Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

PCP‐B class pollen coat proteins are key regulators of the hydration checkpoint in Arabidopsis thaliana pollen–stigma interactions

PCP‐B class pollen coat proteins are key regulators of the hydration checkpoint in Arabidopsis... Research PCP-B class pollen coat proteins are key regulators of the hydration checkpoint in Arabidopsis thaliana pollen–stigma interactions Ludi Wang, Lisa A. Clarke, Russell J. Eason, Christopher C. Parker, Baoxiu Qi, Rod J. Scott and James Doughty Department of Biology and Biochemistry, University of Bath, Claverton Down, Bath, BA2 7AY, UK Summary Author for correspondence: The establishment of pollen–pistil compatibility is strictly regulated by factors derived from James Doughty both male and female reproductive structures. Highly diverse small cysteine-rich proteins Tel: +44 0 1225 383485 (CRPs) have been found to play multiple roles in plant reproduction, including the earliest Email: bssjd@bath.ac.uk stages of the pollen–stigma interaction. Secreted CRPs found in the pollen coat of members Received: 31 May 2016 of the Brassicaceae, the pollen coat proteins (PCPs), are emerging as important signalling Accepted: 23 July 2016 molecules that regulate the pollen–stigma interaction. Using a combination of protein characterization, expression and phylogenetic analyses we New Phytologist (2017) 213: 764–777 identified a novel class of Arabidopsis thaliana pollen-borne CRPs, the PCP-Bs (for pollen coat doi: 10.1111/nph.14162 protein B-class) that are related to embryo surrounding factor (ESF1) developmental regula- tors. Single and multiple PCP-B mutant lines were utilized in bioassays to assess effects on pol- len hydration, adhesion and pollen tube growth. Key words: Arabidopsis thaliana, compati- Our results revealed that pollen hydration is severely impaired when multiple PCP-Bs are bility, pollen coat proteins, pollen hydration, pollen–stigma interaction, reproduction, sig- lost from the pollen coat. The hydration defect also resulted in reduced pollen adhesion and nalling. delayed pollen tube growth in all mutants studied. These results demonstrate that AtPCP-Bs are key regulators of the hydration ‘checkpoint’ in establishment of pollen–stigma compatibility. In addition, we propose that interspecies diver- sity of PCP-Bs may contribute to reproductive barriers in the Brassicaceae. important in maintaining species barriers (Swanson & Vacquier, Introduction 2002; Takeuchi & Higashiyama, 2012). Pollination in angiosperms involves multiple phases of interac- Members of the Brassicaceae family (which includes Brassica tion between female reproductive tissues of the gynoecium and and Arabidopsis species) are characterized by having stigmas of the male reproductive unit, pollen, and subsequently the pollen the ‘dry’ type, which lack sticky secretions such as those present tube it produces on germination (Hiscock & Allen, 2008). The in species of the Solanaceae (Heslop-Harrison & Shivanna, process is highly selective such that the majority of heterospecific 1977). Dry stigmas are highly discriminatory, reducing the prob- pollen fails to effect syngamy and indeed intraspecific pollination ability that heterospecific pollen grains and pathogenic spores will can also be blocked in those species which possess self- be captured, hydrate and germinate on their surfaces. In this sys- incompatibility (SI) systems. Such prezygotic reproductive barri- tem compatibility is established at the stigma surface within min- ers are evolutionarily advantageous as they limit wasted mating utes of pollination when compatible grains gain access to opportunities, contribute to reproductive isolation and facilitate stigmatic water, but incompatible pollen generally fail to fully outbreeding when SI is present (Yost & Kay, 2009; Smith et al., hydrate and germinate. Thus, pollen hydration on the stigma sur- 2013). The establishment of compatibility is complex and face is essential for successful reproduction and is a strictly regu- involves a suite of biophysical and molecular recognition factors lated checkpoint centred in the stigma (Dickinson, 1995; Ma that operate throughout the pollination process; from pollen cap- et al., 2012; Hiroi et al., 2013). Possession of a dry stigma comes ture by the stigma, pollen hydration, germination and stigmatic with the requirement that the exine surface of conspecific pollen penetration, to polarized tube growth through the pistil (reviewed must carry a coating (tryphine) (Dickinson, 1995; Dickinson in Chapman & Goring, 2010). Although much progress has been et al., 2000). Tryphine is a complex mixture of lipids, proteins, made in elucidating the mechanisms that regulate compatibility glycoconjugates and pigments (Piffanelli et al., 1998; Hernandez- in a broad range of species, it is clear that great mechanistic diver- Pinzon et al., 1999) that confers adhesive properties to the grain, sity exists, with no common system in operation (Hiscock & provides a conduit for water to pass from the stigma to effect pol- Allen, 2008; Allen et al., 2011). Such diversity is considered to be len hydration and, importantly, carries factors that determine 764 New Phytologist (2017) 213: 764–777  2016 The Authors www.newphytologist.com New Phytologist  2016 New Phytologist Trust This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. New Phytologist Research 765 compatibility (Dickinson, 1995; Safavian & Goring, 2013). compatibility factor EXO70A1 (Stone et al., 2003; Samuel et al., Pollen access to stigmatic water requires targeted secretion in the 2009). Other CRPs belonging to the PCP-A class of Brassica pol- stigma immediately adjacent to a compatible pollen grain (Dick- len coat proteins such as SLR-BP1 and PCP-A1, bind the stig- inson, 1995) and it is now well established that in the Brassi- matic proteins S-locus related 1 (SLR1) and S-locus glycoprotein caceae this involves exocyst-mediated tethering of secretory (SLG), respectively, and thus are likely to function in the pollen– vesicles to the stigmatic plasma membrane (Samuel et al., 2009; stigma interaction, even though their precise function remains to Safavian & Goring, 2013; Safavian et al., 2015). be determined (Doughty et al., 1998; Takayama et al., 2000). A Despite progress in identifying molecular regulators of com- further class of Brassica pollen coat CRPs, the PCP-Bs, have been patibility in stigmas, relatively little is known about the pollen- described and also are good candidates for regulators of the earli- borne signals that establish it. Components of the pollen coat are est phases of the pollen–stigma interaction (Doughty et al., most likely to mediate compatibility due to the intimate interac- 2000). Thus, there is a growing body of evidence demonstrating tion of this layer with the surface of stigmatic papilla cells and the that the pollen coat carries factors that mediate both incompati- speed of pollen acceptance (Elleman & Dickinson, 1986, 1990; bility and compatibility, and that cysteine-rich pollen coat pro- Preuss et al., 1993). Indeed, application of isolated pollen coat to teins are important to molecular dialogue in the pollen–stigma the stigma surface evokes a rapid expansion of the stigmatic outer interaction. Although many studies have focused on the mecha- wall layer (Elleman & Dickinson, 1990, 1996) and isolated pol- nisms of self-incompatibility (SI) in a range of species, the molec- len stimulates the production of structures resembling vesicles in ular regulation of self-compatibility (SC) in flowering plants is the stigma apoplast beneath the pollen contact site (Elleman & still poorly understood. Dickinson, 1996). In this study, we report on the identification of four Analysis of pollen coat components and mutational studies in A. thaliana PCP-B encoding genes termed AtPCP-Ba Arabidopsis thaliana have shed light on factors that influence the (At5g61605), AtPCP-Bb (At2g29790), AtPCP-Bc (At2g16535) pollen–stigma interaction. A number of these appear to be bio- and AtPCP-Bd (At2g16505), which are expressed gametophyti- physical in nature, for example eceriferum (cer) mutant pollen cally late in pollen development. By utilizing T-DNA insertion fails to hydrate due to the elimination of very long chain lipids lines carrying single, double and triple AtPCP-B gene knockouts, from the pollen coat (Preuss et al., 1993; Hulskamp et al., 1995; we examined the impact of these mutations on pollen morphol- Fiebig et al., 2000). Hydration defects have also been reported in ogy and the pollen–stigma interaction. Phenotypic analyses mutants for extracellular lipase 4 (EXL4) and GRP17, an revealed defects in pollen hydration and delays in pollen tube oleosin-domain-containing glycine-rich protein which may work growth for single and combined mutants compared with wild- cooperatively to alter the lipid composition at the pollen–stigma type. Triple mutant pcp-ba/b/c pollen displayed a substantially interface to facilitate the passage of water to the grain (Mayfield reduced hydration rate on stigmas, delayed pollen tube growth, & Preuss, 2000; Updegraff et al., 2009). Work in Brassica has led as well as weakened anchoring to the stigma surface. Importantly, to the identification of several groups of small cysteine-rich pollen no impact on pollen morphology was revealed in this study coat proteins (Doughty et al., 1993, 1998, 2000; Hiscock et al., though the mutants presented striking effects on early post- 1995; Schopfer et al., 1999; Takayama et al., 2000; Shiba et al., pollination events. Such evidence suggests the AtPCP-Bs act as 2001). Importantly these polypeptides, rather than having major important regulatory factors during the earliest stages of the pol- effects on biophysical properties of the pollen coat, act as ligands len-stigma interaction by establishing a molecular dialogue and have been demonstrated to bind a number of stigmatic between the stigma and pollen grains. proteins. In recent years, a broad range of cysteine-rich proteins (CRPs) Materials and Methods have been identified in plants having functions in cell signalling, development and defence (Silverstein et al., 2007; Li et al., 2014). Plant material and growth conditions They are all characterized by being small (< 160 amino acids), having a conserved N-terminal signal peptide and a C-terminal Brassica oleracea var. alboglabra L. homozygous for incompatibil- cysteine-rich region with the pattern of cysteines determining ity haplotypes S25 and S29 (Horticultural Research Interna- their classification (Marshall et al., 2011). A number function in tional, Wellesbourne, UK) was used for isolation of BoPCP-B1 plant reproduction, including pollen–stigma self-recognition and BoPCP-B2 pollen coat proteins, respectively. Arabidopsis (Schopfer et al., 1999; Shiba et al., 2001), pollen tube growth thaliana (L.) Heynh. T-DNA insertion lines SALK_207087, and guidance (Chae et al., 2009; Okuda et al., 2009), and early SALK_062825, SALK_072366 (Alonso et al., 2003) were embryo development (Marshall et al., 2011; Costa et al., 2014). obtained from the Nottingham Arabidopsis Stock Centre Several families of CRPs have been identified in the pollen coat (NASC). GABI_718B04 was purchased from GABI-KAT of Brassica and Arabidopsis, and some have confirmed roles in the (Kleinboelting et al., 2012). T-DNA insertion sites and their pollen–stigma interaction. In self-incompatible species, the S- respective mutant alleles are detailed in Supporting Information locus cysteine-rich protein (SCR/SP11) (Schopfer et al., 1999; Fig. S1. Single gene T-DNA insertion lines were backcrossed to Shiba et al., 2001), acts as the male determinant and interacts wild-type (WT; Col-0) at least three times before phenotyping. with the S-receptor kinase (SRK) (Takasaki et al., 2000), to trig- pcp-bb/c and pcp-ba/b/c mutant lines were created using stan- ger pollen rejection by targeted degradation of the basal dard crossing procedures. PCP-B transcript status for each 2016 The Authors New Phytologist (2017) 213: 764–777 New Phytologist  2016 New Phytologist Trust www.newphytologist.com New 766 Research Phytologist mutant was confirmed by reverse transcription polymerase chain measured in pixels using IMAGEJ software (Schneider et al., 2012). reaction (RT-PCR) (Fig. S2). Primers for RT-PCR are detailed Pollen hydration (%) was calculated using the equation: pollen in Table S1. The A9-barnase male sterile line was provided by hydration (%) = (pollen diameter  initial pollen diameter)/initial Rod Scott, University of Bath, UK (Paul et al., 1992). GUS pollen diameter. Slopes were determined using 11 data points (b-glucuronidase) reporter lines pAt5g61605:GUS and during the 0–10 min, 10–20 min, or 20–30 min time periods pAt2g16505:GUS were provided by Jose F. Gutierrez-Marcos using the linear regression curve f = a x + b. All statistical analyses (University of Warwick, UK). were carried out using Microsoft Excel 2013. Arabidopsis thaliana plants were propagated in Levington F2 + S compost (Soils HS Limited, Wotton-Under-Edge, UK) in Pollen adhesion assay a controlled environment room with a 16 h : 8 h, light : dark pho- 2 1 toperiod provided by fluorescent lighting (130 lmol m s ). Stigmas of A. thaliana A9-barnase plants were hand-pollinated Temperature was maintained at 21  1°C with 60% relative using freshly dehiscent anthers from WT and pcp-ba/b/c lines. humidity. Brassicas were grown in a glasshouse at 21°C with a Pollen was applied as a monolayer. After 30 min the flower was 16 h : 8 h, light : dark photoperiod. excised from the plant and placed into 0.5 ml of fixative (60% v/v ethanol, 30% v/v chloroform, 10% v/v acetic acid) in a 1.5-ml microfuge tube. The sample was immediately shaken 10 times RT-PCR and RNA gel blot analysis using short sharp strokes to dislodge pollen that was not strongly Anthers were collected from A. thaliana stage 10–12 flower buds adherent to the stigma. The flower was then removed and placed (Smyth et al., 1990), stigmas from open flowers of the A. thaliana into a separate microfuge tube. Both samples were retained for A9-barnase line, roots from 2-wk-old seedlings grown on 0.59 pollen counting. Fifty microlitres of aceto-orcein stain (1%) was MS plates and leaves from fully-grown rosettes. RNA was added to the tubes and incubated overnight at room temperature extracted using a PureLink RNA Mini Kit (Thermo Fisher Sci- (RT). ‘Washed-off’ pollen samples were centrifuged at 13 000 g entific, Loughborough, UK). cDNA synthesis was carried out (10 min), excess fixative was removed and the pellet was resus- using the ProtoScript II First Strand cDNA Synthesis Kit (New pended in 10 ll of 50% glycerol before counting. Stigmas were England Biolabs, Hitchin, UK). DNA amplification utilized excised from stained flowers before mounting in 50% glycerol DreamTaq Green PCR Master Mix (29) (Thermo Fisher Sci- and were squashed on a slide to ensure all pollen was visible for entific). Primers for DNA amplification are detailed in Table S1. counting. RNA gel-blot analysis was carried out as described previously (Doughty et al., 1998) using polyadenylated mRNA (450 ng) Pollen tube growth assay from leaves, stigmas, pollen and anthers derived from a range of bud sizes (whole flower buds for anthers of size < 2 mm). Pollinations were initiated on A. thaliana male sterile A9-barnase Labelling of gene-specific BoPCP-B probes (covering the coding stigmas and allowed to proceed for 2 or 4 h before stigmas were region of the gene from aa residue 40 to the C-terminus) was excised and incubated in fixative (60% v/v ethanol, 30% v/v chlo- conducted using the Prime-a-Gene Labeling System (Promega), roform, 10% v/v acetic acid) overnight. After removal of fixative, with modifications to the deoxynucleotidetriphosphates mix to stigmas were incubated in 8 M NaOH for 20 min then washed in permit double labelling with dATP and dCTP (a- P, 100 lCi, dH O three times, each for 5 min. Samples were transferred to 6000 Ci mmol each). 0.1% decolourized aniline blue (0.1% w/v aniline blue in 0.1 M K PO , pH 11) for 1 h before imaging (Kho & Baer, 1968). 3 4 Pollen hydration assays Microscopy For in vivo hydration assays, pollen grains derived from WT, pcp-ba (SALK_207087), pcp-bb (SALK_062825), pcp-bc Imaging of pollen hydration on stigmas, hydration in a humid (SALK_072366), pcp-bd (GABI_718B04), pcp-bb/c and pcp-ba/ chamber and GUS histochemical staining (Methods S1) of flow- b/c lines were applied to stigmas of the A9-barnase male sterile ers and leaves was carried out using a Nikon (Surrey, UK) line. Freshly opened mature flowers were used (retained on the SMZ1500 dissection microscope coupled to Nikon Digital Sight plant) and pollen was applied in a monolayer with an eyelash. At DS-U1 camera. A Nikon Eclipse 90i epifluorescence microscope least eight independent stigmas were used in assays from each (910 objective) with Nikon Digital Sight DS-U1 camera was line. For in vitro hydration assays, pollen from WT and pcp-ba/b/ used for imaging pollen tubes stained with aniline blue and for c triple knockout lines were placed on a slide in a humid chamber anthers stained for GUS activity. (100% relative humidity). Pollen grains on stigmas were pho- tographed under a dissecting microscope immediately after polli- Scanning and transmission electron microscopy nations were initiated (designated as time point zero). Subsequent images were captured every minute for 30 min. For humid cham- Pollinated stigmas were dry-fixed using the method of Elleman ber assays pollen was photographed 1 min after pollen grains were & Dickinson (1986). Samples were washed with 0.1 M sodium placed, then the chamber was sealed, after which images were cacodylate buffer pH 7.4 and prepared for scanning electron taken every minute for 30 min. Equatorial diameter of pollen was microscopy (SEM) and transmission electron microscopy (TEM) New Phytologist (2017) 213: 764–777  2016 The Authors www.newphytologist.com New Phytologist  2016 New Phytologist Trust New Phytologist Research 767 by a modified method of Villar et al. (1987) using 2.5% glu- described by Doughty et al. (1993, 1998). Both BoPCP-B1 and taraldehyde and low viscosity resin. Samples for SEM were gold- BoPCP-B2 co-purified with the previously characterized PCP-A1 coated and imaged using a Jeol JSM640LV Scanning Electron polypeptide following C18 RP-HPLC and were separated by Microscope (Jeol, Tokyo Japan). Samples for TEM were ultra- cation exchange chromatography (Doughty et al., 1998). BoPCP- thin sectioned (100 nm) on an Ultracut-E ultramicrotome (Leica, B1 and BoPCP-B2 were isolated from S25 and S29 incompatibility London, UK) and imaged by Jeol JEM1200EXII transmission lines of B. oleracea var alboglabra, respectively. BoPCP-B protein electron microscope. samples were prepared for N-terminal sequencing as described previously (Doughty et al., 1998). Multiple sequence alignment and phylogenetic analysis 0 0 5 and 3 rapid amplification of cDNA end (RACE) PCP-B-like sequences were retrieved from available complete polymerase chain reaction cloning of BoPCP-B1 and plant genomes completed to at least scaffold level using TBLASTN BoPCP-B2 database searches (PHYTOZOME, https://phytozome.jgi.doe.gov; Comparative Genomics, COGE, https://genomevolution.org). Polyadenylated RNA was isolated from 100 mg of anthers Nucleotide sequence alignments of PCP-B homologous genes derived from 9 to 11 mm flower buds of B. oleracea var were generated using MUSCLE (codon) (Edgar, 2004). Codons of alboglabra (homozygous for S25 and S29 incompatibility haplo- protein coding sequences were translated into amino acid types) using a QuickPrep Micro mRNA purification kit (Phar- sequences before the alignment was performed. Aligned amino macia Biotech, Piscataway, NJ, USA). Cloning of BoPCP-B1 0 0 acid sequences were then replaced by the original codons. Graph- and BoPCP-B2 cDNA sequences was carried out using a 5 /3 ical output of protein sequence alignment was generated by RACE kit (Boehringer Mannheim) following the manufacturer’s JALVIEW using ‘CLUSTALX’ colour coding. Phylogenetic trees were instructions. One microgram of mRNA was subjected to first- built using the maximum-likelihood statistical method in MEGA strand cDNA synthesis using an oligo(dT) kit primer. First v.6 (Tamura et al., 2013). The initial tree was determined by the round 3 RACE cloning of BoPCP-B1 utilized a degenerate neighbour-joining method (NJ). The phylogeny test was carried primer based on the peptide sequence AGNAAK[P/Q] which is out using the bootstrap method (1000 replications). Phylogenetic common to both BoPCP-B proteins (5 -GC-GGATCC- trees were displayed using iTOL (Letunic & Bork, 2007). GCIGGIAA[C/T]GCIGCIAA[A/G]C-3 , where I represents inosine) in conjunction with a kit anchor primer. This was followed by two rounds of PCR using degenerate nested primers Protein structure prediction and modelling (5 -GC-GGATCC-AA[A/G]CA[A/G]ACICCITG[C/T]CA[C/T] 0 0 Protein sequence alignment of PCP-Bs and 2RU1 was carried G-3 and 5 -GC-GGATCC-AA[A/G]CCIAA[C/T]CA[C/T]AC out with T-COFFEE (Notredame et al., 2000). PCP-B protein ITG) based on the BoPCP-B1 specific peptide sequences models were built using SWISS-MODEL (Arnold et al., 2006; Guex KQTPCHE and KPNHTC, respectively. 5 RACE was con- et al., 2009; Kiefer et al., 2009; Biasini et al., 2014) based on the ducted utilizing sequence-specific primers SP1 and SP2 (derived modified T-COFFEE alignment result. 3D cartoon models and from 3 RACE) for cDNA synthesis and nested amplification of 0 0 electrostatic potential surface models were produced by PYMOL the 5 region of the BoPCP-B1 cDNA (SP1 5 -GCTTGCC 0 0 (v.1.7.4; Schr€odinger, Cambridge, UK). GCACCTACGCG-3 and SP2 5 -CAT GTAGCACATGT 0 0 TTTGAGC-3 ). For BoPCP-B2, first round 3 RACE was carried out as described for BoPCP-B1 followed by one further round of In situ hybridization PCR utilizing a degenerate nested primer (5 -GC-GGATCC- Flower buds were excised from inflorescences and immediately ATGAA[C/T]TG[C/T]GA[C/T]ACICA[A/G] G) based on the fixed in fresh 4% paraformaldehyde for 16 h at 4°C with an ini- BoPCP-B2 specific peptide sequence MNCDTQD. 5 RACE tial 10 min under low vacuum. Tissues were embedded in Para- utilized sequence-specific primers SP1 and SP2 (derived from plast Xtra (Sigma), sectioned (7–10 lm) and prepared for 3 RACE sequence) for cDNA synthesis and nested amplification 0 0 probing as described by Langdale (1994), except for the protease of the 5 region of the BoPCP-B1 cDNA (SP1 5 -GG 0 0 treatment where sections were incubated for 30 min at 37°Cin CTTCCCAGATTTAGTGAC-3 and SP2 5 -GTGACACA 50 lgml proteinase K (Sigma). Both antisense and sense ACAAGAACAACTGCG-3 ). probes were synthesized using a SP6/T7 digoxigenin RNA label- ing kit (Boehringer Mannheim, Lewes, UK), according to the Results manufacturer’s instructions. Probes covered the protein coding sequence of the AtPCP-Bb and AtPCP-Bc cDNAs. The pollen coat of Brassica contains polymorphic PCP-B class cysteine-rich proteins Protein purification and N-terminal sequencing In a previous study that characterized the SLG-binding pollen BoPCP-B1 and BoPCP-B2 proteins were purified from total pol- coat protein PCP-A1 from Brassica oleracea (Doughty et al., len coat proteins by a combination of gel filtration, RP-HPLC 1998) two polypeptides were found to copurify with PCP-A1. and cation exchange chromatography following the protocol These were purified to homogeneity and subjected to N-terminal 2016 The Authors New Phytologist (2017) 213: 764–777 New Phytologist  2016 New Phytologist Trust www.newphytologist.com New 768 Research Phytologist sequencing. Each shared an identical six amino acid N-terminal two distinct clades (Fig. 2). One clade includes ESF1-encoding domain and several conserved cysteine residues arranged in a genes clustering into a group of five sequences. Of the other clade unique pattern with respect to other known pollen coat protein four of the sequences fall into a cluster which, following wider families (Fig. S3). These were subsequently named BoPCP-B1 phylogenetic analysis across the Brassica and Arabidopsis genera, and BoPCP-B2 (for B. oleracea pollen coat protein, class B, 1 and placed them in a clade that included genes encoding pollen coat- 2, respectively). The partial BoPCP-B polypeptide sequences per- derived BoPCP-Bs (Fig. 3). Expression analyses confirmed these mitted cloning of their respective full-length cDNAs by RACE four genes as being largely anther-specific (Fig. 4). Taken PCR (GenBank accession numbers: PCP-B1, KX099662; PCP- together, these data suggest that the Arabidopsis sequences are B2, KX099663). BoPCP-B1 and BoPCP-B2 are predicted to likely orthologues of the B. oleracea PCP-Bs and, hence, we encode proteins of 79 and 84 amino acids, respectively, with both named them AtPCP-Ba (At5g61605), AtPCP-Bb (At2g29790), having a putative 25 amino acid secretory signal peptide AtPCP-Bc (At2g16535) and AtPCP-Bd (At2g16505). (Petersen et al., 2011) and a conserved pattern of eight cysteine Iterative BLAST searches Arabidopsis and Brassica genera iden- residues in the mature protein. Based on the N-terminal sequence tified 46 PCP-B-like sequences in total across four species. Phylo- data, mature BoPCP-B1 and BoPCP-B2 are estimated to have genetic analysis of these protein sequences revealed not only the M s of 5490 and 6109, respectively. The localization of the high degree of polymorphism across the family, but also the pres- BoPCP-Bs to the pollen coat, together with their broad similarity ence of two distinct groupings of sequences, the PCP-B and to other small cysteine-rich proteins such as PCP-A1 (Doughty ESF1-containing clades (Fig. 3). These groupings may reflect et al., 1998) and the pollen self-incompatibility determinant SCR functional specialization into seed and pollination-specific roles. (Shiba et al., 2001), suggested that they could potentially Wider phylogenetic BLAST analyses across all known plant lin- function in the pollen–stigma interaction. eages identified 282 predicted PCP-B-like protein sequences in seven angiosperm families (36 species in total, Fig. S4; Table S2). In addition to the Brassicaceae, PCP-B-like proteins were found PCP-Bs are evolutionarily widespread and have homology in the Poaceae, Nelumbonaceae, Solanaceae, Malvaceae, Phry- to Arabidopsis Embryo Surrounding Factor 1 maceae and Pedaliaceae. Thus, the PCP-Bs are members of a developmental regulators wider family of highly polymorphic, though structurally related In order to facilitate subsequent functional analyses of PCP-B proteins, having an ancient evolutionary origin that predates the class pollen coat proteins putative homologues were identified in split between monocot and eudicot lineages. the model plant Arabidopsis. BLAST searching of the Arabidop- sis genome using BoPCP-B sequences revealed the presence of Arabidopsis and Brassica oleracea PCP-B genes are twelve PCP-B-like genes. All sequences were predicted to encode expressed in maturing pollen small, typically basic secreted proteins that shared a common cys- teine pattern of seven or eight cysteines in the mature polypeptide RNA gel-blot analysis for the Brassica oleracea PCP-B1 and PCP- (Fig. 1). Importantly, three members of this Arabidopsis gene B2 genes indicated high levels of expression in anthers (Fig. 4a). family (At1g10747, At1g10745 and At1g10717) encode the cen- Transcripts were first detected at low levels in anthers derived tral cell-derived Embryo Surrounding Factor 1 (ESF1) signalling from 5 to 7 mm flower buds reaching a maximum by the proteins known to shape early embryo development and pattern- 9–11 mm bud stage by which time pollen is tricellular and tapetal ing (Costa et al., 2014). A phylogenetic analysis of mature pro- cells that line the anther locule are fully degraded (Doughty et al., tein-encoding gene regions, based on prior alignment of amino 1998). This late pattern of expression in anther development acid sequences, revealed that the AtPCP-B-like sequences fall into infers that BoPCP-Bs are likely to be gametophytically derived Fig. 1 Protein sequence alignment of Brassica oleracea PCP-B1 and PCP-B2 with all known Arabidopsis thaliana (Col-0) PCP-B-like proteins. AtPCP-Bs are: AtPCP-Ba (At5g61605), AtPCP-Bb (At2g29790), AtPCP-Bc (At2g16535) and AtPCP-Bd (At2g16505). At1g10747, At1g10745 and At1g10717 are ESF1.1, ESF1.2 and ESF1.3, respectively. Sequence conservation, quality and consensus is displayed below. Colour coding follows the default output for Clustal X (http://www.jalview.org/help/html/colourSchemes/clustal.html). New Phytologist (2017) 213: 764–777  2016 The Authors www.newphytologist.com New Phytologist  2016 New Phytologist Trust New Phytologist Research 769 c) were assessed utilizing the pollen hydration assay. A quadruple mutant could not be generated due to the close genetic linkage of PCP-Bc and PCP-Bd (c. 9 kb apart). Pollen equatorial diameter was recorded for 30 min following placement of pollen on stig- mas. Four time points (0, 10, 20 and 30 min) were selected for analysis of the difference of pollen hydration between WT and mutant lines. In addition, the rate of pollen hydration was assessed for each of the three 10-min periods following pollina- tion. On initiation of pollination (0 min) no significant differ- ence was found between the diameters of pollen derived from WT and mutant lines (Fig. 5a). However, at subsequent time Fig. 2 Phylogenetic analysis of 12 PCP-B class genes in Arabidopsis points pcp-bc and pcp-ba/b/c pollen grains were significantly less thaliana (Col-0). The maximum-likelihood tree was constructed by using hydrated than WT pollen (Fig. 5b–d). Assessment of pollen the nucleotide sequences of predicted mature protein coding regions. Branch lengths are proportional to the bar, defined as 0.1 nucleotide hydration as percentage change in diameter (from time point substitutions per codon. The percentage bootstrap values (1000 zero) demonstrated that the degree of pollen hydration was sig- resamplings) > 50% are shown by interior branches. nificantly lower in the pcp-bb, pcp-bc, pcp-bb/c and pcp-ba/b/c mutant lines compared with WT at each time point (Fig. 5e–g). rather than being products of the tapetum. No expression was Despite there being no statistically significant difference in pollen hydration for pcp-ba and pcp-bd compared with WT, median detected in leaves and stigmas, and only very low transcript levels were detected in mature pollen. This expression pattern exactly pollen diameters, hydration percentage and overall ranges in the mirrors that of the pollen coat protein gene PCP-A1 (Doughty data suggested that pcp-ba and pcp-bd mutations also negatively et al., 1998). In order to determine which of the twelve Arabidop- impact on pollen hydration (Fig. 5e–g). sis PCP-B-like genes were likely orthologues of the B. oleracea We extended the analysis to determine the rate of pollen PCP-B sequences, expression analysis was carried out by RT- hydration on stigmas for mutant and WT pollen. Slopes were PCR (Fig. 4b). Six of the genes were found to be expressed in produced by linear regression based on pollen grain diameter stage 12 anthers though two of these, At1g10747 and during each 10-min period following pollination. During the At1g10745, have previously been characterized as central cell- first 10-min period WT pollen hydrates rapidly with this rate derived ESF1 signalling proteins involved in embryo patterning decreasing dramatically during the second and third 10-min (Costa et al., 2014). The remaining four anther-expressed genes periods. A similar overall pattern was observed for all pcp-b (At5g61605, At2g29790, At2g16535 and At2g16505, AtPCP-Ba mutant lines (Fig. 6a). During the first 10-min period of polli- to d, respectively) were found to share a similar temporal expres- nation, the hydration rate of WT pollen was significantly higher sion pattern to the B. oleracea PCP-Bs (Fig. 4a,c). In addition, than that for pcp-bb and pcp-bc. The pcp-bb/c double mutant RNA–RNA in situ hybridization for AtPCP-Bb (Figs 4d, S5) and also demonstrated a significantly affected hydration rate though AtPCP-Bc and promoter-GUS fusions for AtPCP-Ba and this was not greater than either single mutant. However, intro- AtPCP-Bd (Fig. S6) confirmed high-level expression in pollen, duction of pcp-ba into the pcp-bb/c line creating the pcp-ba/b/c further validating the status of this group as pollen coat protein- triple mutant had a dramatic effect on pollen hydration. pcp- encoding genes. Taken together with the phylogenetic analysis ba/b/c pollen hydrated at a substantially lower rate than either that placed these four genes in the same clade as BoPCP-Bs, it is WT or pollen from the pcp-bb/c double-mutant line (Fig. 6) likely that AtPCP-Ba, AtPCP-Bb, AtPCP-Bc and AtPCP-Bd are despite the fact that the pcp-ba mutant had little discernible orthologous to the BoPCPs. effect on pollen hydration in isolation. Hydration rates for the second and third 10-min periods were not significantly different to WT for any of the mutant lines although the final extent of Pollen hydration is impaired in pcp-b mutants hydration was clearly lower for all lines with the exception of In order to investigate the effects of AtPCP-B gene mutations on pcp-ba. early stages of the pollen–stigma interaction in Arabidopsis In order to determine if the pollen hydration defect resulted thaliana, in vivo pollen hydration assays were carried out by polli- from an inherent inability of mutant pollen grains to absorb nating stigmas of the male sterile A9-barnase WT line with water rather than a defect in the pollen–stigma interaction, an pollen grains derived from WT and pcp-b plants. Four T-DNA in vitro pollen hydration assay was carried out. Using a humid insertion lines were identified as mutant alleles of AtPCP-Ba, b, c chamber (providing 100% relative humidity) hydration of WT and d (Fig. S1) with PCP-B transcripts being undetectable in and pcp-ba/b/c pollen was compared over a 30-min period and anthers for pcp-ba-1, pcp-bb-1 and pcp-bc-1. PCP-Bd expression their hydration characteristics were found to be indistinguishable was found to be substantially downregulated, where the T-DNA (Fig. S7). Interestingly, comparison of WT pollen hydration on insertion was located in the promoter region of the gene (Figs S1, stigmas and in the humid chamber indicated that pollen hydrates S2). No obvious vegetative or reproductive morphological abnor- more rapidly, and attains a greater degree of hydration on stigmas malities were observed in any of the lines. Each individual pcp-b (Figs 5b–d, S7b–d). These data demonstrate that the stigma is line, a double mutant (pcp-bb/c) and a triple mutant (pcp-ba/b/ essential for rapid pollen hydration, and importantly, the absence 2016 The Authors New Phytologist (2017) 213: 764–777 New Phytologist  2016 New Phytologist Trust www.newphytologist.com New 770 Research Phytologist Fig. 3 Phylogenetic analysis of 46 PCP-B class genes in Arabidopsis and Brassica. The maximum-likelihood tree was constructed using nucleotide sequences of the predicted mature protein coding regions. Bootstrap values (1000 resamplings) > 50% are shown for interior branches. Branch length is scaled to the bar defined as 0.1 nucleotide substitutions per codon. The clades indicated by red and blue bars include PCP-Bs and ESFs, respectively. Genes are abbreviated as: AthB, Arabidopsis thaliana PCP-B-like; AlyB, Arabidopsis lyrata PCP-B-like; BoB, Brassica oleracea PCP-B-like; BrapaB, Brassica rapa PCP-B-like. Gene loci or scaffolds are shown adjacent to gene abbreviations. of PCP-B protein from the pollen coat does not impair the bio- which pollen could be washed off the stigma. Significantly higher physical ability of pollen to acquire water. numbers of pollen grains from the pcp-ba/b/c triple mutant (77%) were washed off WT stigmas compared with WT pollen (66%) 30 min post-pollination (Fig. 7a). However, an EM ultra- Pollen adhesion is reduced in pcp-b mutants structural analysis of the pollen from all pcp-b mutant lines In order to further characterize the phenotype of PCP-B mutants, revealed no discernible abnormalities in the characteristics of the a pollen adhesion assay was devised which tested the ease with pollen grain or pollen coat (Figs 7b–g, S8, S9). New Phytologist (2017) 213: 764–777  2016 The Authors www.newphytologist.com New Phytologist  2016 New Phytologist Trust New Phytologist Research 771 Fig. 4 Brassica and Arabidopsis PCP-Bs are gametophytically expressed late in pollen development. (a) mRNA gel blot analysis of Brassica oleracea PCP-B1 and PCP-B2 expression in leaves and reproductive tissue. Anthers from 9 to 11 mm buds have a fully degenerated tapetum and pollen is trinucleate. The arrows indicate the size of the transcript in base pairs. (b) Reverse transcription polymerase chain reaction (RT-PCR) expression analysis of AtPCP-B and AtPCP-B-like genes in Arabidopsis leaves, roots, stigmas and anthers (derived from stage 12 buds). GapC – cDNA input control for RT-PCR. (c) AtPCP-B gene expression in flower buds through development (stages 10–12). S, small (< 1 mm; uninucleate microspores); M, medium (c.1–1.5 mm; binucleate pollen); L, large (> 1.5 mm; unopened buds, trinucleate mature pollen). Arabidopsis flower bud stages are as defined by Smyth et al. (1990). (d) RNA–RNA in situ hybridization study of AtPCP-Bb expression in Arabidopsis thaliana anthers. Left panel: transverse anther section treated with an antisense (+ve) AtPCP-Bb DIG-labelled riboprobe, a clear signal (arrow) is observed within the majority of pollen grains. Right panel: longitudinal anther section treated with a control ‘sense’ (ve) riboprobe with no signal being detectable in pollen grains. Bars, 20 lm. resonance (NMR) and this made it possible to generate 3D struc- Initiation of pollen tube growth is delayed in pcp-b mutants tural predictions for the AtPCP-Bs by homologous alignment In order to determine if the early stages of pollen tube growth (Figs 9, S11, S12). All resulting models were statistically well- were affected by the delay in pollen hydration observed for pcp-b supported (Table S4). mutants, in vivo pollen tube lengths were estimated. After 2 h Based on the predicted 3D structure AtPCP-Bs likely share the WT pollen produced significantly longer tubes than pollen same intramolecular disulphide bonding pattern as ESF1.3 derived from all pcp-b mutant lines (Figs 8a, S10) with this effect (Figs 9a,b, S12) with all possessing a conserved cysteine-stabilized being largely maintained 4 h post-pollination (Fig. 8b). This motif consisting of an a-helix and three-stranded antiparallel result is consistent with data collected from the pollen hydration beta-sheet. In addition all AtPCP-Bs have a conserved aromatic assay where most mutants displayed impairment to the degree residue (Tyr-45 in AtPCP-Bc) that is also present in ESF1.3 and rate of hydration which in turn would likely cause a delay in (Trp-48) and other PCP-B-like proteins in Arabidopsis thaliana pollen tube emergence. Despite the observed post-pollination (Fig. 1). The surface electrostatic potential distribution for defects amongst the pcp-b mutants, there was no significant dif- AtPCP-Bc (Fig. 9c) is characterized by both positively and nega- ference in seed set following self-pollinations compared with WT tively charged domains with a prominent positively charged plants (Table S3), indicating that PCP-B protein function is extended loop held between Cys-36 and Cys-44 which lies in likely restricted to very early post-pollination events. close proximity to the conserved aromatic residue (Tyr-45). These features are broadly shared between all four AtPCP-Bs (Figs 9, S12). Structural prediction of AtPCP-Bs Our analyses have revealed the presence of PCP-B-like proteins Discussion in a wide range of angiosperm lineages with all sequences sharing the characteristic motif of eight cysteine residues in the mature Compatible pollination is a highly regulated process that requires polypeptide (Fig. 1). Costa et al. (2014) recently resolved the a suite of complementary pollen and pistil factors that act from structure of the PCP-B-like protein ESF1.3 by nuclear magnetic the moment of pollen contact through to successful fusion of 2016 The Authors New Phytologist (2017) 213: 764–777 New Phytologist  2016 New Phytologist Trust www.newphytologist.com New 772 Research Phytologist (a) (b) (c) (d) (f) (g) (e) Fig. 5 Mutations in Arabidopsis thaliana PCP-B genes result in altered pollen hydration profiles. Box plots depict the 25% quartile, median, 75% quartile and full range of values. (a–d) Pollen diameter distributions at 0, 10, 20 and 30 min following pollination; 1.8 pixels = 1 lm. *, P < 0.05; **, P < 0.001; ***, P < 0.0005 (Welsh’s t-test). Sample sizes: Col-0 (wild-type), 16; pcp-ba, pcp-bb, pcp-bc, pcp-bd, 15; pcp-ba/b, 16; pcp-ba/b/c, 15. (e–g) Pollen hydration is represented as percentage change in pollen diameter relative to diameter at 0 min (pollen diameter at initial contact with stigma) – distributions shown are for 10, 20 and 30 min post-pollination. *, P < 0.05; **, P < 0.005; ***, P < 0.000005 (Welsh’s t-test). Sample sizes: Col-0 (wild-type), 16; pcp-ba, pcp- bb, pcp-bc, pcp-bd, 15; pcp-ba/b, 16; pcp-ba/b/c, 15. gametes (Edlund et al., 2004; Hiscock & Allen, 2008). One of Our phylogenetic analysis of 46 gene sequences encoding PCP- the earliest events in the establishment of compatibility amongst B-like proteins in Arabidopsis and Brassica (Fig. 3) reveals an evo- species that possess dry stigmas, such as Arabidopsis thaliana,is lutionary history featuring frequent gene duplication events and the ability for pollen to gain access to stigmatic water (Elleman rapid sequence divergence around their conserved cysteine motif. et al., 1992; Safavian & Goring, 2013). This reproductive ‘check- These features are typical for gene families associated with repro- point’ requires activation of a basal stigmatic compatibility sys- duction and importantly can contribute to reproductive isolation tem by factor(s) that must be derived from pollen (Safavian & and speciation (Swanson & Vacquier, 2002; Clark et al., 2006; Goring, 2013). Our investigations reported here into small pol- Cui et al., 2015). Interestingly the PCP-Bs investigated here were len coating-borne cysteine-rich proteins point to an important found to be closely related to the ESF1s that encode embryo role for the PCP-Bs in these earliest stages of pollen–pistil com- developmental regulators (Costa et al., 2014) and these sequences patibility in Arabidopsis, as plants carrying mutations in PCP-B clustered in distinct phylogenetic clades, underlining their func- genes are impaired in pollen hydration. Importantly, PCP-Bs tional specialization (Figs 2, 3). Some AtPCP-B family members bear many hallmarks of intercellular signalling ligands and thus were more similar to genes in the closely related species A. lyrata are likely to be a central component of a pollen molecular ‘signa- suggesting that these have retained a specific function that pre- ture’ that defines compatibility. dates speciation. For example, AtPCP-Ba and AtPCP-Bc are PCP-Bs are structurally related proteins that have an ancient more closely related to the A. lyrata B4 and B1, respectively, than evolutionary origin, being widespread amongst angiosperm taxa. to other AtPCP-Bs (Fig. 3). Whereas putative Brassica New Phytologist (2017) 213: 764–777  2016 The Authors www.newphytologist.com New Phytologist  2016 New Phytologist Trust New Phytologist Research 773 (a) Fig. 6 Rate of pollen hydration is severely decreased in Arabidopsis thaliana pcp-b triple mutants. (a) Curves of mean pollen (b) hydration (% hydration is percentage change in pollen diameter). The vertical lines demark each 10-min period over which slopes were calculated. (b) Rate of change in pollen diameter during the first three 10-min periods of pollination. Average slopes  confidence intervals were produced by linear regression. *, P < 0.05; **, P < 0.005; ***, P < 0.001; ****, P < 0.0005 (Welsh’s t-test). (a) (b) (c) (d) (g) (e) (f) Fig. 7 Pollen morphology is unaffected in Arabidopsis thaliana pcp-ba/b/c pollen grains and pollen–stigma adhesion is weakened. (a) The mean percentage of wild-type (Col-0) and pcp-ba/b/c triple mutant pollen washed off WT stigmas in an adhesion assay 30 min post-pollination. Error bars represent the confidence interval. Sample sizes: WT stigmas, 64; triple mutant, 50. *, P < 0.001 (Welsh’s t-test). (b–g) Scanning electron microscopic (SEM) analysis of exine and pollen coat morphology in (b–d) WT and (e–g) pcp-ba/b/c triple mutant plants. Bars: (b, e), 10 lm; (c, d, f, g) 1 lm. orthologues were found in discrete clades more distant from the class proteins, are small secreted proteins that are encoded by a Arabidopsis PCP-Bs and could point to divergence of recognition rapidly evolving gene family. It is thus tempting to speculate that factors required for compatibility. Species-specific functionaliza- PCP-Bs not only regulate aspects of compatibility but may also tion of plant reproductive proteins that contribute to reproduc- contribute to reproductive barriers within the Brassicaceae. tive isolation have been documented for the pollen tube Our mutational study revealed that absence of AtPCP-Bs from attractant LURE proteins secreted by egg-accompanying synergid the pollen coat caused a series of interlinked phenotypes resulting cells of the embryo sac (Takeuchi & Higashiyama, 2012). from a primary defect in pollen hydration. We ascertained that Heterologous expression of an A. thaliana LURE protein in the pcp-b hydration defect was not caused by gross morphological Torenia fournieri synergid cells enabled A. thaliana pollen tubes perturbation of the pollen coat (Figs 7b–g, S8, S9) and that it to successfully locate and enter the embryo sac of this species. was only evident during the pollen–stigma interaction, as triple LURES are defensin-like CRPs and, in common with the PCP-B mutant pcp-ba/b/c pollen hydrated normally in a humid 2016 The Authors New Phytologist (2017) 213: 764–777 New Phytologist  2016 New Phytologist Trust www.newphytologist.com New 774 Research Phytologist (a) (b) Fig. 8 Extent of in vivo pollen tube growth is reduced for pcp-b mutants. Distance (in pixels) of pollen tube growth for wild-type and pcp-b mutants (a) 2 h and (b) 4 h post- pollination. Pollen was applied to stigmas of the Arabidopsis thaliana Col-0 A9-barnase male sterile line. Error bars represent  SD. Sample sizes: 4. *, P < 0.001; **, P < 0.0001; ***, P < 0.00001; ****, P < 0.000001 (Welsh’s t-test); 1 pixel = 0.625 lm. (a) Fig. 9 AtPCP-Bc structure prediction by SWISS-MODEL. (a) Amino acid sequence of PCP-Bc. Connection arrows, disulphide bonds; blue arrows, beta strands; red bar, (b) (c) alpha helix. (b) Cartoon model of predicted structure of AtPCP-Bc with indicated disulphide bonds and Tyrosine residue. (c) Distribution of electrostatic potential on AtPCP-Bc surface based on the predicted structure. Blue, positive; red, negative; white, hydrophobic residues. chamber (Fig. S7). Hydration rate, the degree of hydration and together, probably through different pollen tube receptors, to resulting pollen tube lengths were all found to be largely impaired ensure appropriate pollen tube guidance to the embryo sac amongst pcp-b single and combined mutants (Figs 5, 6, 8). We (Okuda et al., 2009; Takeuchi & Higashiyama, 2012, 2016; consider that the shorter tubes observed in pistils for pcp-b Wang et al., 2016). The severity of the triple pcp-ba/b/c mutant mutants are most likely the result of delayed pollen tube emer- reduced the degree and rate of pollen hydration to almost one gence rather than slower tube extension, because tube emergence third that of WT and due to the close genetic linkage of PCP-Bd is largely dependent on the degree of pollen hydration and pollen to PCP-Bc (< 10 kb) we were unable to recover and test the effect turgor (Taylor & Hepler, 1997). This inference was supported of a pcp-b quadruple mutant. Thus, it remains to be determined by the observation that triple mutant pollen adhered significantly if a complete hydration block could be achieved by abolishing all less well to stigmatic papillae 30 min post-pollination (Fig. 7a) – PCP-B proteins from the pollen coat. we observed that a significant component of this effect was due to Given the structural features of AtPCP-Bs and their homology WT pollen tubes initiating stigmatic penetration ahead of pcp-b to the ESF1 family of secreted developmental regulators we pro- pollen, thus anchoring them on the stigma, whereas substantially pose that PCP-Bs act as ligands to either directly or indirectly fewer mutant pollen had initiated germination (L. Wang & J. activate stigmatic targets that mediate transfer of water through Doughty, unpublished). the papilla plasma membrane. A substantial body of evidence Comparison of the severity of the hydration defects between now points to targeted stigmatic secretion as being a central fea- single and combined mutants revealed evidence of complex com- ture of compatible pollination in both A. thaliana and Brassica binatorial effects of PCP-Bs in the pollen–stigma interaction. and that the exocyst protein complex is essential to this process Out of the single mutant lines, pcp-bc presented the most statisti- (Samuel et al., 2009; Safavian & Goring, 2013; Safavian et al., cally robust hydration defect over the first 10-min period follow- 2014, 2015). The exocyst mediates tethering of secretory vesicles ing pollination, with pcp-bb having an almost identical hydration to target membranes (Zarsky et al., 2013) and stigmas from Ara- profile (Figs 5e, 6). Interestingly, the phenotype of the double bidopsis that carry mutations in Exo70A1, a key linker compo- pcp-bb/c mutant was not additive, but when combined with the nent of the exocyst-tethering machinery, have severe pollen pcp-ba mutant – which singly had no significant phenotype – hydration defects. Targeted secretion likely delivers factors to the pollen hydration was reduced dramatically (Figs 5e–g, 6). The plasma membrane adjacent to compatible pollen that mediate contrasting combinatorial effects of these mutants suggests that water transport. For instance, aquaporins, membrane-localized PCP-Ba may be acting as a ligand to activate a different stigmatic water transport proteins (Johanson et al., 2001; Quigley et al., hydration effector target to that of PCP-Bb and PCP-Bc, or that 2002; Maurel et al., 2008), could be deposited at the interface PCP-Ba acts to enhance activation of a putative stigmatic target with compatible pollen. A specific role for pollen coat factors working synergistically with other PCP-Bs. Similar complexity triggering such a response is supported by the observation that has been reported for synergid LURE proteins in T. fournieri and isolated B. oleracea pollen coat appears to evoke a secretory Arabidopsis where it seems likely that multiple LUREs work response by stigmatic papillae (Elleman & Dickinson, 1996). New Phytologist (2017) 213: 764–777  2016 The Authors www.newphytologist.com New Phytologist  2016 New Phytologist Trust New Phytologist Research 775 Cao L, Bandelac G, Volgina A, Korostoff J, DiRienzo JM. 2008. Role of Homology modelling of AtPCP-Bs provided strong support aromatic amino acids in receptor binding activity and subunit assembly of the for overall structural similarity with ESF1.3 (Figs 9b, S12). As cytolethal distending toxin of Aggregatibacter actinomycetemcomitans. Infection has been determined for ESF1.3 and other plant regulatory pep- and Immunity 76: 2812–2821. tides it is likely that the disulphide-stabilized cysteine motif is Chae K, Kieslich CA, Morikis D, Kim SC, Lord EM. 2009. A gain-of-function crucial for protein function of PCP-Bs (Ohki et al., 2011; Costa mutation of Arabidopsis lipid transfer protein 5 disturbs pollen tube tip growth and fertilization. Plant Cell 21: 3902–3914. et al., 2014). Intriguingly the AtPCP-Bs shared a functionally Chapman LA, Goring DR. 2010. Pollen–pistil interactions regulating essential aromatic residue with ESF1.3. Aromatic residues are a successful fertilization in the Brassicaceae. Journal of Experimental Botany 61: conserved feature of many plant regulatory peptides (Cao et al., 1987–1999. 2008; Okuda et al., 2009; Sugano et al., 2010; Costa et al., 2012; Clark NL, Aagaard JE, Swanson WJ. 2006. Evolution of reproductive proteins Sprunck et al., 2012) and are likely to be important in protein– from animals and plants. Reproduction 131:11–22. Costa LM, Marshall E, Tesfaye M, Silverstein KA, Mori M, Umetsu Y, protein interactions (Simpson et al., 2000). Otterbach SL, Papareddy R, Dickinson HG, Boutiller K et al. 2014. Central In conclusion, this study shows that AtPCP-Bs are important cell-derived peptides regulate early embryo patterning in flowering plants. mediators of pollen hydration, a key early ‘checkpoint’ of pollen– Science 344: 168–172. stigma compatibility. Their close evolutionary relationship to the Costa LM, Yuan J, Rouster J, Paul W, Dickinson H, Gutierrez-Marcos JF. ESF1 family of embryo developmental regulators, and their 2012. Maternal control of nutrient allocation in plant seeds by genomic imprinting. Current Biology 22: 160–165. broad similarity to other CRP regulatory proteins strongly sug- Cui X, Lv Y, Chen ML, Nikoloski Z, Twell D, Zhang DB. 2015. Young genes gest that they act through interaction with as yet unknown stig- out of the male: an insight from evolutionary age analysis of the pollen matic targets to activate the basal compatibility system. In transcriptome. Molecular Plant 8: 935–945. addition, PCP-B maintenance and diversity within Arabidopsis Dickinson HG. 1995. Dry stigmas, water and self-incompatibility in Brassica. and the Brassicaceae suggest that these proteins have the potential Sexual Plant Reproduction 8:1–10. Dickinson HG, Elleman CJ, Doughty J. 2000. Pollen coatings – chimaeric to contribute to prezygotic hybridization barriers. genetics and new functions. Sexual Plant Reproduction 12: 302–309. Doughty J, Dixon S, Hiscock SJ, Willis AC, Parkin IAP, Dickinson HG. 1998. PCP-A1, a defensin-like Brassica pollen coat protein that binds the S locus Acknowledgements glycoprotein, is the product of gametophytic gene expression. Plant Cell 10: We thank Susan Crennell for assistance with protein structural 1333–1347. Doughty J, Hedderson F, Mccubbin A, Dickinson H. 1993. Interaction between predictions, Ursula Potter for support with electron microscopy, a coating-borne peptide of the Brassica pollen grain and stigmatic S (self- Andrew James for plant maintenance, Jose Gutierrez-Marcos for incompatibility)-locus-specific glycoproteins. Proceedings of the National the kind gift of pAt5g61605:GUS and pAt2g16505:GUS reporter Academy of Sciences, USA 90: 467–471. lines, Tony Willis for protein sequencing, Sue Dixon for techni- Doughty J, Wong HY, Dickinson HG. 2000. Cysteine-rich pollen coat proteins cal assistance, Hugh Dickinson for insightful discussions, and Ed (PCPs) and their interactions with stigmatic S (incompatibility) and S-related proteins in Brassica: putative roles in SI and pollination. Annals of Botany 85: Nottingham, Hanis Jasmine Zaini and Ken Hayes for valuable 161–169. contributions during their undergraduate projects. This research Edgar RC. 2004. MUSCLE: multiple sequence alignment with high accuracy and was funded by a University of Bath postgraduate scholarship to high throughput. Nucleic Acids Research 32: 1792–1797. L.W. and a Nuffield Foundation award to J.D. Edlund AF, Swanson R, Preuss D. 2004. Pollen and stigma structure and function: the role of diversity in pollination. Plant Cell 16: S84–S97. Elleman CJ, Dickinson HG. 1986. Pollen–stigma interactions in Brassica. IV. Author contributions Structural reorganization in the pollen grains during hydration. Journal of Cell Science 80: 141–157. J.D. and L.W. planned and designed the research and wrote the Elleman CJ, Dickinson HG. 1990. The role of the exine coating in pollen– manuscript; L.W. was involved with all aspects of the research stigma interactions in Brassica oleracea L. New Phytologist 114: 511–518. with L.A.C. and R.J.E. contributing to expression analysis and Elleman CJ, Dickinson HG. 1996. Identification of pollen components regulating pollination-specific responses in the stigmatic papillae of Brassica C.C.P. to the SEM work; R.J.S. and B.Q. assisted with experi- oleracea. New Phytologist 133: 197–205. mental design and critical assessment of the manuscript. Elleman CJ, Franklin-Tong V, Dickinson HG. 1992. Pollination in species with dry stigmas: the nature of the early stigmatic response and the pathway taken by pollen tubes. New Phytologist 121: 413–424. References Fiebig A, Mayfield JA, Miley NL, Chau S, Fischer RL, Preuss D. 2000. Allen AM, Thorogood CJ, Hegarty MJ, Lexer C, Hiscock SJ. 2011. Pollen–pistil Alterations in CER6, a gene identical to CUT1, differentially affect long-chain interactions and self-incompatibility in the Asteraceae: new insights from lipid content on the surface of pollen and stems. Plant Cell 12: 2001–2008. studies of Senecio squalidus (Oxford ragwort). Annals of Botany 108: 687–698. Guex N, Peitsch MC, Schwede T. 2009. Automated comparative protein Alonso JM, Stepanova AN, Leisse TJ, Kim CJ, Chen HM, Shinn P, Stevenson structure modeling with SWISS-MODEL and Swiss-PdbViewer: a historical DK, Zimmerman J, Barajas P, Cheuk R et al. 2003. Genome-wide insertional perspective. Electrophoresis 30: S162–S173. mutagenesis of Arabidopsis thaliana. Science 301: 653–657. Hernandez-Pinzon I, Ross JH, Barnes KA, Damant AP, Murphy DJ. 1999. Arnold K, Bordoli L, Kopp J, Schwede T. 2006. The SWISS-MODEL Composition and role of tapetal lipid bodies in the biogenesis of the pollen coat workspace: a web-based environment for protein structure homology of Brassica napus. Planta 208: 588–598. modelling. Bioinformatics 22: 195–201. Heslop-Harrison Y, Shivanna KR. 1977. The receptive surface of the angiosperm Biasini M, Bienert S, Waterhouse A, Arnold K, Studer G, Schmidt T, Kiefer F, stigma. Annals of Botany 41: 1233–1258. Cassarino TG, Bertoni M, Bordoli L et al. 2014. SWISS-MODEL: modelling Hiroi K, Song M, Sakazono S, Osaka M, Masuko-Suzuki H, Matsuda T, Suzuki protein tertiary and quaternary structure using evolutionary information. G, Suwabe K, Watanabe M. 2013. Time-lapse imaging of self- and cross- Nucleic Acids Research 42: W252–W258. pollination in Brassica rapa. Annals of Botany 112: 115–122. 2016 The Authors New Phytologist (2017) 213: 764–777 New Phytologist  2016 New Phytologist Trust www.newphytologist.com New 776 Research Phytologist Hiscock SJ, Allen AM. 2008. Diverse cell signalling pathways regulate pollen– stigmatic defect for accepting compatible pollen. Plant Reproduction 27: stigma interactions: the search for consensus. New Phytologist 179: 286–317. 121–127. Hiscock SJ, Doughty J, Willis AC, Dickinson HG. 1995. A 7-kDa pollen Safavian D, Zayed Y, Indriolo E, Chapman L, Ahmed A, Goring DR. 2015. coating-borne peptide from Brassica napus interacts with S-locus glycoprotein RNA silencing of exocyst genes in the stigma impairs the acceptance of and S-locus-related glycoprotein. Planta 196: 367–374. compatible pollen in Arabidopsis. Plant Physiology 169: 2526–2538. Hulskamp M, Kopczak SD, Horejsi TF, Kihl BK, Pruitt RE. 1995. Samuel MA, Chong YT, Haasen KE, Aldea-Brydges MG, Stone SL, Goring Identification of genes required for pollen–stigma recognition in Arabidopsis DR. 2009. Cellular pathways regulating responses to compatible and self- thaliana. Plant Journal 8: 703–714. incompatible pollen in Brassica and Arabidopsis stigmas intersect at Johanson U, Karlsson M, Johansson I, Gustavsson S, Sjovall S, Fraysse L, Weig Exo70A1, a putative component of the exocyst complex. Plant Cell 21: AR, Kjellbom P. 2001. The complete set of genes encoding major intrinsic 2655–2671. proteins in Arabidopsis provides a framework for a new nomenclature for Schneider CA, Rasband WS, Eliceiri KW. 2012. NIH Image to ImageJ: 25 years major intrinsic proteins in plants. Plant Physiology 126: 1358–1369. of image analysis. Nature Methods 9: 671–675. Kho YO, Baer J. 1968. Observing pollen tubes by means of fluorescence. Schopfer CR, Nasrallah ME, Nasrallah JB. 1999. The male determinant of self- Euphytica 17: 298–302. incompatibility in Brassica. Science 286: 1697–1700. Kiefer F, Arnold K, Kuenzli M, Bordoli L, Schwede T. 2009. The SWISS- Shiba H, Takayama S, Iwano M, Shimosato H, Funato M, Nakagawa T, Che MODEL Repository and associated resources. Nucleic Acids Research 37: FS, Suzuki G, Watanabe M, Hinata K et al. 2001. A pollen coat protein, D387–D392. SP11/SCR, determines the pollen S-specificity in the self-incompatibility of Kleinboelting N, Huep G, Kloetgen A, Viehoever P, Weisshaar B. 2012. GABI- Brassica species. Plant Physiology 125: 2095–2103. Kat SimpleSearch: new features of the Arabidopsis thaliana T-DNA mutant Silverstein KAT, Moskal WA Jr, Wu HC, Underwood BA, Graham MA, Town database. Nucleic Acids Research 40: D1211–D1215. CD, VandenBosch KA. 2007. Small cysteine-rich peptides resembling Langdale JA. 1994. In situ hybridization. In: Freeling M, Walbot V, eds. The antimicrobial peptides have been under-predicted in plants. Plant Journal 51: maize handbook. New York, USA: Springer-Verlag, 165–179. 262–280. Letunic I, Bork P. 2007. Interactive Tree Of Life (iTOL): an online tool for Simpson PJ, Xie HF, Bolam DN, Gilbert HJ, Williamson MP. 2000. The phylogenetic tree display and annotation. Bioinformatics 23: 127–128. structural basis for the ligand specificity of family 2 carbohydrate-binding Li YL, Dai XR, Yue X, Gao XQ, Zhang XS. 2014. Identification of small modules. Journal of Biological Chemistry 275: 41137–41142. secreted peptides (SSPs) in maize and expression analysis of partial SSP genes in Smith AG, Eberle CA, Moss NG, Anderson NO, Clasen BM, Hegeman AD. reproductive tissues. Planta 240: 713–728. 2013. The transmitting tissue of Nicotiana tabacum is not essential to pollen Ma J-F, Liu Z-H, Chu C-P, Hu Z-Y, Wang X-L, Zhang XS. 2012. Different tube growth, and its ablation can reverse prezygotic interspecific barriers. Plant regulatory processes control pollen hydration and germination in Arabidopsis. Reproduction 26: 339–350. Sexual Plant Reproduction 25:77–82. Smyth DR, Bowman JL, Meyerowitz EM. 1990. Early flower development in Marshall E, Costa LM, Gutierrez-Marcos J. 2011. Cysteine-rich peptides (CRPs) Arabidopsis. Plant Cell 2: 755–767. mediate diverse aspects of cell–cell communication in plant reproduction and Sprunck S, Rademacher S, Vogler F, Gheyselinck J, Grossniklaus U, development. Journal of Experimental Botany 62: 1677–1686. Dresselhaus T. 2012. Egg cell-secreted EC1 triggers sperm cell activation Maurel C, Verdoucq L, Luu DT, Santoni V. 2008. Plant aquaporins: membrane during double fertilization. Science 338: 1093–1097. channels with multiple integrated functions. Annual Review of Plant Biology 59: Stone SL, Anderson EM, Mullen RT, Goring DR. 2003. ARC1 is an E3 595–624. ubiquitin ligase and promotes the ubiquitination of proteins during the Mayfield JA, Preuss D. 2000. Rapid initiation of Arabidopsis pollination requires rejection of self-incompatible Brassica pollen. Plant Cell 15: 885–898. the oleosin-domain protein GRP17. Nature Cell Biology 2: 128–130. Sugano SS, Shimada T, Imai Y, Okawa K, Tamai A, Mori M, Hara-Nishimura Notredame C, Higgins DG, Heringa J. 2000. T-Coffee: a novel method for fast I. 2010. Stomagen positively regulates stomatal density in Arabidopsis. Nature and accurate multiple sequence alignment. Journal of Molecular Biology 302: 463: 241–244. 205–217. Swanson WJ, Vacquier VD. 2002. The rapid evolution of reproductive proteins. Ohki S, Takeuchi M, Mori M. 2011. The NMR structure of stomagen reveals Nature Reviews Genetics 3: 137–144. the basis of stomatal density regulation by plant peptide hormones. Nature Takasaki T, Hatakeyama K, Suzuki G, Watanabe M, Isogai A, Hinata K. 2000. Communications 2: 512. The S receptor kinase determines self-incompatibility in Brassica stigma. Nature Okuda S, Tsutsui H, Shiina K, Sprunck S, Takeuchi H, Yui R, Kasahara RD, 403: 913–916. Hamamura Y, Mizukami A, Susaki D et al. 2009. Defensin-like polypeptide Takayama S, Shiba H, Iwano M, Asano K, Hara M, Che FS, Watanabe M, LUREs are pollen tube attractants secreted from synergid cells. Nature 458: Hinata K, Isogai A. 2000. Isolation and characterization of pollen coat proteins 357–361. of Brassica campestris that interact with S locus-related glycoprotein 1 involved Paul W, Hodge R, Smartt S, Draper J, Scott R. 1992. The isolation and in pollen-stigma adhesion. Proceedings of the National Academy of Sciences, USA characterisation of the tapetum-specific Arabidopsis thaliana A9 gene. Plant 97: 3765–3770. Molecular Biology 19: 611–622. Takeuchi H, Higashiyama T. 2012. A species-specific cluster of defensin-like Petersen TN, Brunak S, von Heijne G, Nielsen H. 2011. SignalP 4.0: genes encodes diffusible pollen tube attractants in Arabidopsis. PLoS Biology 10: discriminating signal peptides from transmembrane regions. Nature Methods 8: e1001449. 785–786. Takeuchi H, Higashiyama T. 2016. Tip-localized receptors control pollen tube Piffanelli P, Ross JHE, Murphy DJ. 1998. Biogenesis and function of the lipidic growth and LURE sensing in Arabidopsis. Nature 531: 245–248. structures of pollen grains. Sexual Plant Reproduction 11:65–80. Tamura K, Stecher G, Peterson D, Filipski A, Kumar S. 2013. MEGA6: Preuss D, Lemieux B, Yen G, Davis RW. 1993. A conditional sterile mutation molecular evolutionary genetics analysis version 6.0. Molecular Biology and eliminates surface components from Arabidopsis pollen and disrupts cell Evolution 30: 2725–2729. signaling during fertilization. Genes & Development 7: 974–985. Taylor LP, Hepler PK. 1997. Pollen germination and tube growth. Annual Quigley F, Rosenberg JM, Shachar-Hill Y, Bohnert HJ. 2002. From genome to Review of Plant Physiology and Plant Molecular Biology 48: 461–491. function: the Arabidopsis aquaporins. Genome Biology 3: RESEARCH0001. Updegraff EP, Zhao F, Preuss D. 2009. The extracellular lipase EXL4 is required Safavian D, Goring DR. 2013. Secretory activity is rapidly induced in stigmatic for efficient hydration of Arabidopsis pollen. Sexual Plant Reproduction 22: papillae by compatible pollen, but inhibited for self-incompatible pollen in the 197–204. Brassicaceae. PLoS ONE 8: e84286. Villar M, Gaget M, Said C, Knox RB, Dumas C. 1987. Incompatibility in Safavian D, Jamshed M, Sankaranarayanan S, Indriolo E, Samuel MA, Goring Populus: structural and cytochemical characteristics of the receptive stigmas of DR. 2014. High humidity partially rescues the Arabidopsis thaliana exo70A1 Populus alba and Populus nigra. Journal of Cell Science 87: 483–490. New Phytologist (2017) 213: 764–777  2016 The Authors www.newphytologist.com New Phytologist  2016 New Phytologist Trust New Phytologist Research 777 Wang T, Liang L, Xue Y, Jia PF, Chen W, Zhang MX, Wang YC, Li HJ, Yang Fig. S8 SEM analysis of exine layer and pollen coat morphology. WC. 2016. A receptor heteromer mediates the male perception of female attractants in plants. Nature 531: 241–244. Fig. S9 TEM analysis of exine layer and pollen coat morphology. Yost JM, Kay KM. 2009. The evolution of postpollination reproductive isolation in Costus. Sexual Plant Reproduction 22: 247–255. Fig. S10 Comparison of pollen tube growth for wild-type and Zarsky V, Kulich I, Fendrych M, Pecenkova T. 2013. Exocyst complexes multiple functions in plant cells secretory pathways. Current Opinion in Plant pcp-b triple mutant plants. Biology 16: 726–733. Fig. S11 Homologous alignments of ESF1.3 and AtPCP-Bs for protein structure predictions. Supporting Information Additional Supporting Information may be found online in the Fig. S12 Predicted protein structure homology models of Supporting Information tab for this article: AtPCP-Ba, b and d. Fig. S1 Locations of T-DNA insertions. Table S1 PCR primers used in this study Fig. S2 RT-PCR analysis results of stage 12 anthers in pcp-b Table S2 Numbers and abbreviations of predicted PCP-B-like mutants. proteins in species and families Fig. S3 N-terminal sequencing of two PCP-B proteins purified Table S3 Average seed count values of Arabidopsis wild-type and from Brassica oleracea pollen coat. pcp-b mutants Fig. S4 Phylogeny of 282 predicted PCP-B-like protein Table S4 Statistics for AtPCP-B protein structural predictions sequences. Methods S1 Histochemical staining for b-glucuronidase activity. Fig. S5 RNA–RNA in situ hybridization study of AtPCP-Bc expression in Arabidopsis thaliana anthers. Please note: Wiley Blackwell are not responsible for the content or functionality of any Supporting Information supplied by the Fig. S6 Histochemical staining for GUS activity driven by authors. Any queries (other than missing material) should be AtPCP-Ba and AtPCP-Bd promoters in Arabidopsis tissues. directed to the New Phytologist Central Office. Fig. S7 Pollen hydration profiles of wild-type and pcp-b triple mutant grains in a humid chamber. New Phytologist is an electronic (online-only) journal owned by the New Phytologist Trust, a not-for-profit organization dedicated to the promotion of plant science, facilitating projects from symposia to free access for our Tansley reviews. Regular papers, Letters, Research reviews, Rapid reports and both Modelling/Theory and Methods papers are encouraged. We are committed to rapid processing, from online submission through to publication ‘as ready’ via Early View – our average time to decision is <28 days. There are no page or colour charges and a PDF version will be provided for each article. The journal is available online at Wiley Online Library. Visit www.newphytologist.com to search the articles and register for table of contents email alerts. If you have any questions, do get in touch with Central Office (np-centraloffice@lancaster.ac.uk) or, if it is more convenient, our USA Office (np-usaoffice@lancaster.ac.uk) For submission instructions, subscription and all the latest information visit www.newphytologist.com 2016 The Authors New Phytologist (2017) 213: 764–777 New Phytologist  2016 New Phytologist Trust www.newphytologist.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png New Phytologist Wiley

PCP‐B class pollen coat proteins are key regulators of the hydration checkpoint in Arabidopsis thaliana pollen–stigma interactions

Loading next page...
 
/lp/wiley/pcp-b-class-pollen-coat-proteins-are-key-regulators-of-the-hydration-2X2Vb2r2Aa

References (81)

Publisher
Wiley
Copyright
Copyright © 2017 New Phytologist Trust
ISSN
0028-646X
eISSN
1469-8137
DOI
10.1111/nph.14162
pmid
27596924
Publisher site
See Article on Publisher Site

Abstract

Research PCP-B class pollen coat proteins are key regulators of the hydration checkpoint in Arabidopsis thaliana pollen–stigma interactions Ludi Wang, Lisa A. Clarke, Russell J. Eason, Christopher C. Parker, Baoxiu Qi, Rod J. Scott and James Doughty Department of Biology and Biochemistry, University of Bath, Claverton Down, Bath, BA2 7AY, UK Summary Author for correspondence: The establishment of pollen–pistil compatibility is strictly regulated by factors derived from James Doughty both male and female reproductive structures. Highly diverse small cysteine-rich proteins Tel: +44 0 1225 383485 (CRPs) have been found to play multiple roles in plant reproduction, including the earliest Email: bssjd@bath.ac.uk stages of the pollen–stigma interaction. Secreted CRPs found in the pollen coat of members Received: 31 May 2016 of the Brassicaceae, the pollen coat proteins (PCPs), are emerging as important signalling Accepted: 23 July 2016 molecules that regulate the pollen–stigma interaction. Using a combination of protein characterization, expression and phylogenetic analyses we New Phytologist (2017) 213: 764–777 identified a novel class of Arabidopsis thaliana pollen-borne CRPs, the PCP-Bs (for pollen coat doi: 10.1111/nph.14162 protein B-class) that are related to embryo surrounding factor (ESF1) developmental regula- tors. Single and multiple PCP-B mutant lines were utilized in bioassays to assess effects on pol- len hydration, adhesion and pollen tube growth. Key words: Arabidopsis thaliana, compati- Our results revealed that pollen hydration is severely impaired when multiple PCP-Bs are bility, pollen coat proteins, pollen hydration, pollen–stigma interaction, reproduction, sig- lost from the pollen coat. The hydration defect also resulted in reduced pollen adhesion and nalling. delayed pollen tube growth in all mutants studied. These results demonstrate that AtPCP-Bs are key regulators of the hydration ‘checkpoint’ in establishment of pollen–stigma compatibility. In addition, we propose that interspecies diver- sity of PCP-Bs may contribute to reproductive barriers in the Brassicaceae. important in maintaining species barriers (Swanson & Vacquier, Introduction 2002; Takeuchi & Higashiyama, 2012). Pollination in angiosperms involves multiple phases of interac- Members of the Brassicaceae family (which includes Brassica tion between female reproductive tissues of the gynoecium and and Arabidopsis species) are characterized by having stigmas of the male reproductive unit, pollen, and subsequently the pollen the ‘dry’ type, which lack sticky secretions such as those present tube it produces on germination (Hiscock & Allen, 2008). The in species of the Solanaceae (Heslop-Harrison & Shivanna, process is highly selective such that the majority of heterospecific 1977). Dry stigmas are highly discriminatory, reducing the prob- pollen fails to effect syngamy and indeed intraspecific pollination ability that heterospecific pollen grains and pathogenic spores will can also be blocked in those species which possess self- be captured, hydrate and germinate on their surfaces. In this sys- incompatibility (SI) systems. Such prezygotic reproductive barri- tem compatibility is established at the stigma surface within min- ers are evolutionarily advantageous as they limit wasted mating utes of pollination when compatible grains gain access to opportunities, contribute to reproductive isolation and facilitate stigmatic water, but incompatible pollen generally fail to fully outbreeding when SI is present (Yost & Kay, 2009; Smith et al., hydrate and germinate. Thus, pollen hydration on the stigma sur- 2013). The establishment of compatibility is complex and face is essential for successful reproduction and is a strictly regu- involves a suite of biophysical and molecular recognition factors lated checkpoint centred in the stigma (Dickinson, 1995; Ma that operate throughout the pollination process; from pollen cap- et al., 2012; Hiroi et al., 2013). Possession of a dry stigma comes ture by the stigma, pollen hydration, germination and stigmatic with the requirement that the exine surface of conspecific pollen penetration, to polarized tube growth through the pistil (reviewed must carry a coating (tryphine) (Dickinson, 1995; Dickinson in Chapman & Goring, 2010). Although much progress has been et al., 2000). Tryphine is a complex mixture of lipids, proteins, made in elucidating the mechanisms that regulate compatibility glycoconjugates and pigments (Piffanelli et al., 1998; Hernandez- in a broad range of species, it is clear that great mechanistic diver- Pinzon et al., 1999) that confers adhesive properties to the grain, sity exists, with no common system in operation (Hiscock & provides a conduit for water to pass from the stigma to effect pol- Allen, 2008; Allen et al., 2011). Such diversity is considered to be len hydration and, importantly, carries factors that determine 764 New Phytologist (2017) 213: 764–777  2016 The Authors www.newphytologist.com New Phytologist  2016 New Phytologist Trust This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. New Phytologist Research 765 compatibility (Dickinson, 1995; Safavian & Goring, 2013). compatibility factor EXO70A1 (Stone et al., 2003; Samuel et al., Pollen access to stigmatic water requires targeted secretion in the 2009). Other CRPs belonging to the PCP-A class of Brassica pol- stigma immediately adjacent to a compatible pollen grain (Dick- len coat proteins such as SLR-BP1 and PCP-A1, bind the stig- inson, 1995) and it is now well established that in the Brassi- matic proteins S-locus related 1 (SLR1) and S-locus glycoprotein caceae this involves exocyst-mediated tethering of secretory (SLG), respectively, and thus are likely to function in the pollen– vesicles to the stigmatic plasma membrane (Samuel et al., 2009; stigma interaction, even though their precise function remains to Safavian & Goring, 2013; Safavian et al., 2015). be determined (Doughty et al., 1998; Takayama et al., 2000). A Despite progress in identifying molecular regulators of com- further class of Brassica pollen coat CRPs, the PCP-Bs, have been patibility in stigmas, relatively little is known about the pollen- described and also are good candidates for regulators of the earli- borne signals that establish it. Components of the pollen coat are est phases of the pollen–stigma interaction (Doughty et al., most likely to mediate compatibility due to the intimate interac- 2000). Thus, there is a growing body of evidence demonstrating tion of this layer with the surface of stigmatic papilla cells and the that the pollen coat carries factors that mediate both incompati- speed of pollen acceptance (Elleman & Dickinson, 1986, 1990; bility and compatibility, and that cysteine-rich pollen coat pro- Preuss et al., 1993). Indeed, application of isolated pollen coat to teins are important to molecular dialogue in the pollen–stigma the stigma surface evokes a rapid expansion of the stigmatic outer interaction. Although many studies have focused on the mecha- wall layer (Elleman & Dickinson, 1990, 1996) and isolated pol- nisms of self-incompatibility (SI) in a range of species, the molec- len stimulates the production of structures resembling vesicles in ular regulation of self-compatibility (SC) in flowering plants is the stigma apoplast beneath the pollen contact site (Elleman & still poorly understood. Dickinson, 1996). In this study, we report on the identification of four Analysis of pollen coat components and mutational studies in A. thaliana PCP-B encoding genes termed AtPCP-Ba Arabidopsis thaliana have shed light on factors that influence the (At5g61605), AtPCP-Bb (At2g29790), AtPCP-Bc (At2g16535) pollen–stigma interaction. A number of these appear to be bio- and AtPCP-Bd (At2g16505), which are expressed gametophyti- physical in nature, for example eceriferum (cer) mutant pollen cally late in pollen development. By utilizing T-DNA insertion fails to hydrate due to the elimination of very long chain lipids lines carrying single, double and triple AtPCP-B gene knockouts, from the pollen coat (Preuss et al., 1993; Hulskamp et al., 1995; we examined the impact of these mutations on pollen morphol- Fiebig et al., 2000). Hydration defects have also been reported in ogy and the pollen–stigma interaction. Phenotypic analyses mutants for extracellular lipase 4 (EXL4) and GRP17, an revealed defects in pollen hydration and delays in pollen tube oleosin-domain-containing glycine-rich protein which may work growth for single and combined mutants compared with wild- cooperatively to alter the lipid composition at the pollen–stigma type. Triple mutant pcp-ba/b/c pollen displayed a substantially interface to facilitate the passage of water to the grain (Mayfield reduced hydration rate on stigmas, delayed pollen tube growth, & Preuss, 2000; Updegraff et al., 2009). Work in Brassica has led as well as weakened anchoring to the stigma surface. Importantly, to the identification of several groups of small cysteine-rich pollen no impact on pollen morphology was revealed in this study coat proteins (Doughty et al., 1993, 1998, 2000; Hiscock et al., though the mutants presented striking effects on early post- 1995; Schopfer et al., 1999; Takayama et al., 2000; Shiba et al., pollination events. Such evidence suggests the AtPCP-Bs act as 2001). Importantly these polypeptides, rather than having major important regulatory factors during the earliest stages of the pol- effects on biophysical properties of the pollen coat, act as ligands len-stigma interaction by establishing a molecular dialogue and have been demonstrated to bind a number of stigmatic between the stigma and pollen grains. proteins. In recent years, a broad range of cysteine-rich proteins (CRPs) Materials and Methods have been identified in plants having functions in cell signalling, development and defence (Silverstein et al., 2007; Li et al., 2014). Plant material and growth conditions They are all characterized by being small (< 160 amino acids), having a conserved N-terminal signal peptide and a C-terminal Brassica oleracea var. alboglabra L. homozygous for incompatibil- cysteine-rich region with the pattern of cysteines determining ity haplotypes S25 and S29 (Horticultural Research Interna- their classification (Marshall et al., 2011). A number function in tional, Wellesbourne, UK) was used for isolation of BoPCP-B1 plant reproduction, including pollen–stigma self-recognition and BoPCP-B2 pollen coat proteins, respectively. Arabidopsis (Schopfer et al., 1999; Shiba et al., 2001), pollen tube growth thaliana (L.) Heynh. T-DNA insertion lines SALK_207087, and guidance (Chae et al., 2009; Okuda et al., 2009), and early SALK_062825, SALK_072366 (Alonso et al., 2003) were embryo development (Marshall et al., 2011; Costa et al., 2014). obtained from the Nottingham Arabidopsis Stock Centre Several families of CRPs have been identified in the pollen coat (NASC). GABI_718B04 was purchased from GABI-KAT of Brassica and Arabidopsis, and some have confirmed roles in the (Kleinboelting et al., 2012). T-DNA insertion sites and their pollen–stigma interaction. In self-incompatible species, the S- respective mutant alleles are detailed in Supporting Information locus cysteine-rich protein (SCR/SP11) (Schopfer et al., 1999; Fig. S1. Single gene T-DNA insertion lines were backcrossed to Shiba et al., 2001), acts as the male determinant and interacts wild-type (WT; Col-0) at least three times before phenotyping. with the S-receptor kinase (SRK) (Takasaki et al., 2000), to trig- pcp-bb/c and pcp-ba/b/c mutant lines were created using stan- ger pollen rejection by targeted degradation of the basal dard crossing procedures. PCP-B transcript status for each 2016 The Authors New Phytologist (2017) 213: 764–777 New Phytologist  2016 New Phytologist Trust www.newphytologist.com New 766 Research Phytologist mutant was confirmed by reverse transcription polymerase chain measured in pixels using IMAGEJ software (Schneider et al., 2012). reaction (RT-PCR) (Fig. S2). Primers for RT-PCR are detailed Pollen hydration (%) was calculated using the equation: pollen in Table S1. The A9-barnase male sterile line was provided by hydration (%) = (pollen diameter  initial pollen diameter)/initial Rod Scott, University of Bath, UK (Paul et al., 1992). GUS pollen diameter. Slopes were determined using 11 data points (b-glucuronidase) reporter lines pAt5g61605:GUS and during the 0–10 min, 10–20 min, or 20–30 min time periods pAt2g16505:GUS were provided by Jose F. Gutierrez-Marcos using the linear regression curve f = a x + b. All statistical analyses (University of Warwick, UK). were carried out using Microsoft Excel 2013. Arabidopsis thaliana plants were propagated in Levington F2 + S compost (Soils HS Limited, Wotton-Under-Edge, UK) in Pollen adhesion assay a controlled environment room with a 16 h : 8 h, light : dark pho- 2 1 toperiod provided by fluorescent lighting (130 lmol m s ). Stigmas of A. thaliana A9-barnase plants were hand-pollinated Temperature was maintained at 21  1°C with 60% relative using freshly dehiscent anthers from WT and pcp-ba/b/c lines. humidity. Brassicas were grown in a glasshouse at 21°C with a Pollen was applied as a monolayer. After 30 min the flower was 16 h : 8 h, light : dark photoperiod. excised from the plant and placed into 0.5 ml of fixative (60% v/v ethanol, 30% v/v chloroform, 10% v/v acetic acid) in a 1.5-ml microfuge tube. The sample was immediately shaken 10 times RT-PCR and RNA gel blot analysis using short sharp strokes to dislodge pollen that was not strongly Anthers were collected from A. thaliana stage 10–12 flower buds adherent to the stigma. The flower was then removed and placed (Smyth et al., 1990), stigmas from open flowers of the A. thaliana into a separate microfuge tube. Both samples were retained for A9-barnase line, roots from 2-wk-old seedlings grown on 0.59 pollen counting. Fifty microlitres of aceto-orcein stain (1%) was MS plates and leaves from fully-grown rosettes. RNA was added to the tubes and incubated overnight at room temperature extracted using a PureLink RNA Mini Kit (Thermo Fisher Sci- (RT). ‘Washed-off’ pollen samples were centrifuged at 13 000 g entific, Loughborough, UK). cDNA synthesis was carried out (10 min), excess fixative was removed and the pellet was resus- using the ProtoScript II First Strand cDNA Synthesis Kit (New pended in 10 ll of 50% glycerol before counting. Stigmas were England Biolabs, Hitchin, UK). DNA amplification utilized excised from stained flowers before mounting in 50% glycerol DreamTaq Green PCR Master Mix (29) (Thermo Fisher Sci- and were squashed on a slide to ensure all pollen was visible for entific). Primers for DNA amplification are detailed in Table S1. counting. RNA gel-blot analysis was carried out as described previously (Doughty et al., 1998) using polyadenylated mRNA (450 ng) Pollen tube growth assay from leaves, stigmas, pollen and anthers derived from a range of bud sizes (whole flower buds for anthers of size < 2 mm). Pollinations were initiated on A. thaliana male sterile A9-barnase Labelling of gene-specific BoPCP-B probes (covering the coding stigmas and allowed to proceed for 2 or 4 h before stigmas were region of the gene from aa residue 40 to the C-terminus) was excised and incubated in fixative (60% v/v ethanol, 30% v/v chlo- conducted using the Prime-a-Gene Labeling System (Promega), roform, 10% v/v acetic acid) overnight. After removal of fixative, with modifications to the deoxynucleotidetriphosphates mix to stigmas were incubated in 8 M NaOH for 20 min then washed in permit double labelling with dATP and dCTP (a- P, 100 lCi, dH O three times, each for 5 min. Samples were transferred to 6000 Ci mmol each). 0.1% decolourized aniline blue (0.1% w/v aniline blue in 0.1 M K PO , pH 11) for 1 h before imaging (Kho & Baer, 1968). 3 4 Pollen hydration assays Microscopy For in vivo hydration assays, pollen grains derived from WT, pcp-ba (SALK_207087), pcp-bb (SALK_062825), pcp-bc Imaging of pollen hydration on stigmas, hydration in a humid (SALK_072366), pcp-bd (GABI_718B04), pcp-bb/c and pcp-ba/ chamber and GUS histochemical staining (Methods S1) of flow- b/c lines were applied to stigmas of the A9-barnase male sterile ers and leaves was carried out using a Nikon (Surrey, UK) line. Freshly opened mature flowers were used (retained on the SMZ1500 dissection microscope coupled to Nikon Digital Sight plant) and pollen was applied in a monolayer with an eyelash. At DS-U1 camera. A Nikon Eclipse 90i epifluorescence microscope least eight independent stigmas were used in assays from each (910 objective) with Nikon Digital Sight DS-U1 camera was line. For in vitro hydration assays, pollen from WT and pcp-ba/b/ used for imaging pollen tubes stained with aniline blue and for c triple knockout lines were placed on a slide in a humid chamber anthers stained for GUS activity. (100% relative humidity). Pollen grains on stigmas were pho- tographed under a dissecting microscope immediately after polli- Scanning and transmission electron microscopy nations were initiated (designated as time point zero). Subsequent images were captured every minute for 30 min. For humid cham- Pollinated stigmas were dry-fixed using the method of Elleman ber assays pollen was photographed 1 min after pollen grains were & Dickinson (1986). Samples were washed with 0.1 M sodium placed, then the chamber was sealed, after which images were cacodylate buffer pH 7.4 and prepared for scanning electron taken every minute for 30 min. Equatorial diameter of pollen was microscopy (SEM) and transmission electron microscopy (TEM) New Phytologist (2017) 213: 764–777  2016 The Authors www.newphytologist.com New Phytologist  2016 New Phytologist Trust New Phytologist Research 767 by a modified method of Villar et al. (1987) using 2.5% glu- described by Doughty et al. (1993, 1998). Both BoPCP-B1 and taraldehyde and low viscosity resin. Samples for SEM were gold- BoPCP-B2 co-purified with the previously characterized PCP-A1 coated and imaged using a Jeol JSM640LV Scanning Electron polypeptide following C18 RP-HPLC and were separated by Microscope (Jeol, Tokyo Japan). Samples for TEM were ultra- cation exchange chromatography (Doughty et al., 1998). BoPCP- thin sectioned (100 nm) on an Ultracut-E ultramicrotome (Leica, B1 and BoPCP-B2 were isolated from S25 and S29 incompatibility London, UK) and imaged by Jeol JEM1200EXII transmission lines of B. oleracea var alboglabra, respectively. BoPCP-B protein electron microscope. samples were prepared for N-terminal sequencing as described previously (Doughty et al., 1998). Multiple sequence alignment and phylogenetic analysis 0 0 5 and 3 rapid amplification of cDNA end (RACE) PCP-B-like sequences were retrieved from available complete polymerase chain reaction cloning of BoPCP-B1 and plant genomes completed to at least scaffold level using TBLASTN BoPCP-B2 database searches (PHYTOZOME, https://phytozome.jgi.doe.gov; Comparative Genomics, COGE, https://genomevolution.org). Polyadenylated RNA was isolated from 100 mg of anthers Nucleotide sequence alignments of PCP-B homologous genes derived from 9 to 11 mm flower buds of B. oleracea var were generated using MUSCLE (codon) (Edgar, 2004). Codons of alboglabra (homozygous for S25 and S29 incompatibility haplo- protein coding sequences were translated into amino acid types) using a QuickPrep Micro mRNA purification kit (Phar- sequences before the alignment was performed. Aligned amino macia Biotech, Piscataway, NJ, USA). Cloning of BoPCP-B1 0 0 acid sequences were then replaced by the original codons. Graph- and BoPCP-B2 cDNA sequences was carried out using a 5 /3 ical output of protein sequence alignment was generated by RACE kit (Boehringer Mannheim) following the manufacturer’s JALVIEW using ‘CLUSTALX’ colour coding. Phylogenetic trees were instructions. One microgram of mRNA was subjected to first- built using the maximum-likelihood statistical method in MEGA strand cDNA synthesis using an oligo(dT) kit primer. First v.6 (Tamura et al., 2013). The initial tree was determined by the round 3 RACE cloning of BoPCP-B1 utilized a degenerate neighbour-joining method (NJ). The phylogeny test was carried primer based on the peptide sequence AGNAAK[P/Q] which is out using the bootstrap method (1000 replications). Phylogenetic common to both BoPCP-B proteins (5 -GC-GGATCC- trees were displayed using iTOL (Letunic & Bork, 2007). GCIGGIAA[C/T]GCIGCIAA[A/G]C-3 , where I represents inosine) in conjunction with a kit anchor primer. This was followed by two rounds of PCR using degenerate nested primers Protein structure prediction and modelling (5 -GC-GGATCC-AA[A/G]CA[A/G]ACICCITG[C/T]CA[C/T] 0 0 Protein sequence alignment of PCP-Bs and 2RU1 was carried G-3 and 5 -GC-GGATCC-AA[A/G]CCIAA[C/T]CA[C/T]AC out with T-COFFEE (Notredame et al., 2000). PCP-B protein ITG) based on the BoPCP-B1 specific peptide sequences models were built using SWISS-MODEL (Arnold et al., 2006; Guex KQTPCHE and KPNHTC, respectively. 5 RACE was con- et al., 2009; Kiefer et al., 2009; Biasini et al., 2014) based on the ducted utilizing sequence-specific primers SP1 and SP2 (derived modified T-COFFEE alignment result. 3D cartoon models and from 3 RACE) for cDNA synthesis and nested amplification of 0 0 electrostatic potential surface models were produced by PYMOL the 5 region of the BoPCP-B1 cDNA (SP1 5 -GCTTGCC 0 0 (v.1.7.4; Schr€odinger, Cambridge, UK). GCACCTACGCG-3 and SP2 5 -CAT GTAGCACATGT 0 0 TTTGAGC-3 ). For BoPCP-B2, first round 3 RACE was carried out as described for BoPCP-B1 followed by one further round of In situ hybridization PCR utilizing a degenerate nested primer (5 -GC-GGATCC- Flower buds were excised from inflorescences and immediately ATGAA[C/T]TG[C/T]GA[C/T]ACICA[A/G] G) based on the fixed in fresh 4% paraformaldehyde for 16 h at 4°C with an ini- BoPCP-B2 specific peptide sequence MNCDTQD. 5 RACE tial 10 min under low vacuum. Tissues were embedded in Para- utilized sequence-specific primers SP1 and SP2 (derived from plast Xtra (Sigma), sectioned (7–10 lm) and prepared for 3 RACE sequence) for cDNA synthesis and nested amplification 0 0 probing as described by Langdale (1994), except for the protease of the 5 region of the BoPCP-B1 cDNA (SP1 5 -GG 0 0 treatment where sections were incubated for 30 min at 37°Cin CTTCCCAGATTTAGTGAC-3 and SP2 5 -GTGACACA 50 lgml proteinase K (Sigma). Both antisense and sense ACAAGAACAACTGCG-3 ). probes were synthesized using a SP6/T7 digoxigenin RNA label- ing kit (Boehringer Mannheim, Lewes, UK), according to the Results manufacturer’s instructions. Probes covered the protein coding sequence of the AtPCP-Bb and AtPCP-Bc cDNAs. The pollen coat of Brassica contains polymorphic PCP-B class cysteine-rich proteins Protein purification and N-terminal sequencing In a previous study that characterized the SLG-binding pollen BoPCP-B1 and BoPCP-B2 proteins were purified from total pol- coat protein PCP-A1 from Brassica oleracea (Doughty et al., len coat proteins by a combination of gel filtration, RP-HPLC 1998) two polypeptides were found to copurify with PCP-A1. and cation exchange chromatography following the protocol These were purified to homogeneity and subjected to N-terminal 2016 The Authors New Phytologist (2017) 213: 764–777 New Phytologist  2016 New Phytologist Trust www.newphytologist.com New 768 Research Phytologist sequencing. Each shared an identical six amino acid N-terminal two distinct clades (Fig. 2). One clade includes ESF1-encoding domain and several conserved cysteine residues arranged in a genes clustering into a group of five sequences. Of the other clade unique pattern with respect to other known pollen coat protein four of the sequences fall into a cluster which, following wider families (Fig. S3). These were subsequently named BoPCP-B1 phylogenetic analysis across the Brassica and Arabidopsis genera, and BoPCP-B2 (for B. oleracea pollen coat protein, class B, 1 and placed them in a clade that included genes encoding pollen coat- 2, respectively). The partial BoPCP-B polypeptide sequences per- derived BoPCP-Bs (Fig. 3). Expression analyses confirmed these mitted cloning of their respective full-length cDNAs by RACE four genes as being largely anther-specific (Fig. 4). Taken PCR (GenBank accession numbers: PCP-B1, KX099662; PCP- together, these data suggest that the Arabidopsis sequences are B2, KX099663). BoPCP-B1 and BoPCP-B2 are predicted to likely orthologues of the B. oleracea PCP-Bs and, hence, we encode proteins of 79 and 84 amino acids, respectively, with both named them AtPCP-Ba (At5g61605), AtPCP-Bb (At2g29790), having a putative 25 amino acid secretory signal peptide AtPCP-Bc (At2g16535) and AtPCP-Bd (At2g16505). (Petersen et al., 2011) and a conserved pattern of eight cysteine Iterative BLAST searches Arabidopsis and Brassica genera iden- residues in the mature protein. Based on the N-terminal sequence tified 46 PCP-B-like sequences in total across four species. Phylo- data, mature BoPCP-B1 and BoPCP-B2 are estimated to have genetic analysis of these protein sequences revealed not only the M s of 5490 and 6109, respectively. The localization of the high degree of polymorphism across the family, but also the pres- BoPCP-Bs to the pollen coat, together with their broad similarity ence of two distinct groupings of sequences, the PCP-B and to other small cysteine-rich proteins such as PCP-A1 (Doughty ESF1-containing clades (Fig. 3). These groupings may reflect et al., 1998) and the pollen self-incompatibility determinant SCR functional specialization into seed and pollination-specific roles. (Shiba et al., 2001), suggested that they could potentially Wider phylogenetic BLAST analyses across all known plant lin- function in the pollen–stigma interaction. eages identified 282 predicted PCP-B-like protein sequences in seven angiosperm families (36 species in total, Fig. S4; Table S2). In addition to the Brassicaceae, PCP-B-like proteins were found PCP-Bs are evolutionarily widespread and have homology in the Poaceae, Nelumbonaceae, Solanaceae, Malvaceae, Phry- to Arabidopsis Embryo Surrounding Factor 1 maceae and Pedaliaceae. Thus, the PCP-Bs are members of a developmental regulators wider family of highly polymorphic, though structurally related In order to facilitate subsequent functional analyses of PCP-B proteins, having an ancient evolutionary origin that predates the class pollen coat proteins putative homologues were identified in split between monocot and eudicot lineages. the model plant Arabidopsis. BLAST searching of the Arabidop- sis genome using BoPCP-B sequences revealed the presence of Arabidopsis and Brassica oleracea PCP-B genes are twelve PCP-B-like genes. All sequences were predicted to encode expressed in maturing pollen small, typically basic secreted proteins that shared a common cys- teine pattern of seven or eight cysteines in the mature polypeptide RNA gel-blot analysis for the Brassica oleracea PCP-B1 and PCP- (Fig. 1). Importantly, three members of this Arabidopsis gene B2 genes indicated high levels of expression in anthers (Fig. 4a). family (At1g10747, At1g10745 and At1g10717) encode the cen- Transcripts were first detected at low levels in anthers derived tral cell-derived Embryo Surrounding Factor 1 (ESF1) signalling from 5 to 7 mm flower buds reaching a maximum by the proteins known to shape early embryo development and pattern- 9–11 mm bud stage by which time pollen is tricellular and tapetal ing (Costa et al., 2014). A phylogenetic analysis of mature pro- cells that line the anther locule are fully degraded (Doughty et al., tein-encoding gene regions, based on prior alignment of amino 1998). This late pattern of expression in anther development acid sequences, revealed that the AtPCP-B-like sequences fall into infers that BoPCP-Bs are likely to be gametophytically derived Fig. 1 Protein sequence alignment of Brassica oleracea PCP-B1 and PCP-B2 with all known Arabidopsis thaliana (Col-0) PCP-B-like proteins. AtPCP-Bs are: AtPCP-Ba (At5g61605), AtPCP-Bb (At2g29790), AtPCP-Bc (At2g16535) and AtPCP-Bd (At2g16505). At1g10747, At1g10745 and At1g10717 are ESF1.1, ESF1.2 and ESF1.3, respectively. Sequence conservation, quality and consensus is displayed below. Colour coding follows the default output for Clustal X (http://www.jalview.org/help/html/colourSchemes/clustal.html). New Phytologist (2017) 213: 764–777  2016 The Authors www.newphytologist.com New Phytologist  2016 New Phytologist Trust New Phytologist Research 769 c) were assessed utilizing the pollen hydration assay. A quadruple mutant could not be generated due to the close genetic linkage of PCP-Bc and PCP-Bd (c. 9 kb apart). Pollen equatorial diameter was recorded for 30 min following placement of pollen on stig- mas. Four time points (0, 10, 20 and 30 min) were selected for analysis of the difference of pollen hydration between WT and mutant lines. In addition, the rate of pollen hydration was assessed for each of the three 10-min periods following pollina- tion. On initiation of pollination (0 min) no significant differ- ence was found between the diameters of pollen derived from WT and mutant lines (Fig. 5a). However, at subsequent time Fig. 2 Phylogenetic analysis of 12 PCP-B class genes in Arabidopsis points pcp-bc and pcp-ba/b/c pollen grains were significantly less thaliana (Col-0). The maximum-likelihood tree was constructed by using hydrated than WT pollen (Fig. 5b–d). Assessment of pollen the nucleotide sequences of predicted mature protein coding regions. Branch lengths are proportional to the bar, defined as 0.1 nucleotide hydration as percentage change in diameter (from time point substitutions per codon. The percentage bootstrap values (1000 zero) demonstrated that the degree of pollen hydration was sig- resamplings) > 50% are shown by interior branches. nificantly lower in the pcp-bb, pcp-bc, pcp-bb/c and pcp-ba/b/c mutant lines compared with WT at each time point (Fig. 5e–g). rather than being products of the tapetum. No expression was Despite there being no statistically significant difference in pollen hydration for pcp-ba and pcp-bd compared with WT, median detected in leaves and stigmas, and only very low transcript levels were detected in mature pollen. This expression pattern exactly pollen diameters, hydration percentage and overall ranges in the mirrors that of the pollen coat protein gene PCP-A1 (Doughty data suggested that pcp-ba and pcp-bd mutations also negatively et al., 1998). In order to determine which of the twelve Arabidop- impact on pollen hydration (Fig. 5e–g). sis PCP-B-like genes were likely orthologues of the B. oleracea We extended the analysis to determine the rate of pollen PCP-B sequences, expression analysis was carried out by RT- hydration on stigmas for mutant and WT pollen. Slopes were PCR (Fig. 4b). Six of the genes were found to be expressed in produced by linear regression based on pollen grain diameter stage 12 anthers though two of these, At1g10747 and during each 10-min period following pollination. During the At1g10745, have previously been characterized as central cell- first 10-min period WT pollen hydrates rapidly with this rate derived ESF1 signalling proteins involved in embryo patterning decreasing dramatically during the second and third 10-min (Costa et al., 2014). The remaining four anther-expressed genes periods. A similar overall pattern was observed for all pcp-b (At5g61605, At2g29790, At2g16535 and At2g16505, AtPCP-Ba mutant lines (Fig. 6a). During the first 10-min period of polli- to d, respectively) were found to share a similar temporal expres- nation, the hydration rate of WT pollen was significantly higher sion pattern to the B. oleracea PCP-Bs (Fig. 4a,c). In addition, than that for pcp-bb and pcp-bc. The pcp-bb/c double mutant RNA–RNA in situ hybridization for AtPCP-Bb (Figs 4d, S5) and also demonstrated a significantly affected hydration rate though AtPCP-Bc and promoter-GUS fusions for AtPCP-Ba and this was not greater than either single mutant. However, intro- AtPCP-Bd (Fig. S6) confirmed high-level expression in pollen, duction of pcp-ba into the pcp-bb/c line creating the pcp-ba/b/c further validating the status of this group as pollen coat protein- triple mutant had a dramatic effect on pollen hydration. pcp- encoding genes. Taken together with the phylogenetic analysis ba/b/c pollen hydrated at a substantially lower rate than either that placed these four genes in the same clade as BoPCP-Bs, it is WT or pollen from the pcp-bb/c double-mutant line (Fig. 6) likely that AtPCP-Ba, AtPCP-Bb, AtPCP-Bc and AtPCP-Bd are despite the fact that the pcp-ba mutant had little discernible orthologous to the BoPCPs. effect on pollen hydration in isolation. Hydration rates for the second and third 10-min periods were not significantly different to WT for any of the mutant lines although the final extent of Pollen hydration is impaired in pcp-b mutants hydration was clearly lower for all lines with the exception of In order to investigate the effects of AtPCP-B gene mutations on pcp-ba. early stages of the pollen–stigma interaction in Arabidopsis In order to determine if the pollen hydration defect resulted thaliana, in vivo pollen hydration assays were carried out by polli- from an inherent inability of mutant pollen grains to absorb nating stigmas of the male sterile A9-barnase WT line with water rather than a defect in the pollen–stigma interaction, an pollen grains derived from WT and pcp-b plants. Four T-DNA in vitro pollen hydration assay was carried out. Using a humid insertion lines were identified as mutant alleles of AtPCP-Ba, b, c chamber (providing 100% relative humidity) hydration of WT and d (Fig. S1) with PCP-B transcripts being undetectable in and pcp-ba/b/c pollen was compared over a 30-min period and anthers for pcp-ba-1, pcp-bb-1 and pcp-bc-1. PCP-Bd expression their hydration characteristics were found to be indistinguishable was found to be substantially downregulated, where the T-DNA (Fig. S7). Interestingly, comparison of WT pollen hydration on insertion was located in the promoter region of the gene (Figs S1, stigmas and in the humid chamber indicated that pollen hydrates S2). No obvious vegetative or reproductive morphological abnor- more rapidly, and attains a greater degree of hydration on stigmas malities were observed in any of the lines. Each individual pcp-b (Figs 5b–d, S7b–d). These data demonstrate that the stigma is line, a double mutant (pcp-bb/c) and a triple mutant (pcp-ba/b/ essential for rapid pollen hydration, and importantly, the absence 2016 The Authors New Phytologist (2017) 213: 764–777 New Phytologist  2016 New Phytologist Trust www.newphytologist.com New 770 Research Phytologist Fig. 3 Phylogenetic analysis of 46 PCP-B class genes in Arabidopsis and Brassica. The maximum-likelihood tree was constructed using nucleotide sequences of the predicted mature protein coding regions. Bootstrap values (1000 resamplings) > 50% are shown for interior branches. Branch length is scaled to the bar defined as 0.1 nucleotide substitutions per codon. The clades indicated by red and blue bars include PCP-Bs and ESFs, respectively. Genes are abbreviated as: AthB, Arabidopsis thaliana PCP-B-like; AlyB, Arabidopsis lyrata PCP-B-like; BoB, Brassica oleracea PCP-B-like; BrapaB, Brassica rapa PCP-B-like. Gene loci or scaffolds are shown adjacent to gene abbreviations. of PCP-B protein from the pollen coat does not impair the bio- which pollen could be washed off the stigma. Significantly higher physical ability of pollen to acquire water. numbers of pollen grains from the pcp-ba/b/c triple mutant (77%) were washed off WT stigmas compared with WT pollen (66%) 30 min post-pollination (Fig. 7a). However, an EM ultra- Pollen adhesion is reduced in pcp-b mutants structural analysis of the pollen from all pcp-b mutant lines In order to further characterize the phenotype of PCP-B mutants, revealed no discernible abnormalities in the characteristics of the a pollen adhesion assay was devised which tested the ease with pollen grain or pollen coat (Figs 7b–g, S8, S9). New Phytologist (2017) 213: 764–777  2016 The Authors www.newphytologist.com New Phytologist  2016 New Phytologist Trust New Phytologist Research 771 Fig. 4 Brassica and Arabidopsis PCP-Bs are gametophytically expressed late in pollen development. (a) mRNA gel blot analysis of Brassica oleracea PCP-B1 and PCP-B2 expression in leaves and reproductive tissue. Anthers from 9 to 11 mm buds have a fully degenerated tapetum and pollen is trinucleate. The arrows indicate the size of the transcript in base pairs. (b) Reverse transcription polymerase chain reaction (RT-PCR) expression analysis of AtPCP-B and AtPCP-B-like genes in Arabidopsis leaves, roots, stigmas and anthers (derived from stage 12 buds). GapC – cDNA input control for RT-PCR. (c) AtPCP-B gene expression in flower buds through development (stages 10–12). S, small (< 1 mm; uninucleate microspores); M, medium (c.1–1.5 mm; binucleate pollen); L, large (> 1.5 mm; unopened buds, trinucleate mature pollen). Arabidopsis flower bud stages are as defined by Smyth et al. (1990). (d) RNA–RNA in situ hybridization study of AtPCP-Bb expression in Arabidopsis thaliana anthers. Left panel: transverse anther section treated with an antisense (+ve) AtPCP-Bb DIG-labelled riboprobe, a clear signal (arrow) is observed within the majority of pollen grains. Right panel: longitudinal anther section treated with a control ‘sense’ (ve) riboprobe with no signal being detectable in pollen grains. Bars, 20 lm. resonance (NMR) and this made it possible to generate 3D struc- Initiation of pollen tube growth is delayed in pcp-b mutants tural predictions for the AtPCP-Bs by homologous alignment In order to determine if the early stages of pollen tube growth (Figs 9, S11, S12). All resulting models were statistically well- were affected by the delay in pollen hydration observed for pcp-b supported (Table S4). mutants, in vivo pollen tube lengths were estimated. After 2 h Based on the predicted 3D structure AtPCP-Bs likely share the WT pollen produced significantly longer tubes than pollen same intramolecular disulphide bonding pattern as ESF1.3 derived from all pcp-b mutant lines (Figs 8a, S10) with this effect (Figs 9a,b, S12) with all possessing a conserved cysteine-stabilized being largely maintained 4 h post-pollination (Fig. 8b). This motif consisting of an a-helix and three-stranded antiparallel result is consistent with data collected from the pollen hydration beta-sheet. In addition all AtPCP-Bs have a conserved aromatic assay where most mutants displayed impairment to the degree residue (Tyr-45 in AtPCP-Bc) that is also present in ESF1.3 and rate of hydration which in turn would likely cause a delay in (Trp-48) and other PCP-B-like proteins in Arabidopsis thaliana pollen tube emergence. Despite the observed post-pollination (Fig. 1). The surface electrostatic potential distribution for defects amongst the pcp-b mutants, there was no significant dif- AtPCP-Bc (Fig. 9c) is characterized by both positively and nega- ference in seed set following self-pollinations compared with WT tively charged domains with a prominent positively charged plants (Table S3), indicating that PCP-B protein function is extended loop held between Cys-36 and Cys-44 which lies in likely restricted to very early post-pollination events. close proximity to the conserved aromatic residue (Tyr-45). These features are broadly shared between all four AtPCP-Bs (Figs 9, S12). Structural prediction of AtPCP-Bs Our analyses have revealed the presence of PCP-B-like proteins Discussion in a wide range of angiosperm lineages with all sequences sharing the characteristic motif of eight cysteine residues in the mature Compatible pollination is a highly regulated process that requires polypeptide (Fig. 1). Costa et al. (2014) recently resolved the a suite of complementary pollen and pistil factors that act from structure of the PCP-B-like protein ESF1.3 by nuclear magnetic the moment of pollen contact through to successful fusion of 2016 The Authors New Phytologist (2017) 213: 764–777 New Phytologist  2016 New Phytologist Trust www.newphytologist.com New 772 Research Phytologist (a) (b) (c) (d) (f) (g) (e) Fig. 5 Mutations in Arabidopsis thaliana PCP-B genes result in altered pollen hydration profiles. Box plots depict the 25% quartile, median, 75% quartile and full range of values. (a–d) Pollen diameter distributions at 0, 10, 20 and 30 min following pollination; 1.8 pixels = 1 lm. *, P < 0.05; **, P < 0.001; ***, P < 0.0005 (Welsh’s t-test). Sample sizes: Col-0 (wild-type), 16; pcp-ba, pcp-bb, pcp-bc, pcp-bd, 15; pcp-ba/b, 16; pcp-ba/b/c, 15. (e–g) Pollen hydration is represented as percentage change in pollen diameter relative to diameter at 0 min (pollen diameter at initial contact with stigma) – distributions shown are for 10, 20 and 30 min post-pollination. *, P < 0.05; **, P < 0.005; ***, P < 0.000005 (Welsh’s t-test). Sample sizes: Col-0 (wild-type), 16; pcp-ba, pcp- bb, pcp-bc, pcp-bd, 15; pcp-ba/b, 16; pcp-ba/b/c, 15. gametes (Edlund et al., 2004; Hiscock & Allen, 2008). One of Our phylogenetic analysis of 46 gene sequences encoding PCP- the earliest events in the establishment of compatibility amongst B-like proteins in Arabidopsis and Brassica (Fig. 3) reveals an evo- species that possess dry stigmas, such as Arabidopsis thaliana,is lutionary history featuring frequent gene duplication events and the ability for pollen to gain access to stigmatic water (Elleman rapid sequence divergence around their conserved cysteine motif. et al., 1992; Safavian & Goring, 2013). This reproductive ‘check- These features are typical for gene families associated with repro- point’ requires activation of a basal stigmatic compatibility sys- duction and importantly can contribute to reproductive isolation tem by factor(s) that must be derived from pollen (Safavian & and speciation (Swanson & Vacquier, 2002; Clark et al., 2006; Goring, 2013). Our investigations reported here into small pol- Cui et al., 2015). Interestingly the PCP-Bs investigated here were len coating-borne cysteine-rich proteins point to an important found to be closely related to the ESF1s that encode embryo role for the PCP-Bs in these earliest stages of pollen–pistil com- developmental regulators (Costa et al., 2014) and these sequences patibility in Arabidopsis, as plants carrying mutations in PCP-B clustered in distinct phylogenetic clades, underlining their func- genes are impaired in pollen hydration. Importantly, PCP-Bs tional specialization (Figs 2, 3). Some AtPCP-B family members bear many hallmarks of intercellular signalling ligands and thus were more similar to genes in the closely related species A. lyrata are likely to be a central component of a pollen molecular ‘signa- suggesting that these have retained a specific function that pre- ture’ that defines compatibility. dates speciation. For example, AtPCP-Ba and AtPCP-Bc are PCP-Bs are structurally related proteins that have an ancient more closely related to the A. lyrata B4 and B1, respectively, than evolutionary origin, being widespread amongst angiosperm taxa. to other AtPCP-Bs (Fig. 3). Whereas putative Brassica New Phytologist (2017) 213: 764–777  2016 The Authors www.newphytologist.com New Phytologist  2016 New Phytologist Trust New Phytologist Research 773 (a) Fig. 6 Rate of pollen hydration is severely decreased in Arabidopsis thaliana pcp-b triple mutants. (a) Curves of mean pollen (b) hydration (% hydration is percentage change in pollen diameter). The vertical lines demark each 10-min period over which slopes were calculated. (b) Rate of change in pollen diameter during the first three 10-min periods of pollination. Average slopes  confidence intervals were produced by linear regression. *, P < 0.05; **, P < 0.005; ***, P < 0.001; ****, P < 0.0005 (Welsh’s t-test). (a) (b) (c) (d) (g) (e) (f) Fig. 7 Pollen morphology is unaffected in Arabidopsis thaliana pcp-ba/b/c pollen grains and pollen–stigma adhesion is weakened. (a) The mean percentage of wild-type (Col-0) and pcp-ba/b/c triple mutant pollen washed off WT stigmas in an adhesion assay 30 min post-pollination. Error bars represent the confidence interval. Sample sizes: WT stigmas, 64; triple mutant, 50. *, P < 0.001 (Welsh’s t-test). (b–g) Scanning electron microscopic (SEM) analysis of exine and pollen coat morphology in (b–d) WT and (e–g) pcp-ba/b/c triple mutant plants. Bars: (b, e), 10 lm; (c, d, f, g) 1 lm. orthologues were found in discrete clades more distant from the class proteins, are small secreted proteins that are encoded by a Arabidopsis PCP-Bs and could point to divergence of recognition rapidly evolving gene family. It is thus tempting to speculate that factors required for compatibility. Species-specific functionaliza- PCP-Bs not only regulate aspects of compatibility but may also tion of plant reproductive proteins that contribute to reproduc- contribute to reproductive barriers within the Brassicaceae. tive isolation have been documented for the pollen tube Our mutational study revealed that absence of AtPCP-Bs from attractant LURE proteins secreted by egg-accompanying synergid the pollen coat caused a series of interlinked phenotypes resulting cells of the embryo sac (Takeuchi & Higashiyama, 2012). from a primary defect in pollen hydration. We ascertained that Heterologous expression of an A. thaliana LURE protein in the pcp-b hydration defect was not caused by gross morphological Torenia fournieri synergid cells enabled A. thaliana pollen tubes perturbation of the pollen coat (Figs 7b–g, S8, S9) and that it to successfully locate and enter the embryo sac of this species. was only evident during the pollen–stigma interaction, as triple LURES are defensin-like CRPs and, in common with the PCP-B mutant pcp-ba/b/c pollen hydrated normally in a humid 2016 The Authors New Phytologist (2017) 213: 764–777 New Phytologist  2016 New Phytologist Trust www.newphytologist.com New 774 Research Phytologist (a) (b) Fig. 8 Extent of in vivo pollen tube growth is reduced for pcp-b mutants. Distance (in pixels) of pollen tube growth for wild-type and pcp-b mutants (a) 2 h and (b) 4 h post- pollination. Pollen was applied to stigmas of the Arabidopsis thaliana Col-0 A9-barnase male sterile line. Error bars represent  SD. Sample sizes: 4. *, P < 0.001; **, P < 0.0001; ***, P < 0.00001; ****, P < 0.000001 (Welsh’s t-test); 1 pixel = 0.625 lm. (a) Fig. 9 AtPCP-Bc structure prediction by SWISS-MODEL. (a) Amino acid sequence of PCP-Bc. Connection arrows, disulphide bonds; blue arrows, beta strands; red bar, (b) (c) alpha helix. (b) Cartoon model of predicted structure of AtPCP-Bc with indicated disulphide bonds and Tyrosine residue. (c) Distribution of electrostatic potential on AtPCP-Bc surface based on the predicted structure. Blue, positive; red, negative; white, hydrophobic residues. chamber (Fig. S7). Hydration rate, the degree of hydration and together, probably through different pollen tube receptors, to resulting pollen tube lengths were all found to be largely impaired ensure appropriate pollen tube guidance to the embryo sac amongst pcp-b single and combined mutants (Figs 5, 6, 8). We (Okuda et al., 2009; Takeuchi & Higashiyama, 2012, 2016; consider that the shorter tubes observed in pistils for pcp-b Wang et al., 2016). The severity of the triple pcp-ba/b/c mutant mutants are most likely the result of delayed pollen tube emer- reduced the degree and rate of pollen hydration to almost one gence rather than slower tube extension, because tube emergence third that of WT and due to the close genetic linkage of PCP-Bd is largely dependent on the degree of pollen hydration and pollen to PCP-Bc (< 10 kb) we were unable to recover and test the effect turgor (Taylor & Hepler, 1997). This inference was supported of a pcp-b quadruple mutant. Thus, it remains to be determined by the observation that triple mutant pollen adhered significantly if a complete hydration block could be achieved by abolishing all less well to stigmatic papillae 30 min post-pollination (Fig. 7a) – PCP-B proteins from the pollen coat. we observed that a significant component of this effect was due to Given the structural features of AtPCP-Bs and their homology WT pollen tubes initiating stigmatic penetration ahead of pcp-b to the ESF1 family of secreted developmental regulators we pro- pollen, thus anchoring them on the stigma, whereas substantially pose that PCP-Bs act as ligands to either directly or indirectly fewer mutant pollen had initiated germination (L. Wang & J. activate stigmatic targets that mediate transfer of water through Doughty, unpublished). the papilla plasma membrane. A substantial body of evidence Comparison of the severity of the hydration defects between now points to targeted stigmatic secretion as being a central fea- single and combined mutants revealed evidence of complex com- ture of compatible pollination in both A. thaliana and Brassica binatorial effects of PCP-Bs in the pollen–stigma interaction. and that the exocyst protein complex is essential to this process Out of the single mutant lines, pcp-bc presented the most statisti- (Samuel et al., 2009; Safavian & Goring, 2013; Safavian et al., cally robust hydration defect over the first 10-min period follow- 2014, 2015). The exocyst mediates tethering of secretory vesicles ing pollination, with pcp-bb having an almost identical hydration to target membranes (Zarsky et al., 2013) and stigmas from Ara- profile (Figs 5e, 6). Interestingly, the phenotype of the double bidopsis that carry mutations in Exo70A1, a key linker compo- pcp-bb/c mutant was not additive, but when combined with the nent of the exocyst-tethering machinery, have severe pollen pcp-ba mutant – which singly had no significant phenotype – hydration defects. Targeted secretion likely delivers factors to the pollen hydration was reduced dramatically (Figs 5e–g, 6). The plasma membrane adjacent to compatible pollen that mediate contrasting combinatorial effects of these mutants suggests that water transport. For instance, aquaporins, membrane-localized PCP-Ba may be acting as a ligand to activate a different stigmatic water transport proteins (Johanson et al., 2001; Quigley et al., hydration effector target to that of PCP-Bb and PCP-Bc, or that 2002; Maurel et al., 2008), could be deposited at the interface PCP-Ba acts to enhance activation of a putative stigmatic target with compatible pollen. A specific role for pollen coat factors working synergistically with other PCP-Bs. Similar complexity triggering such a response is supported by the observation that has been reported for synergid LURE proteins in T. fournieri and isolated B. oleracea pollen coat appears to evoke a secretory Arabidopsis where it seems likely that multiple LUREs work response by stigmatic papillae (Elleman & Dickinson, 1996). New Phytologist (2017) 213: 764–777  2016 The Authors www.newphytologist.com New Phytologist  2016 New Phytologist Trust New Phytologist Research 775 Cao L, Bandelac G, Volgina A, Korostoff J, DiRienzo JM. 2008. Role of Homology modelling of AtPCP-Bs provided strong support aromatic amino acids in receptor binding activity and subunit assembly of the for overall structural similarity with ESF1.3 (Figs 9b, S12). As cytolethal distending toxin of Aggregatibacter actinomycetemcomitans. Infection has been determined for ESF1.3 and other plant regulatory pep- and Immunity 76: 2812–2821. tides it is likely that the disulphide-stabilized cysteine motif is Chae K, Kieslich CA, Morikis D, Kim SC, Lord EM. 2009. A gain-of-function crucial for protein function of PCP-Bs (Ohki et al., 2011; Costa mutation of Arabidopsis lipid transfer protein 5 disturbs pollen tube tip growth and fertilization. Plant Cell 21: 3902–3914. et al., 2014). Intriguingly the AtPCP-Bs shared a functionally Chapman LA, Goring DR. 2010. Pollen–pistil interactions regulating essential aromatic residue with ESF1.3. Aromatic residues are a successful fertilization in the Brassicaceae. Journal of Experimental Botany 61: conserved feature of many plant regulatory peptides (Cao et al., 1987–1999. 2008; Okuda et al., 2009; Sugano et al., 2010; Costa et al., 2012; Clark NL, Aagaard JE, Swanson WJ. 2006. Evolution of reproductive proteins Sprunck et al., 2012) and are likely to be important in protein– from animals and plants. Reproduction 131:11–22. Costa LM, Marshall E, Tesfaye M, Silverstein KA, Mori M, Umetsu Y, protein interactions (Simpson et al., 2000). Otterbach SL, Papareddy R, Dickinson HG, Boutiller K et al. 2014. Central In conclusion, this study shows that AtPCP-Bs are important cell-derived peptides regulate early embryo patterning in flowering plants. mediators of pollen hydration, a key early ‘checkpoint’ of pollen– Science 344: 168–172. stigma compatibility. Their close evolutionary relationship to the Costa LM, Yuan J, Rouster J, Paul W, Dickinson H, Gutierrez-Marcos JF. ESF1 family of embryo developmental regulators, and their 2012. Maternal control of nutrient allocation in plant seeds by genomic imprinting. Current Biology 22: 160–165. broad similarity to other CRP regulatory proteins strongly sug- Cui X, Lv Y, Chen ML, Nikoloski Z, Twell D, Zhang DB. 2015. Young genes gest that they act through interaction with as yet unknown stig- out of the male: an insight from evolutionary age analysis of the pollen matic targets to activate the basal compatibility system. In transcriptome. Molecular Plant 8: 935–945. addition, PCP-B maintenance and diversity within Arabidopsis Dickinson HG. 1995. Dry stigmas, water and self-incompatibility in Brassica. and the Brassicaceae suggest that these proteins have the potential Sexual Plant Reproduction 8:1–10. Dickinson HG, Elleman CJ, Doughty J. 2000. Pollen coatings – chimaeric to contribute to prezygotic hybridization barriers. genetics and new functions. Sexual Plant Reproduction 12: 302–309. Doughty J, Dixon S, Hiscock SJ, Willis AC, Parkin IAP, Dickinson HG. 1998. PCP-A1, a defensin-like Brassica pollen coat protein that binds the S locus Acknowledgements glycoprotein, is the product of gametophytic gene expression. Plant Cell 10: We thank Susan Crennell for assistance with protein structural 1333–1347. Doughty J, Hedderson F, Mccubbin A, Dickinson H. 1993. Interaction between predictions, Ursula Potter for support with electron microscopy, a coating-borne peptide of the Brassica pollen grain and stigmatic S (self- Andrew James for plant maintenance, Jose Gutierrez-Marcos for incompatibility)-locus-specific glycoproteins. Proceedings of the National the kind gift of pAt5g61605:GUS and pAt2g16505:GUS reporter Academy of Sciences, USA 90: 467–471. lines, Tony Willis for protein sequencing, Sue Dixon for techni- Doughty J, Wong HY, Dickinson HG. 2000. Cysteine-rich pollen coat proteins cal assistance, Hugh Dickinson for insightful discussions, and Ed (PCPs) and their interactions with stigmatic S (incompatibility) and S-related proteins in Brassica: putative roles in SI and pollination. Annals of Botany 85: Nottingham, Hanis Jasmine Zaini and Ken Hayes for valuable 161–169. contributions during their undergraduate projects. This research Edgar RC. 2004. MUSCLE: multiple sequence alignment with high accuracy and was funded by a University of Bath postgraduate scholarship to high throughput. Nucleic Acids Research 32: 1792–1797. L.W. and a Nuffield Foundation award to J.D. Edlund AF, Swanson R, Preuss D. 2004. Pollen and stigma structure and function: the role of diversity in pollination. Plant Cell 16: S84–S97. Elleman CJ, Dickinson HG. 1986. Pollen–stigma interactions in Brassica. IV. Author contributions Structural reorganization in the pollen grains during hydration. Journal of Cell Science 80: 141–157. J.D. and L.W. planned and designed the research and wrote the Elleman CJ, Dickinson HG. 1990. The role of the exine coating in pollen– manuscript; L.W. was involved with all aspects of the research stigma interactions in Brassica oleracea L. New Phytologist 114: 511–518. with L.A.C. and R.J.E. contributing to expression analysis and Elleman CJ, Dickinson HG. 1996. Identification of pollen components regulating pollination-specific responses in the stigmatic papillae of Brassica C.C.P. to the SEM work; R.J.S. and B.Q. assisted with experi- oleracea. New Phytologist 133: 197–205. mental design and critical assessment of the manuscript. Elleman CJ, Franklin-Tong V, Dickinson HG. 1992. Pollination in species with dry stigmas: the nature of the early stigmatic response and the pathway taken by pollen tubes. New Phytologist 121: 413–424. References Fiebig A, Mayfield JA, Miley NL, Chau S, Fischer RL, Preuss D. 2000. Allen AM, Thorogood CJ, Hegarty MJ, Lexer C, Hiscock SJ. 2011. Pollen–pistil Alterations in CER6, a gene identical to CUT1, differentially affect long-chain interactions and self-incompatibility in the Asteraceae: new insights from lipid content on the surface of pollen and stems. Plant Cell 12: 2001–2008. studies of Senecio squalidus (Oxford ragwort). Annals of Botany 108: 687–698. Guex N, Peitsch MC, Schwede T. 2009. Automated comparative protein Alonso JM, Stepanova AN, Leisse TJ, Kim CJ, Chen HM, Shinn P, Stevenson structure modeling with SWISS-MODEL and Swiss-PdbViewer: a historical DK, Zimmerman J, Barajas P, Cheuk R et al. 2003. Genome-wide insertional perspective. Electrophoresis 30: S162–S173. mutagenesis of Arabidopsis thaliana. Science 301: 653–657. Hernandez-Pinzon I, Ross JH, Barnes KA, Damant AP, Murphy DJ. 1999. Arnold K, Bordoli L, Kopp J, Schwede T. 2006. The SWISS-MODEL Composition and role of tapetal lipid bodies in the biogenesis of the pollen coat workspace: a web-based environment for protein structure homology of Brassica napus. Planta 208: 588–598. modelling. Bioinformatics 22: 195–201. Heslop-Harrison Y, Shivanna KR. 1977. The receptive surface of the angiosperm Biasini M, Bienert S, Waterhouse A, Arnold K, Studer G, Schmidt T, Kiefer F, stigma. Annals of Botany 41: 1233–1258. Cassarino TG, Bertoni M, Bordoli L et al. 2014. SWISS-MODEL: modelling Hiroi K, Song M, Sakazono S, Osaka M, Masuko-Suzuki H, Matsuda T, Suzuki protein tertiary and quaternary structure using evolutionary information. G, Suwabe K, Watanabe M. 2013. Time-lapse imaging of self- and cross- Nucleic Acids Research 42: W252–W258. pollination in Brassica rapa. Annals of Botany 112: 115–122. 2016 The Authors New Phytologist (2017) 213: 764–777 New Phytologist  2016 New Phytologist Trust www.newphytologist.com New 776 Research Phytologist Hiscock SJ, Allen AM. 2008. Diverse cell signalling pathways regulate pollen– stigmatic defect for accepting compatible pollen. Plant Reproduction 27: stigma interactions: the search for consensus. New Phytologist 179: 286–317. 121–127. Hiscock SJ, Doughty J, Willis AC, Dickinson HG. 1995. A 7-kDa pollen Safavian D, Zayed Y, Indriolo E, Chapman L, Ahmed A, Goring DR. 2015. coating-borne peptide from Brassica napus interacts with S-locus glycoprotein RNA silencing of exocyst genes in the stigma impairs the acceptance of and S-locus-related glycoprotein. Planta 196: 367–374. compatible pollen in Arabidopsis. Plant Physiology 169: 2526–2538. Hulskamp M, Kopczak SD, Horejsi TF, Kihl BK, Pruitt RE. 1995. Samuel MA, Chong YT, Haasen KE, Aldea-Brydges MG, Stone SL, Goring Identification of genes required for pollen–stigma recognition in Arabidopsis DR. 2009. Cellular pathways regulating responses to compatible and self- thaliana. Plant Journal 8: 703–714. incompatible pollen in Brassica and Arabidopsis stigmas intersect at Johanson U, Karlsson M, Johansson I, Gustavsson S, Sjovall S, Fraysse L, Weig Exo70A1, a putative component of the exocyst complex. Plant Cell 21: AR, Kjellbom P. 2001. The complete set of genes encoding major intrinsic 2655–2671. proteins in Arabidopsis provides a framework for a new nomenclature for Schneider CA, Rasband WS, Eliceiri KW. 2012. NIH Image to ImageJ: 25 years major intrinsic proteins in plants. Plant Physiology 126: 1358–1369. of image analysis. Nature Methods 9: 671–675. Kho YO, Baer J. 1968. Observing pollen tubes by means of fluorescence. Schopfer CR, Nasrallah ME, Nasrallah JB. 1999. The male determinant of self- Euphytica 17: 298–302. incompatibility in Brassica. Science 286: 1697–1700. Kiefer F, Arnold K, Kuenzli M, Bordoli L, Schwede T. 2009. The SWISS- Shiba H, Takayama S, Iwano M, Shimosato H, Funato M, Nakagawa T, Che MODEL Repository and associated resources. Nucleic Acids Research 37: FS, Suzuki G, Watanabe M, Hinata K et al. 2001. A pollen coat protein, D387–D392. SP11/SCR, determines the pollen S-specificity in the self-incompatibility of Kleinboelting N, Huep G, Kloetgen A, Viehoever P, Weisshaar B. 2012. GABI- Brassica species. Plant Physiology 125: 2095–2103. Kat SimpleSearch: new features of the Arabidopsis thaliana T-DNA mutant Silverstein KAT, Moskal WA Jr, Wu HC, Underwood BA, Graham MA, Town database. Nucleic Acids Research 40: D1211–D1215. CD, VandenBosch KA. 2007. Small cysteine-rich peptides resembling Langdale JA. 1994. In situ hybridization. In: Freeling M, Walbot V, eds. The antimicrobial peptides have been under-predicted in plants. Plant Journal 51: maize handbook. New York, USA: Springer-Verlag, 165–179. 262–280. Letunic I, Bork P. 2007. Interactive Tree Of Life (iTOL): an online tool for Simpson PJ, Xie HF, Bolam DN, Gilbert HJ, Williamson MP. 2000. The phylogenetic tree display and annotation. Bioinformatics 23: 127–128. structural basis for the ligand specificity of family 2 carbohydrate-binding Li YL, Dai XR, Yue X, Gao XQ, Zhang XS. 2014. Identification of small modules. Journal of Biological Chemistry 275: 41137–41142. secreted peptides (SSPs) in maize and expression analysis of partial SSP genes in Smith AG, Eberle CA, Moss NG, Anderson NO, Clasen BM, Hegeman AD. reproductive tissues. Planta 240: 713–728. 2013. The transmitting tissue of Nicotiana tabacum is not essential to pollen Ma J-F, Liu Z-H, Chu C-P, Hu Z-Y, Wang X-L, Zhang XS. 2012. Different tube growth, and its ablation can reverse prezygotic interspecific barriers. Plant regulatory processes control pollen hydration and germination in Arabidopsis. Reproduction 26: 339–350. Sexual Plant Reproduction 25:77–82. Smyth DR, Bowman JL, Meyerowitz EM. 1990. Early flower development in Marshall E, Costa LM, Gutierrez-Marcos J. 2011. Cysteine-rich peptides (CRPs) Arabidopsis. Plant Cell 2: 755–767. mediate diverse aspects of cell–cell communication in plant reproduction and Sprunck S, Rademacher S, Vogler F, Gheyselinck J, Grossniklaus U, development. Journal of Experimental Botany 62: 1677–1686. Dresselhaus T. 2012. Egg cell-secreted EC1 triggers sperm cell activation Maurel C, Verdoucq L, Luu DT, Santoni V. 2008. Plant aquaporins: membrane during double fertilization. Science 338: 1093–1097. channels with multiple integrated functions. Annual Review of Plant Biology 59: Stone SL, Anderson EM, Mullen RT, Goring DR. 2003. ARC1 is an E3 595–624. ubiquitin ligase and promotes the ubiquitination of proteins during the Mayfield JA, Preuss D. 2000. Rapid initiation of Arabidopsis pollination requires rejection of self-incompatible Brassica pollen. Plant Cell 15: 885–898. the oleosin-domain protein GRP17. Nature Cell Biology 2: 128–130. Sugano SS, Shimada T, Imai Y, Okawa K, Tamai A, Mori M, Hara-Nishimura Notredame C, Higgins DG, Heringa J. 2000. T-Coffee: a novel method for fast I. 2010. Stomagen positively regulates stomatal density in Arabidopsis. Nature and accurate multiple sequence alignment. Journal of Molecular Biology 302: 463: 241–244. 205–217. Swanson WJ, Vacquier VD. 2002. The rapid evolution of reproductive proteins. Ohki S, Takeuchi M, Mori M. 2011. The NMR structure of stomagen reveals Nature Reviews Genetics 3: 137–144. the basis of stomatal density regulation by plant peptide hormones. Nature Takasaki T, Hatakeyama K, Suzuki G, Watanabe M, Isogai A, Hinata K. 2000. Communications 2: 512. The S receptor kinase determines self-incompatibility in Brassica stigma. Nature Okuda S, Tsutsui H, Shiina K, Sprunck S, Takeuchi H, Yui R, Kasahara RD, 403: 913–916. Hamamura Y, Mizukami A, Susaki D et al. 2009. Defensin-like polypeptide Takayama S, Shiba H, Iwano M, Asano K, Hara M, Che FS, Watanabe M, LUREs are pollen tube attractants secreted from synergid cells. Nature 458: Hinata K, Isogai A. 2000. Isolation and characterization of pollen coat proteins 357–361. of Brassica campestris that interact with S locus-related glycoprotein 1 involved Paul W, Hodge R, Smartt S, Draper J, Scott R. 1992. The isolation and in pollen-stigma adhesion. Proceedings of the National Academy of Sciences, USA characterisation of the tapetum-specific Arabidopsis thaliana A9 gene. Plant 97: 3765–3770. Molecular Biology 19: 611–622. Takeuchi H, Higashiyama T. 2012. A species-specific cluster of defensin-like Petersen TN, Brunak S, von Heijne G, Nielsen H. 2011. SignalP 4.0: genes encodes diffusible pollen tube attractants in Arabidopsis. PLoS Biology 10: discriminating signal peptides from transmembrane regions. Nature Methods 8: e1001449. 785–786. Takeuchi H, Higashiyama T. 2016. Tip-localized receptors control pollen tube Piffanelli P, Ross JHE, Murphy DJ. 1998. Biogenesis and function of the lipidic growth and LURE sensing in Arabidopsis. Nature 531: 245–248. structures of pollen grains. Sexual Plant Reproduction 11:65–80. Tamura K, Stecher G, Peterson D, Filipski A, Kumar S. 2013. MEGA6: Preuss D, Lemieux B, Yen G, Davis RW. 1993. A conditional sterile mutation molecular evolutionary genetics analysis version 6.0. Molecular Biology and eliminates surface components from Arabidopsis pollen and disrupts cell Evolution 30: 2725–2729. signaling during fertilization. Genes & Development 7: 974–985. Taylor LP, Hepler PK. 1997. Pollen germination and tube growth. Annual Quigley F, Rosenberg JM, Shachar-Hill Y, Bohnert HJ. 2002. From genome to Review of Plant Physiology and Plant Molecular Biology 48: 461–491. function: the Arabidopsis aquaporins. Genome Biology 3: RESEARCH0001. Updegraff EP, Zhao F, Preuss D. 2009. The extracellular lipase EXL4 is required Safavian D, Goring DR. 2013. Secretory activity is rapidly induced in stigmatic for efficient hydration of Arabidopsis pollen. Sexual Plant Reproduction 22: papillae by compatible pollen, but inhibited for self-incompatible pollen in the 197–204. Brassicaceae. PLoS ONE 8: e84286. Villar M, Gaget M, Said C, Knox RB, Dumas C. 1987. Incompatibility in Safavian D, Jamshed M, Sankaranarayanan S, Indriolo E, Samuel MA, Goring Populus: structural and cytochemical characteristics of the receptive stigmas of DR. 2014. High humidity partially rescues the Arabidopsis thaliana exo70A1 Populus alba and Populus nigra. Journal of Cell Science 87: 483–490. New Phytologist (2017) 213: 764–777  2016 The Authors www.newphytologist.com New Phytologist  2016 New Phytologist Trust New Phytologist Research 777 Wang T, Liang L, Xue Y, Jia PF, Chen W, Zhang MX, Wang YC, Li HJ, Yang Fig. S8 SEM analysis of exine layer and pollen coat morphology. WC. 2016. A receptor heteromer mediates the male perception of female attractants in plants. Nature 531: 241–244. Fig. S9 TEM analysis of exine layer and pollen coat morphology. Yost JM, Kay KM. 2009. The evolution of postpollination reproductive isolation in Costus. Sexual Plant Reproduction 22: 247–255. Fig. S10 Comparison of pollen tube growth for wild-type and Zarsky V, Kulich I, Fendrych M, Pecenkova T. 2013. Exocyst complexes multiple functions in plant cells secretory pathways. Current Opinion in Plant pcp-b triple mutant plants. Biology 16: 726–733. Fig. S11 Homologous alignments of ESF1.3 and AtPCP-Bs for protein structure predictions. Supporting Information Additional Supporting Information may be found online in the Fig. S12 Predicted protein structure homology models of Supporting Information tab for this article: AtPCP-Ba, b and d. Fig. S1 Locations of T-DNA insertions. Table S1 PCR primers used in this study Fig. S2 RT-PCR analysis results of stage 12 anthers in pcp-b Table S2 Numbers and abbreviations of predicted PCP-B-like mutants. proteins in species and families Fig. S3 N-terminal sequencing of two PCP-B proteins purified Table S3 Average seed count values of Arabidopsis wild-type and from Brassica oleracea pollen coat. pcp-b mutants Fig. S4 Phylogeny of 282 predicted PCP-B-like protein Table S4 Statistics for AtPCP-B protein structural predictions sequences. Methods S1 Histochemical staining for b-glucuronidase activity. Fig. S5 RNA–RNA in situ hybridization study of AtPCP-Bc expression in Arabidopsis thaliana anthers. Please note: Wiley Blackwell are not responsible for the content or functionality of any Supporting Information supplied by the Fig. S6 Histochemical staining for GUS activity driven by authors. Any queries (other than missing material) should be AtPCP-Ba and AtPCP-Bd promoters in Arabidopsis tissues. directed to the New Phytologist Central Office. Fig. S7 Pollen hydration profiles of wild-type and pcp-b triple mutant grains in a humid chamber. New Phytologist is an electronic (online-only) journal owned by the New Phytologist Trust, a not-for-profit organization dedicated to the promotion of plant science, facilitating projects from symposia to free access for our Tansley reviews. Regular papers, Letters, Research reviews, Rapid reports and both Modelling/Theory and Methods papers are encouraged. We are committed to rapid processing, from online submission through to publication ‘as ready’ via Early View – our average time to decision is <28 days. There are no page or colour charges and a PDF version will be provided for each article. The journal is available online at Wiley Online Library. Visit www.newphytologist.com to search the articles and register for table of contents email alerts. If you have any questions, do get in touch with Central Office (np-centraloffice@lancaster.ac.uk) or, if it is more convenient, our USA Office (np-usaoffice@lancaster.ac.uk) For submission instructions, subscription and all the latest information visit www.newphytologist.com 2016 The Authors New Phytologist (2017) 213: 764–777 New Phytologist  2016 New Phytologist Trust www.newphytologist.com

Journal

New PhytologistWiley

Published: Jan 1, 2017

Keywords: ; ; ; ; ; ;

There are no references for this article.