export and Golgi retention signals at its C terminus. Plant Cell, 24, 2086–2104. , , , , , and (
- 2010) How to shape a cylinder: pollen tube as a model system for the generation of complex cellular geometry. Sex. Plant Reprod. 23, 63–71. (
- 2005) Glycosylphosphatidylinositol-anchored proteins are required for cell wall synthesis and morphogenesis in Arabidopsis. Plant Cell, 17, 1128–1140. , , , , , and (
- 2004) Dynamic redistribution of raft domains as an organizing platform for signaling during cell chemotaxis. J. Cell Biol. 164, 759–768. , , , , , , and (
- 2008) Cellulose binding protein from the parasitic nematode Heterodera schachtii interacts with Arabidopsis pectin methylesterase: cooperative cell wall modification during parasitism. Plant Cell, 20, 3080–3093. , , , , , and (
- 2001) Pollen tube attraction by the synergid cell. Science, 293, 1480–1483. , , , , , and (
- 2008) The maize (Zea mays L.) roothairless3 gene encodes a putative GPI-anchored, monocot-specific, COBRA-like protein that significantly affects grain yield. Plant J. 54, 888–898. , , , , , , , and (
- 2004) Transcriptome analysis of haploid male gametophyte development in Arabidopsis. Genome Biol. 5, R85. and (
- 2002) Glycosylphosphatidylinositol (GPI)-anchored proteins. Biol. Pharm. Bull. 25, 409–417. (
- 2002) Plotting a course: multiple signals guide pollen tubes to their targets. Dev. Cell, 2, 273–281. and (
- 2004) A compendium of methods useful for characterizing Arabidopsis pollen mutants and gametophytically-expressed genes. Plant J. 39, 761–775. and (
- 2006) Analysis of the root-hair morphogenesis transcriptome reveals the molecular identity of six genes with roles in root-hair development in Arabidopsis. Plant J. 45, 83–100. , and (
- 2004) Membrane trafficking in plants. Annu. Rev. Cell Dev. Biol. 20, 481–504. (
- 2006) Loss of function of COBRA, a determinant of oriented cell expansion, invokes cellular defence responses in Arabidopsis thaliana. J. Exp. Bot. 57, 2923–2936. , , , and (
- 2008) The regulation of cell motility and chemotaxis by phospholipid signaling. J. Cell Sci. 121, 551–559. , and (
- 2005) Angiotensin-converting enzyme is a GPI-anchored protein releasing factor crucial for fertilization. Nat. Med. 11, 160–166. , , et al. (
- 2004) SETH1 and SETH2, two components of the glycosylphosphatidylinositol anchor biosynthetic pathway, are required for pollen germination and tube growth in Arabidopsis. Plant Cell, 16, 229–240. , , , , , , and (
- 2003) BRITTLE CULM1, which encodes a COBRA-like protein, affects the mechanical properties of rice plants. Plant Cell, 15, 2020–2031. , , et al. (
- 2011) POD1 regulates pollen tube guidance in response to micropylar female signaling and acts in early embryo patterning in Arabidopsis. Plant Cell, 23, 3288–3302. , , , , , , , , and (
- 2011) Pollen tubes lacking a pair of K+ transporters fail to target ovules in Arabidopsis. Plant Cell, 23, 81–93. , , , , , and (
- 2005) Micropylar pollen tube guidance by egg apparatus 1 of maize. Science, 307, 573–576. , , and (
- 1993) Male Gametophyte Development. Plant Cell, 5, 1265–1275. (
- 1998) Glycosylphosphatidylinositol-dependent secretory transport in Trypanosoma brucei. Biochem J. 335, (Pt 3) 681–689. , and (
- 2011) ATTED-II updates: condition-specific gene coexpression to extend coexpression analyses and applications to a broad range of flowering plants. Plant Cell Physiol. 52, 213–219. , , and (
- 2009) Defensin-like polypeptide LUREs are pollen tube attractants secreted from synergid cells. Nature, 458, 357–361. , , et al. (
- 2006) Distinct short-range ovule signals attract or repel Arabidopsis thaliana pollen tubes in vitro. BMC Plant Biol. 6, 7. and (
- 2003) Pollen tube growth and guidance is regulated by POP2, an Arabidopsis gene that controls GABA levels. Cell, 114, 47–59. , , and (
- 2007) Cell-cell membrane fusion during mammalian fertilization. FEBS Lett. 581, 2174–2180. and (
- 2009) Penetration of the stigma and style elicits a novel transcriptome in pollen tubes, pointing to genes critical for growth in a pistil. PLoS Genet. 5, e1000621. , , , , , , , and (
- 2002) The COBRA family of putative GPI-anchored proteins in Arabidopsis. A new fellowship in expansion. Plant Physiol. 130, 538–548. , , and (
- 2005) COBRA, an Arabidopsis extracellular glycosyl-phosphatidyl inositol-anchored protein, specifically controls highly anisotropic expansion through its involvement in cellulose microfibril orientation. Plant Cell, 17, 1749–1763. , , et al. (
- 2001) COBRA encodes a putative GPI-anchored protein, which is polarly localized and necessary for oriented cell expansion in Arabidopsis. Genes Dev. 15, 1115–1127. , , , , , , and (
- 2007) Maize Brittle stalk2 encodes a COBRA-like protein expressed in early organ development but required for tissue flexibility at maturity. Plant Physiol. 145, 1444–1459. , , , , , , and (
- 2008) Arabidopsis phosphatidylinositol-4-monophosphate 5-kinase 4 regulates pollen tube growth and polarity by modulating membrane recycling. Plant Cell, 20, 3050–3064. , and (
- 2009) The Rab GTPase RabA4d regulates pollen tube tip growth in Arabidopsis thaliana. Plant Cell, 21, 526–544. and (
- 2004) LeSTIG1, an extracellular binding partner for the pollen receptor kinases LePRK1 and LePRK2, promotes pollen tube growth in vitro. Plant J. 39, 343–353. , , , and (
- 2004) High-throughput fluorescent tagging of full-length Arabidopsis gene products in planta. Plant Physiol. 135, 25–38. , , et al. (
- 2010) A role for LORELEI, a putative glycosylphosphatidylinositol-anchored protein, in Arabidopsis thaliana double fertilization and early seed development. Plant J. 62, 571–588. , , , and (
- 2011) Arabidopsis CSLD1 and CSLD4 are required for cellulose deposition and normal growth of pollen tubes. J. Exp. Bot. 62, 5161–5177. , , , , , , , , and (
- 2000) A pollen tube growth-promoting arabinogalactan protein from nicotiana alata is similar to the tobacco TTS protein. Plant J. 22, 165–176. , , and (
- 2009) Arabidopsis formin3 directs the formation of actin cables and polarized growth in pollen tubes. Plant Cell, 21, 3868–3884. , , , , , and (
- 2007) A distinct mechanism regulating a pollen-specific guanine nucleotide exchange factor for the small GTPase Rop in Arabidopsis thaliana. Proc. Natl Acad. Sci. USA, 104, 18830–18835. and (
- 2009) Two Arabidopsis AGC kinases are critical for the polarized growth of pollen tubes. Plant J. 58, 474–484. , and (
- 2010) Interdependence of endomembrane trafficking and actin dynamics during polarized growth of Arabidopsis pollen tubes. Plant Physiol. 152, 2200–2210. , , and (
- Top of page
- Experimental procedures
- Supporting Information
|tpj12139-sup-0001-FigS1.tif||image/tif||8436K||Figure S1. cobl10 has larger seeds than wild type due to increased embryonic cell size.|
|tpj12139-sup-0002-FigS2.tif||image/tif||5017K||Figure S2. cobl10 has normal pollen development, hydration and germination.|
|tpj12139-sup-0003-FigS3.tif||image/tif||2580K||Figure S3. cobl10 pollen showed normal germination and growth in vitro.|
|tpj12139-sup-0004-FigS4.tif||image/tif||1705K||Figure S4. Pollen tubes of cobl10 did not differ from those of wild type in carbohydrate deposition in vitro.|
|tpj12139-sup-0005-FigS5.tif||image/tif||1003K||Figure S5. COBL10ΔC-citrine relocalized to the endoplasmic reticulum.|
|tpj12139-sup-0006-FigS6.tif||image/tif||1571K||Figure S6. The apical plasma membrane localization of COBL10 relies on an intact PIG complex.|
|tpj12139-sup-0007-FigS7.tif||image/tif||1517K||Figure S7. COBL10 function is partly redundant with COBL11.|
|tpj12139-sup-0008-TableS1.doc||Word document||49K||Table S1. Primers used in this study.|
|tpj12139-sup-0009-MovieS1.AVI||video/avi||4255K||Movie S1. Pollen tubes expressing SP-Citrine-COBL10 during in vitro growth.|
|tpj12139-sup-0010-MethodsS1.doc||Word document||48K||Methods S1. Supporting experimental procedures.|
Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.