Cell–cell communications and molecular mechanisms in plant sexual reproduction
- 296 Downloads
Sexual reproduction is achieved by precise interactions between male and female reproductive organs. In plant fertilization, sperm cells are carried to ovules by pollen tubes. Signals from the pistil are involved in elongation and control of the direction of the pollen tube. Genetic, reverse genetic, and cell biological analyses using model plants have identified various factors related to the regulation of pollen tube growth and guidance. In this review, I summarize the mechanisms and molecules controlling pollen tube growth to the ovule, micropylar guidance, reception of the guidance signal in the pollen tube, rupture of the pollen tube to release sperm cells, and cessation of the tube guidance signal. I also briefly introduce various techniques used to analyze pollen tube guidance in vitro.
KeywordsPlant sexual reproduction Pollen tube guidance Synergid cell Gamete fusion Lab-on-a-chip
Angiosperms have evolved their own sexual reproduction mechanisms, in which female gametophytes are formed within ovules. In general gametophyte formation (Polygonum type), a megaspore is generated via meiosis of the megaspore mother cell. In the megaspore, three cycles of mitosis are induced, and eight nuclei are formed. Among these, three nuclei on the carazal side differentiate into antipodal cells, and three nuclei on the micropylar side differentiate into two synergid cells and one egg cell. The two polar nuclei fuse together to form the nucleus of the central cell. The male pollen grain comprises two sperm cells contained in a vegetative cell. The pollen tube germinates from the pollen grain and delivers two sperm cells to the ovule. Pollen tube elongation occurs very rapidly, reaching a rate of 1 cm/h in some plant species. The pollen tube enters the ovule from the micropyle and then ruptures to release the sperm cells. Sperm cells fertilize egg or central cells, forming an embryo and endosperm, respectively. This fertilization mechanism, unique to angiosperms, is known as double fertilization.
Since pollen tubes were discovered approximately 150 years ago, much research has been conducted to determine how pollen tubes reach the ovules for fertilization. Signals are exchanged between male and female cells during extension of the pollen tube (Higashiyama and Takeuchi 2015). These signals guide the pollen tube to enable proper fertilization, reject multiple pollen tubes, and prevent fertilization from unwanted (extraspecific) pollen.
Many small secreted peptides have been identified as signaling substances so far (Kanaoka and Higashiyama 2015). Among them, one of a key family of proteins is called Cysteine-Rich Polypeptide (CRP). In general, CRP has four or more cysteine residues in the molecule, and disulfide bonds are formed between the cysteine residues. There are many genes encoding CRP in the genome of plants. Up to now, 800 CRP genes have been reported in Arabidopsis thaliana and 600 in rice (Silverstein et al. 2007). In addition, more than 600 receptor-like kinase genes, which are thought to be involved in reception of CRPs, have been reported in the genome of A. thaliana (Shiu and Bleecker 2001). Furthermore, recently it has been reported that structures such as sugar chains, which are added to proteins by posttranslational modification, act as signaling molecules (Pereira et al. 2016a).
This review introduces recent topics in pollen tube guidance research (Fig. 1). I will describe the interaction between the pollen tube and female stylar tissue, guidance signals from ovules and their reception by pollen tubes, and mechanisms involved in pollen tube rupture, gamete fusion, and cessation of the pollen tube guidance signal. I will also explore microdevices used for in vitro analysis of pollen tube behavior. For topics that could not be precisely introduced here are found in recent reviews (Hafidh et al. 2016a; Higashiyama and Takeuchi 2015; Higashiyama and Yang 2017; Kanaoka and Higashiyama 2015; Maruyama and Higashiyama 2016; Pereira et al. 2016a; Qu et al. 2015; Zhou et al. 2017).
Preovular pollen tube guidance
The first interaction between male and female cells during fertilization is pollination and germination of pollen tubes. In the self-incompatibility response of Brassicaceae, SP11/SCR and SRK act as a ligand–receptor pair (Takayama et al. 2001). SP11/SCR is a CRP with eight cysteine residues and belongs to difensin-like superfamily. SP11/SCR is rich in sequence diversity among alleles, and its diversity plays a critical role in the conformational difference and recognition with the receptor (Mishima et al. 2003). LePRK1 and LePRK2, first identified in tomato, are pollen-specific receptor kinases involved in pollen tube growth (Gui et al. 2014). The extracellularly secreted protein LAT52, specifically expressed in pollen grains, LeSHY, a leucine-rich protein, and LeSTIG1, expressed specifically in the pistil, have been reported to bind to LePRKs (Tang et al. 2004). LAT52 binds to LePRK2 only before pollen tube germination. Binding between LAT52 and LePRK2 is inhibited by LeSTIG1, a protein containing 16 cysteine residues. LeSTIG1 interacts with LePRK2 or phosphatidylinositol 3-phosphate (PI(3)P) at different sites. When LeSTIG1 is applied to a pollen tube on medium, elongation of the pollen tube is promoted, suggesting interaction of these proteins and the importance of PI(3)P for pollen tube growth (Huang et al. 2014).
Arabinogaractan proteins (AGPs) are a family of hydroxyproline-rich glycoproteins that bind to the cell membrane; they possess a sugar chain with a very large, complex structure. In tobacco, the TTS protein, which is expressed in the style, promotes pollen tube growth (Cheung et al. 1995; Wu et al. 2000). In the evolutionarily old angiosperm magnolia (Magnolia virginiana), two types of AGPs are expressed once the stigma is able to accept pollen (Losada et al. 2014). In A. thaliana, PECTIC ARABINOGALACTAN SYNTHESIS-RELATED (PAGR) is involved in the biosynthesis of RG-1 arabinogalactan, and inhibition of the function of this protein suppresses elongation of the pollen tube (Stonebloom et al. 2016).
In A. thaliana, the style is filled with cells. When the pollen tube grows through the style, programmed cell death occurs in the style, the connections between cells weaken, and a space is created where the pollen tube elongates. This programmed cell death process involves the NO TRANSMITTING TRACT (NTT), HECATE1 (HEC1), HEC2, HEC3, HALF FILLED (HAF), and SPATULA (SPT) genes, which encode transcription factors, and AtWSCP, which encodes a Kunitz-type protease inhibitor (Alvarez and Smyth 1999; Boex-Fontvieille et al. 2015; Crawford et al. 2007; Crawford and Yanofsky 2011). In rice, mutation of CNGC13, which encodes a cyclic nucleotide-gated channel family protein, results in defects in calcium uptake into the cells of the style and delays in stylar cell death. As a result, pollen tube growth is arrested at the style (Xu et al. 2017). The lily style is hollow, and the pollen tube grows along its surface. Chemocyanin, which exhibits homology with a copper-binding plantacyanin protein, has been isolated as a factor involved in lily pollen tube growth in in vitro experiments (Kim et al. 2003). Knockout of the A. thaliana plantacyanin gene (At2g08250) does not result in a remarkable phenotype; however, abnormalities in pollen tube elongation are observed after overexpression of the gene (Dong et al. 2005). The activity of lily chemocyanin is enhanced by adding Stylar Cysteine-rich Adhesin (SCA), which contains eight cysteines and interacts with pectin on the style surface (Kim et al. 2003; Mollet et al. 2000; Park et al. 2000). In A. thaliana, overexpression of the SCA-like protein LTP5 causes abnormal elongation of the pollen tube and a decrease in fertility (Chae et al. 2009).
Signaling molecules involved in micropylar guidance
When the pollen tube passed through the style and entered the transmitting tract, it changes the growth direction to the ovule. Pollen tube is attracted by the ovule-derived attractant and enters the ovule from the micropyle. In vitro fertilization experiments of Torenia fournieri revealed that the synergid cell secretes pollen tube attraction signals (Higashiyama et al. 2001). The LURE genes encoding Defensin-like CRPs were isolated as genes highly expressed in the synergids from EST analysis. LURE1 and LURE2 proteins synthesized in E. coli attracted pollen tubes in a concentration-dependent manner on the medium (Okuda et al. 2009). Pollen tube attraction by LUREs has species preferentiality. TcLURE1 (TcCRP1) in T. concolor attracts conspecific pollen tubes more frequently than T. fournieri pollen tubes (Kanaoka et al. 2011). Since TfLURE1 and TcLURE1 differ in eight out of the 62 amino acids of the mature peptide, it is considered that these amino acids contribute for species preferentiality among them. AtLUREs also encode Defensin-like CRPs and function as pollen tube attractors in A. thaliana. Ovules of T. fournieri do not attract pollen tubes of A. thaliana, but transgenic T. fournieri ovules expressing AtLURE1.2 do attract them (Takeuchi and Higashiyama 2012). That is, although the origin of the gene used as the attractant is different, the secretory pathway from the synergid cells are conserved. In addition, it is considered that three dimensional structure of the attractants and the binding to the receptor on the pollen tube are different between attractants, which would be the key to the mechanism of rejection of other species’ pollen tubes and of fertilization of conspecific pollen tubes (Higashiyama et al. 2006; Kanaoka and Higashiyama 2015).
Interestingly, TfLUREs alone are not sufficient as attraction cues in T. fournieri (Mizukami et al. 2016). When the usual pollen tube germination medium was used to grow pollen tubes, the pollen tubes were not attracted to ovules or purified TfLUREs. On the other hand, when a conditioned medium containing the ovule extract was used to grow pollen tubes, the pollen tubes were attracted to the ovules or purified TfLUREs. This finding suggests that secretions derived from ovules impart the ability to respond to attractants in pollen tubes, and this factor was named Activation Molecule for Response Capability (AMOR) (Mizukami et al. 2016). Biochemical analysis revealed that AMOR activity is present in the sugar chains of AGPs. Further analysis revealed that removal of the 4-O-methyl-glucronosyl-galactose (4-Me-GlcA-β-(1→6)-Gal) structure at the terminal end of the sugar chain eliminated AMOR activity, and that chemically synthesized 4-Me-GlcA-β-(1→6)-Gal also showed AMOR activity (Jiao et al. 2017; Mizukami et al. 2016). From these results, the substances that has AMOR activity was shown to be a 4-Me-GlcA-β-(1→6)-Gal. Since AGPs with a sugar chain structure similar to that of 4-Me-GlcA-β-(1→6)-Gal are widely distributed on the surface of the placenta, the pollen tube accepts substances with AMOR activity during growth on the placenta, and its properties change so as to response to the ovule-derived attractant. Requirement of AMOR activity for pollen tube attraction has only been reported in T. fournieri currently. Whether AMOR activity is also involved in the attraction of pollen tubes in A. thaliana and other plant species, or whether the structure of AMOR has diverged to generate species specificity are subjects for future research.
myb98, maa3 and siz1 have been reported as mutants involved in micropylar guidance (Kasahara et al. 2005; Ling et al. 2012; Shimizu and Okada 2000). MYB98, MAA3 and SIZ1 encode R2R3 MYB transcription factor (Kasahara et al. 2005), Sen1p-like RNA helicase (Shimizu et al. 2008), SUMO E3 ligase (Ling et al. 2012), respectively. Expression or secretion of AtLURE proteins is greatly reduced in myb98 and maa3 mutants (Takeuchi and Higashiyama 2012). The CENTRAL CELL GUIDANCE (CCG) expressed in the central cell contributes to pollen tube guidance through expression of AtLURE1 (Chen et al. 2007; Takeuchi and Higashiyama 2012). CCG BINDING PROTEIN 1 (CBP1) directly interacts with CCG to regulate the expression of the CRP genes including LURE1 in female gametophyte (Li et al. 2015b). When the function of N-acatylglutamate kinase (NAGK), which catalyzes the second step of arginine biosynthesis, is inhibited, the form of female gametophytes is normal, but pollen tube guidance is abnormal (Huang et al. 2017). These would be upstream factors that control guidance signals from synergid cells via interactions between female gametophytes.
Unlike dicotyledonous plants, CRP has not been reported as an attractant in monocotyledonous plants so far. Instead, in maize, the secreted peptide EGG APPARATUS 1 (ZmEA1) has been reported as a guidance factor to the micropyle (Marton et al. 2005; Marton and Dresselhaus 2010). ZmEA1 binds to pollen tubes of maize, but not to those of Tripsacum dactylides or Nicotiana benthamiana. These findings indicate that the pollen tube guidance factor and its receptor mechanism differ greatly among species (Uebler et al. 2013).
Pollen tube factors for guidance
It has been reported that CRPs bind directly to receptor kinases localized on the cell membrane (Lee et al. 2012; Takayama et al. 2001). Research on receptor-like kinases highly expressed in A. thaliana pollen tubes suggests that POLLEN RECEPTOR KINASE (PRK) family members are candidate receptors for CRPs. An attraction assay of pollen tubes from knockout lines revealed that pollen tube attraction was absent in the PRK6 knockout line. PRK6 is located at the tip of the pollen tube. When LURE was applied to the pollen tube, the localization of PRK6 at the pollen tube tip was biased toward the higher LURE protein concentration before the pollen tube changed its orientation. PRK6 interacts directly with PRK3 receptor kinase expressed in the pollen tube, the Rho family of plant guanine nucleotide exchange factors (ROP-GEFs), and LOST IN POLLEN TUBE GUIDANCE 1 (LIP1) and LIP2, which are intracellularly localized receptor-like kinases (Takeuchi and Higashiyama 2016). The lip1 lip2 double mutant exhibits an abnormality in pollen tube guidance to the micropyle, and its responsiveness to the LURE1.2 protein is low (Liu et al. 2013). PRK6, PRK3, LIP1, and LIP2 may form a receptor complex to accept LURE, and signaling through ROP-GEFs controls the direction of pollen tube growth. Additionally, MALE-DISCOVERER1 (MDIS1) was linked to abnormalities in pollen tube guidance when the function of the kinase domain was lost. Yeast two-hybrid screening identified MDIS2, MDIS1-INTERACTING RECEPTOR LIKE KINASE1 (MIK1), and MIK2 as interacting partners of MDIS1. mdis1 mdis2 and mik1 mik2 double mutants exhibited abnormal pollen tube guidance, and MDIS1, MIK1, and MIK2 bind to LURE1.2, suggesting that these proteins are also receptors for LURE (Wang et al. 2016b). Thus, the MDIS1 and PRK6 complexes are involved in the binding of LUREs.
Several protein involved in pollen tube guidance on the pollen side, apart from the LURE receptors, have been reported. POLLEN DEFECTIVE IN GUIDANCE 1 (POD1) is localized in the endoplasmic reticulum and interacts with CALRETICULIN3 (CRT3). POD1 functions in membrane protein folding (Li et al. 2011). Large numbers of CRT proteins are localized near the transmitting tract and synergid cells, cells surrounding the embryo sac, and pollen tubes. Because CRT regulates calcium storage, it may be involved in the accumulation of calcium used for pollen tube elongation (Wasag et al. 2017). CNGC18 is involved in calcium ion uptake into pollen tubes and pollen tube guidance (Gao et al. 2016). Pollen tubes of the cobra-like 10 (cobl10) mutant can extend to the transmitting tract but cannot orient toward the ovule. COBL10 encodes a glycosylphosphatidylinositol (GPI)-anchored protein localized at the tip of the pollen tube that is involved in the formation of the apical pectin cap (Li et al. 2013). An A. thaliana mutant with deletion of two aspartic protease-encoding genes (A36 and A39) exhibited abnormal pollen tube guidance to the micropyle. A36 and A39 are bound to the cell membrane via a GPI anchor and are co-localized with COBL10 (Gao et al. 2017). The a36 a39 double mutant also exhibits an abnormality in the cell wall component at the tip of the pollen tube; however, the relationship between A36/A39 and COBL10 is unknown. ABNORMAL POLLEN TUBE GUIDANCE 1 (APTG1) encodes a mannosyltransferase localized in the endoplasmic reticulum. In aptg1, pollen tube guidance near the micropyle is abnormal. Since localization of COBL10 is altered in aptg1, it may be that APTG1 is involved in production of the GPI-anchored protein encoded by COBL10 (Dai et al. 2014).
Reception of pollen tube and termination of the attraction signal
Upon entering the ovule, the pollen tube elongates to the appropriate position enabling contact with the synergid cells (Leshem et al. 2013). An increase in calcium concentration is observed in one of the synergid cells (the receptive synergid cell), which subsequently collapses. At the same time, the pollen tube ruptures, and the two sperm cells move to the area between the egg and central cell (Denninger et al. 2014; Hamamura et al. 2014). Ethylene is involved in synergid cell collapse (Volz et al. 2013).
In feronia (fer)/sirene mutant, a pollen tube overgrowth phenotype is observed in the mutant embryo (Huck et al. 2003; Rotman et al. 2003). FER encodes a receptor-like kinase localized in the cell membranes of synergid cells (Escobar-Restrepo et al. 2007). A similar phenotype, in which the pollen tube continues to grow in the ovule, is seen after mutation of NORTIA (NTA), which encodes a plant-specific mildew resistance locus O family protein localized in the filiform apparatus of the synergid cells, and after mutation of LORELEI (LRE), which encodes a GPI-anchored proteins (Capron et al. 2008; Kessler et al. 2010; Tsukamoto et al. 2010). LRE and a similar protein, LORELEI-like GPI-anchored protein 1 (LLG1), interact with FER and function as chaperones, helping guide FER to the cell membrane (Li et al. 2015a). LRE is also expressed in eggs and central cells, and a maternal LRE allele is involved in early seed development (Wang et al. 2017).
It is not yet clear what type of signal is received by FER. Rapid alkalinization factor (RALF), a type of CRPs, has been reported to bind to FER (Haruta et al. 2014); however, the function of RALF in fertilization is not clear. Some A. thaliana homologues of early nodulin (ENOD), which functions in nodule formation, act specifically in synergid cells. When the function of these ENODs is inhibited, the pollen tubes continue growing without rupturing in the ovule as shown in fer. Because ENODs bind directly to the extracellular domain of FER, ENODs may be involved in FER signaling (Hou et al. 2016). FER regulates the production of reactive oxygen species (ROS) in synergid cells. When ROS are applied to the pollen tube, calcium flows into the pollen tube causing the pollen tube to burst (Duan et al. 2014). ROS are also involved in the proper elongation of pollen tubes. In the pollen tube, two genes encoding NAD(P)H oxidases, AtRbohH and AtRbohJ, and two receptor-like kinase proteins that closely resemble FER, ANXUR1 (ANX1) and ANX2, regulate pollen tube elongation via ROS production (Boisson-Dernier et al. 2013; Kaya et al. 2014; Lassig et al. 2014).
Pollen tube mutations resulting in phenotypes similar to those associated with FER are also reported. In a triple mutant harboring mutations in the MYB97, 101, 120 transcription factors, the pollen tube remains growing within the ovule (Leydon et al. 2013; Liang et al. 2013). A mutation in IV2 (At2g23590), which encodes a methylesterase, shows a similar phenotype (Lin et al. 2014). Mutations in turan (tun) and evan (evn), encoding a UDP-glycosyltransferase and dolichol kinase, respectively, also result in a phenotype similar to associated with fer mutation. These proteins regulate protein modifications involved in pollen tube–synergid cell interactions (Lindner et al. 2015).
Gamete fusion and termination of pollen tube guidance
Cell membrane fusion between the sperm cell and egg or central cells occurs on average 8.5 min after the rupture of pollen tubes (Hamamura et al. 2011). A CRP protein EGG CELL1 (EC1), which is expressed in the egg cell, is secreted extracellularly after sperm cells are released (Sprunck et al. 2012). When EC1 peptide is applied to the pollen tube, GENERATIVE CELL SPECIFIC 1 (GCS1), which is essential for gametophyte fusion, is localized on the surface of sperm cells. A mutation in GAMETE EXPRESSED 2 (GEX2), expressed in sperm cells, also results in membrane fusion (plasmogamy) abnormalities in a manner similar to that associated with gcs1 (Mori et al. 2014, 2006).
The pollen tube of the gcs1 mutant is normally attracted to the ovule (Takahashi et al. 2017), but the ovule that receives the pollen tube fails to become fertilized, because the gametophytes cannot fuse. It has been observed that ovules failing to fertilize will accept a second pollen tube. This was suggested to be due to attraction of the second pollen tube by another remaining synergid cell, known as a fertilization recovery system (Kasahara et al. 2012). In other words, in wild-type situations, gamete cell communication including the interaction between EC1 and GCS1 is required not only for gamete membrane fusion but also for rejection of multiple pollen tubes (polytubey) (Beale et al. 2012; Sprunck et al. 2012). Interestingly, ovules were also enlarged, and seed coat development was observed in ovules that failed to fertilize via a gsc1 pollen tube. Hence, fertilization is not necessarily required for ovule enlargement and seed coat development, and pollen tube contents released at the same time as sperm cells signal this phenomenon (Kasahara et al. 2016).
The HAP2/GCS1 homolog is conserved in many organisms and is thought to be an ancestral gamete fusogen (Wong and Johnson 2010). The conformation of the extracellular domain of the Tetrahymena HAP2/GCS1 ortholog is structurally similar to those of the dengue virus E glycoprotein and its related class II viral fusogen (Fedry et al. 2017; Pinello et al. 2017). When EFF-1, a somatic cell fusion factor of Caenorhabditis elegans, was expressed in hamster BHK cells, these cells were able to fuse. Similarly, BHK cells expressing HAP2/GCS1 of A. thaliana also fused (Valansi et al. 2017). These results suggest that eukaryotes acquired the cell fusion system early in evolution. The gametophyte fusion mechanism involved in reproduction may have evolved from an ancient cell fusion system. Izumo is expressed in mammalian spermatozoa, and Juno is expressed in eggs. Izumo is a membrane protein belonging to the immunoglobulin superfamily, whereas Juno is anchored to the membrane via a GPI anchor domain, functioning as a receptor for Izumo (Bianchi et al. 2014). Although a HAP2/GCS1 receptor specifically expressed in A. thaliana egg cells has not been identified yet, membrane proteins expressed specifically in eggs are candidates for further investigation.
How is the pollen tube guidance signal terminated? Detailed observations of the ovules after fertilization showed that the remaining synergid cell (persistent synergid cell) was fused to the endosperm. The LURE signal expressed in the synergid cells was diluted by endosperm fusion, making it impossible to attract the pollen tube. This fusion between the synergid and endosperm cells is the third cellular fusion phenomenon to be identified in plants outside of gametophyte fusion events (egg cell with one sperm cell and central cell with another sperm cell) (Maruyama et al. 2015). In an AGP4/JAGGER loss-of-function mutant, degradation of persistent synergid cells was inhibited, resulting in attraction of multiple pollen tubes (Pereira et al. 2016b, c). This finding also indicates that the loss of function in synergid cells is important for cessation of the signal attracting the pollen tube.
Lab-on-a-chip technique to analyze pollen tube guidance in vitro
Conventional pollen tube guidance studies have included genetic techniques such as isolation of mutants, reverse genetics techniques such as analysis of genes expressed in specific cells, and in vitro analysis such as applying specific substances to pollen tubes growing on medium. From an in vitro pollen tube guidance assay, it was discovered that the pollen tubes of T. fournieri and A. thaliana are attracted by the LURE protein (Okuda et al. 2009; Takeuchi and Higashiyama 2012). In this assay, gelatin beads containing the LURE protein were placed obliquely in front of the pollen tube using a glass needle (Fig. 2a, b). Handling gelatin beads using a glass needle under a microscope requires great skill. Additionally, pollen tubes grow randomly on the medium, which increases the difficulty associated with guidance assays. A micro-electro-mechanical system (MEMS) is a device integrating electronic circuits and mechanical parts. Although this technology is used to create sensors, etc., because it can handle substances on the micrometer/nanometer scale, it is now used in biology fields. In plant science, a system called RootChip was developed using MEMS technology to observe root elongation and apply arbitrary substances to specific parts of roots (Grossmann et al. 2011). A platform to analyze pollen tube guidance has also been developed using this technology (Horade et al. 2013; Yetisen et al. 2011). Horade and colleagues developed a microfluidic device made of a type of silicon called dimethylpolysiloxane (PDMS). By creating a mold designed using the CAD software and solidifying it by PDMS resin application to the mold, they were able to create a flow channel standardized precisely on a micrometer scale. When ovules of T. fournieri were placed on one side of the T-shaped channel and the buffer on the other, the pollen tubes grew toward the ovule (Fig. 2c). When the ovule was applied with the embryo sac removed by ultraviolet laser, the attraction rate of the pollen tube decreased. These results showed that the pollen tube senses substances secreted from the ovule, and that the direction of elongation is influenced by the signal (Horade et al. 2013).
Devices using PDMS can be freely designed in terms of size and shape, such that they can be applied to various types of pollen tube analyses (Agudelo et al. 2013, 2016; Ghanbari et al. 2014; Hu et al. 2016, 2017; Sanati Nezhad et al. 2014a, b; Sato et al. 2015; Shamsudhin et al. 2016). For example, by designing a very narrow channel, it was shown that pollen tubes and root hairs can pass through a width of only 1 μm (Yanagisawa et al. 2017). When a fluorescently labeled molecule was added to dextran (molecular weight 10 kDa) and passed through one of these devices, a concentration gradient formed in the flow channel (Horade et al. 2013). A bead assay using LURE directly labeled with a fluorescent dye revealed that approximately 1,000 molecules of the LURE protein were required for pollen tube attraction (Goto et al. 2011). However, it remains unknown whether the absolute concentration of the attractant or the concentration gradient is important for attraction of the pollen tube. The application of MEMS technology can create devices that can strictly control the absolute concentration and concentration gradient of the attractant. Such a device enables analysis of the influence of the attractant on the pollen tube during fertilization.
Recent studies have led to a deeper understanding of how pollen tubes near the micropyle receive signals, enter the embryo sac from the micropyle, and rupture to release sperm and begin fertilization (Fig. 1). However, the entire process of pollen tube guidance has yet to be revealed. Pollination causes large-scale gene expression changes (Rao et al. 2017). For example, the pollen transcriptome and pollen tube transcriptome are very different. Furthermore, pollen tubes germinated in vitro and those that pass through the pistil differ greatly in terms of gene expression and the proteins secreted from the pollen tube (Hafidh et al. 2016b). Differences in the genes expressed in the pistil in conspecific pollination versus pollination of incompatible pollen grains have been reported (Broz et al. 2017). In A. thaliana and rice, the pollen tube reaches the ovule several hours after pollination; however, in some plants, fertilization occurs over a period of several months, after pollen tube elongation ceases in the middle of the pistil (Chen and Fang 2016). Because the pollen tube and pistil tissue affect each other in these plants, it is important to investigate their gene expression profiles in detail.
There are still many uncertainties as to when and how the pollen tube will reach the ovule in the pistil. In A. thaliana, after the pollen tube elongates to the surface of the transmitting tract, it extends along the funiculus and enters the ovule from the micropyle. The timing of pollen tube movement from the inside to the surface of the transmitting tract, and the type of molecules involved, remain unknown.
Mutants defective in micropylar pollen tube guidance and transgenic plants with suppression of AtLURE expression show abnormal pollen tube guidance to the micropyle; however, pollen tube growth to the funiculus is normal (Takeuchi and Higashiyama 2012). This suggests that there is a signal guiding the approach of the pollen tube to the micropyle, apart from the signaling by LURE. In the T. fournieri ovule-chasing experiment, the effective distance between the attracting signal and synergid cells was estimated to be ~ 100 μm (Higashiyama and Hamamura 2008). In experiments using microdevices, pollen tubes are attracted to ovules located several millimeters away (Fig. 2c) (Horade et al. 2013). If ovule development is abnormal, then pollen tube guidance is also abnormal (Wang et al. 2016a). For these reasons, factors derived from sporophyte tissue may induce pollen tubes to travel long distances.
Interspecific crossing experiments in which A. thaliana was pollinated by pollen grains from closely related species showed recognition between male and female cells in various processes of pollen tube elongation (Shimizu 2002); however, most of the specific factors affecting their interaction are unknown. Elongation of the pollen tube to a different region than usual has been observed as a result of heterogeneous pollination between Sagittaria pygmaea and S. trifolia (Lyu et al. 2017). Experiments using a microfluidic device suggest the existence of a factor that attracts pollen tubes to ovules over long distances in T. fournieri, but its regulating factor is also unknown (Horade et al. 2013).
Techniques to observe deep within the pistil are necessary to understand these attraction processes and to identify attractants. In recent years, the reagent ClearSee, which enables the transparency and observation of plant tissues, was developed. Since ClearSee does not abolish GFP fluorescence, it is now possible to observe GFP expression deep within the pistil by making the tissue transparent (Kurihara et al. 2015). Using this method, it may be possible to identify genes that are expressed at specific sites in pistil tissues. Additionally, the excitation of fluorescent proteins by multiphoton lasers is an effective means to observe deeply situated tissues in live plant specimens. Using a two-photon excitation confocal laser microscope, live imaging of pollen tubes growing within the pistil can be achieved (Mizuta et al. 2015). In vivo observations using these newly developed technologies and semi-in vitro analyses using microfluidic devices and other techniques combined with conventional genetic/reverse genetic analyses will enable elucidation of the entire process of pollen tube guidance.
Author thank Ms. Ryoko Tsukamoto for providing pictures in Figure 2. This work was supported in part by Grants-in-Aid from the Ministry of Education, Culture, Sports, Science and Technology for Innovative Areas (15H04385 and 16H06465) and Grant-in-Aid for Scientific Research (B) (15H01231) to MMK.
- Boisson-Dernier A, Lituiev DS, Nestorova A, Franck CM, Thirugnanarajah S, Grossniklaus U (2013) ANXUR receptor-like kinases coordinate cell wall integrity with growth at the pollen tube tip via NADPH oxidases. PLoS Biol 11:e1001719. https://doi.org/10.1371/journal.pbio.1001719 PubMedPubMedCentralCrossRefGoogle Scholar
- Dai XR, Gao XQ, Chen GH, Tang LL, Wang H, Zhang XS (2014) ABNORMAL POLLEN TUBE GUIDANCE1, an endoplasmic reticulum-localized mannosyltransferase homolog of GLYCOSYLPHOSPHATIDYLINOSITOL10 in yeast and PHOSPHATIDYLINOSITOL GLYCAN ANCHOR BIOSYNTHESIS B in human, is required for Arabidopsis pollen tube micropylar guidance and embryo development. Plant Physiol 165:1544–1556. https://doi.org/10.1104/pp.114.236133 PubMedPubMedCentralCrossRefGoogle Scholar
- Hafidh S, Potesil D, Fila J, Capkova V, Zdrahal Z, Honys D (2016b) Quantitative proteomics of the tobacco pollen tube secretome identifies novel pollen tube guidance proteins important for fertilization. Genome Biol 17:81. https://doi.org/10.1186/s13059-016-0928-x PubMedPubMedCentralCrossRefGoogle Scholar
- Higashiyama T, Takeuchi H (2015) The mechanism and key molecules involved in pollen tube guidance. Annu Rev Plant Biol 66:393–413. https://doi.org/10.1146/annurev-arplant-043014-115635 PubMedCrossRefGoogle Scholar
- Huang WJ, Liu HK, McCormick S, Tang WH (2014) Tomato pistil factor STIG1 promotes in vivo pollen tube growth by binding to phosphatidylinositol 3-phosphate and the extracellular domain of the pollen receptor kinase LePRK2. Plant Cell 26:2505–2523. https://doi.org/10.1105/tpc.114.123281 PubMedPubMedCentralCrossRefGoogle Scholar
- Lin SY, Chen PW, Chuang MH, Juntawong P, Bailey-Serres J, Jauh GY (2014) Profiling of translatomes of in vivo-grown pollen tubes reveals genes with roles in micropylar guidance during pollination in Arabidopsis. Plant Cell 26:602–618. https://doi.org/10.1105/tpc.113.121335 PubMedPubMedCentralCrossRefGoogle Scholar
- Pereira AM, Nobre MS, Pinto SC, Lopes AL, Costa ML, Masiero S, Coimbra S (2016c) “Love is strong, and you’re so sweet”: JAGGER is essential for persistent synergid degeneration and polytubey block in Arabidopsis thaliana. Mol Plant 9:601–614. https://doi.org/10.1016/j.molp.2016.01.002 PubMedCrossRefGoogle Scholar
- Tsukamoto T, Qin Y, Huang Y, Dunatunga D, Palanivelu R (2010) A role for LORELEI, a putative glycosylphosphatidylinositol-anchored protein, in Arabidopsis thaliana double fertilization and early seed development. Plant J 62:571–588. https://doi.org/10.1111/j.1365-313X.2010.04177.x PubMedCrossRefGoogle Scholar