https://orcid.org/0000-0003-1827-4459
Abstract
In the anthers and ovaries of flowers, pollen grains and embryo sacs are produced with uniform cell compositions. This stable gametogenesis enables elaborate interactions between male and female gametophytes after pollination, forming the highly successful sexual reproduction system in flowering plants. As most ovules are fertilized with a single pollen tube, the resulting genome set in the embryo and endosperm is determined in a single pattern by independent fertilization of the egg cell and central cell by two sperm cells. However, if ovules receive four sperm cells from two pollen tubes, the expected options for genome sets in the developing seeds would more than double. In wild-type Arabidopsis thaliana plants, around 5% of ovules receive two pollen tubes. Recent studies have elucidated the abnormal fertilization in supernumerary pollen tubes and sperm cells related to polytubey, polyspermy, heterofertilization and fertilization recovery. Analyses of model plants have begun to uncover the mechanisms underlying this new pollen tube biology. Here, we review unusual fertilization phenomena and propose several breeding applications for flowering plants. These arguments contribute to the remodeling of plant reproduction, a challenging concept that alters typical plant fertilization by utilizing the current genetic toolbox.
Introduction
In sexually reproducing organisms, male and female gametes fuse to produce diploid zygotes. If eggs are fertilized by more than one sperm, also called polyspermy, the zygote contains polyploid genomes that may cause imbalanced paternal–maternal gene expression. In animals, polyspermy often causes mitotic defects due to excessive sperm-derived centrosomes, and thus, polyspermy blocking systems have evolved to avoid multiple fertilization (Bhakta et al. 2019). To our knowledge, however, polyspermy-induced mitotic defects have not been reported in flowering plants that do not produce centrosomes. Recent studies have investigated the sophisticated systems that prevent polyspermy and establish a rigid, one-to-one relationship between male and female tissues prior to gamete association.
Sexual reproduction in flowering plants occurs by the pollen tube growing into the pistil and delivering two sperm cells to the female gametophyte in the ovule (Fig. 1A). Female gametophytes are tiny tissues containing two dimorphic female gametes: an egg cell and a central cell. The other accessory female gametophyte cells are highly differentiated and vary in number or ploidy. In the Polygonum-type female gametophyte, which is observed most frequently, there are two synergid cells and three antipodal cells; thus, a seven-celled female gametophyte is formed with the egg and central cells. The synergid cells secrete pollen tube attractant peptides from the micropylar end of plasma membrane invaginations, termed the filiform apparatus, and control the number of pollen tubes participating in fertilization. After the pollen tube arrives at a synergid cell, sperm cells are released toward degenerating receptive synergids, and each sperm cell independently fertilizes the egg and central cells. Fertilized egg cells develop into embryos, and fertilized central cells develop into the endosperm. In double fertilization, the ovule only requires the insertion of a single pollen tube because of the high efficiency of pollen tube guidance and gamete fusions.
Polytubey and polytubey blocking in A. thaliana. (A) Pollen tube guidance in the pistil. In the vicinity of the ovules, pollen tubes precisely enter the micropyle responding to attractant peptides secreted by the synergid cells. (B) Polytubey blocking events before pollen tube discharge. (C) Polytubey blocking events after pollen tube discharge. (D) Heterofertilization: the embryo and endosperm carry paternal genomes from different fathers. (E) Polyspermy: the egg cell is fertilized by two sperm cells from different pollen tubes.
Abnormal fertilization has been observed in various plants, in which ovules receive more than one pollen tube, also called polytubey (Maheshwari 1950, Beale et al. 2012). Polytubey provides excess sperm cells in the ovule. However, it is unknown what happens after fertilization by multiple pollen tubes because of the low frequency of polytubey in wild-type ovules. Recent advances in pollen tube guidance have gradually uncovered the mechanism of male–female communication underlying polytubey. Here, we review unusual fertilization phenomena through an analysis of mutant model plants and indicate possible applications for breeding technologies. These ideas propose remodeling plant reproduction by altering typical plant fertilization, utilizing the current genetic toolbox, which may generate novel reproduction pathways that are useful for human beings.
Polytubey and Polytubey Blocking
The relatively low frequency of polytubey after a competitive race of pollen tubes implies the existence of a sophisticated polytubey blocking mechanism. Growing genetic data have provided evidence of several steps of gating systems in pollen tube guidance. After pollination, pollen tubes in the transmitting tract penetrate the septum and emerge on the surface of the placenta in response to attractants from the ovules (Zhong et al. 2019a). The septum receives RAPID ALKALINIZATION FACTOR (RALF) peptides from the pollen tube via the Catharanthus roseus Receptor-Like Kinase 1–type receptors, including FERONIA, ANJEA and HERCULES1, where the signals locally restrict secondary pollen tube penetration until the first pollen tube is discharged and terminate RALF secretion (Zhong et al. 2022a).
Other polytubey blocking mechanisms are activated immediately before or after double fertilization. During pollen tube reception, synergid cells produce nitric oxide (NO) in a FERONIA- and pectin-dependent manner, where the NO signal stops AtLURE1 secretion and induces the post-translational nitrosation of AtLURE1 to reduce its attraction activity (Duan et al. 2020). Immediately after fertilization, egg cells secrete EGG CELL SPECIFIC 1 (ECS1) and ECS2, which are aspartic endopeptidases that cleave AtLURE1 and contribute to the rapid clearance of attractant signals around the receptive synergid cells (Yu et al. 2021). Fertilization of egg cells also accelerates nuclear degeneration of the non-receptive persistent synergid via nuclear accumulation of ETHYLENE INSENSITIVE 3 (EIN3) and ETHYLENE INSENSITIVE LIKE 1 in the ovules, terminating pollen tube attraction (Volz et al. 2013, Maruyama et al. 2015, Li et al. 2021, Heydlauff et al. 2022). Persistent synergid inactivation is also controlled by fertilization of the central cell, independent of the egg cell–mediated pathway (Maruyama et al. 2013). Persistent synergids are absorbed by fertilized central cells, and the rapid efflux of the synergid cytoplasm likely reflects decreasing pollen tube attraction activity (Maruyama et al. 2015). Arabinogalactan Protein 4 (AGP4), a highly glycosylated protein also known as JAGGER, is predominantly expressed in ovular integuments and was shown to regulate nuclear degeneration of the persistent synergid (Pereira et al. 2016a, 2016b). However, the roles of this polysaccharide signal in the synergid inactivation and possible interaction with fertilization-dependent other polytubey blocking pathways are largely unknown.
Fertilization Recovery and Heterofertilization
Any abnormalities in polytubey blocking mechanisms increase the polytubey rate. The most severe polytubey phenotype has been observed when wild-type ovules receive fertilization-defective sperm cells. For example, in homozygous mutants of the HAPLESS2/GENERATIVE CELL SPECIFIC 1, which encodes a crucial fusogenic protein in the sperm cell, around 70% of ovules received a second pollen tube (Mori et al. 2006, Nagahara et al. 2015, Fedry et al. 2017, Zhong et al. 2022a). In polytubey, fertilization failure by the first mutant pollen tube can be recovered by a fertilization-competent wild-type second pollen tube (Beale et al. 2012, Kasahara et al. 2012). Fertilization recovery is initiated only when the pistil is pollinated with more pollen grains than there are ovules, but the fertilization success rate is less efficient than that of normal fertilization (Kasahara et al. 2013, Li et al. 2022a). Nevertheless, fertilization recovery may be widely conserved among flowering plants to maximize fertility because ovules usually produce two synergid cells, and polytubey is observed in a variety of flowering plants.
A class of mutant sperm cells with reduced fertility (e.g. kokopelli and gex2) often induce a single fertilization of either the egg cell or the central cell (Ron et al. 2010, Mori et al. 2014). After a single fertilization, ovules partially maintain persistent synergid activity, and fertilization is recovered by the attraction of a second pollen tube. In this type of fertilization recovery, the embryo and endosperm carry different paternal genomes; normal double fertilization introduces genetically identical paternal genomes from single pollen tubes (Maruyama et al. 2013). The peculiar fertilization producing heterogeneity of the embryo and endosperm is termed heterofertilization and was originally found in a study of maize (Sprague 1929, 1932), which found that heterofertilization rates can be increased by dual pollination using special pollen-carrying chemically induced single sperm cells or pollen from haploid inducer lines; this is consistent with findings on heterofertilization in Arabidopsis thaliana (Kato 2001, Tian et al. 2018).
Polyspermy and Polyspermy Blocking
In contrast to normal fertilization, polytubey supplies four sperm cells to a single ovule and may increase the risk of polyspermy, multiple fertilization of the egg cell or central cell by more than one sperm cell. Scott et al. (2008) tried to detect polyspermy in planta using tetraspore (tes) mutant A. thaliana plants. The TES gene encodes a kinesin that delivers a vesicle for phragmoplast formation (Yang et al. 2003). Due to aberrant cell plate formation after the tetrad stage, the tes mutant produces a large pollen grain that carries supernumerary sperm cells (Spielman et al. 1997). Analysis of ploidy in the F1 generation of the tes mutant revealed that the excess sperm cell supply increased polyspermy in the central cell by 26%, while the egg cell was not susceptible to polyspermy (Scott et al. 2008). The discrepancy in polyspermy competence between the egg cell and the central cell was confirmed by a semi–in vivo fertilization assay of tes mutants (Nagahara et al. 2020). Polyspermy blocking in the egg cell is partially explained by a physical barrier of the cell wall that emerges after fertilization in an in vitro fertilization assay of maize gametes (Kranz et al. 1995). Recently, egg cells have been shown to be a predominant fertilization target for a single sperm cell in the cdka;1 mutant (Nowack et al. 2006, Aw et al. 2010, Li et al. 2022a). The strong polyspermy blocking and predominant fertility may be key egg cell features that determine rigid gamete pairs during double fertilization (Fig. 1B).
Despite the many layers of blocking mechanisms, some ovules still receive multiple pollen tubes and undergo polyspermy. A dual pollination experiment using different pollen donors demonstrated that Arabidopsis wild-type plants produced 0.012% triploid seeds via polyspermy (Nakel et al. 2017). In contrast to animal cells that produce lethal multipolar spindles from excessive centrosomes under polyspermy, mitosis in centrosome-lacking flowering plants is unlikely to be affected in polyploid zygotes or endosperms. Consistently, the electrofusion of two sperm cells and an egg cell could recover viable triploid zygotes in rice plants (Toda et al. 2016, Toda and Okamoto 2016). The in vitro fertilization-mediated polyspermy significantly induced developmental arrest during embryonic cell proliferation, which may be due to the aberrant ratio of paternal and maternal genomes (Toda et al. 2018). Compared with the embryo, the endosperm is more susceptible to an imbalance in paternal–maternal genome dosage in interploid crosses (Kohler et al. 2010). Because of the triploid blocking mechanism, viable seeds containing endosperms with higher ploidy should not be obtained from wild-type plants. To adjust the paternal–maternal ratio in the endosperm after polyspermy, Grossniklaus performed dual pollination using tetraploid female plants and confirmed that polyspermy-derived endosperms were enriched in viable seeds in maize (Grossniklaus 2017).
Distinct from the tes-induced polyspermy, polyspermy detectable with dual pollination has a time lag between the approach of the first and second pollen tubes. Thus, it is still difficult to monitor whether the second pair of sperm cells shows preference or co-ordination in egg cell- and central cell fertilization as ordinal double fertilization completed by single pollen tube reception. In Arabidopsis dual pollination, seeds containing triploid embryos often produce triploid endosperms. In addition, triploid seeds have been shown to not drastically increase due to the loss of the paternally expressed triploid blocking gene ADMETOS. These data indicate that the seeds avoid triploid blocks via selective polyspermy of the egg cell (Kradolfer et al. 2013, Mao et al. 2020). In maize dual pollination, the polyspermy endosperm usually grows with embryos after normal fertilization (Grossniklaus 2017). Although these observations did not directly compare polyspermy rates between the egg and central cells, the independence of the polyspermy target was evident.
Triploid Blocking Avoidance Using Heterofertilization
Growing knowledge of the physiology and molecular mechanism of polytubey and polyspermy in recent years has enabled the production of new breeding technologies that may overcome problems in typical plant reproduction pathways. For example, triploid blocking is one of the major causes of seed abortion in interspecies or interploidy crosses. Based on molecular studies of the endosperm in model plants, this hybridization barrier was shown to be regulated by epigenetically imprinted genes, MADS transcription factors and auxin biosynthesis (Kohler et al. 2021). Genetic manipulation of these factors could inhibit triploid blocking and increase the penetration of interspecies or interploidy crosses (Feng and Xue 2006, Tonosaki et al. 2016, 2018, Kohler et al. 2021, Wang et al. 2021b). This straightforward approach would require optimization of individual endosperm gene manipulation in each case of cross-breeding. However, the heterofertilization approach only requires a single-fertilization-inducible plant and could become a ubiquitous tool that bypasses triploid blocking. First, a single fertilization inducer pollen tube fertilizes the central cell and produces a fully functional triploid endosperm with a normal 2:1 maternal–paternal genome ratio, and fertilization of the egg cell is recovered by the second pollen tube. Theoretically, the father of the hybrid seeds can be changed by any strain or species that has been unable to produce viable seeds due to triploid blocking in normal crosses (Fig. 2A). Although heterofertilization bypassing has never been tested before, triploid seeds resulting from polyspermy avoid seed abortion in a heterofertilization manner and could be considered proof-of-concept data (Mao et al. 2020).
Applications of polytubey biology. (A) Triploid blocking avoidance by heterofertilization: pollen grains from a triploid block–inducing plant are pollinated with pollen from a single fertilization inducer, such as the dmp8 dmp9 double mutant. The single fertilization inducer can produce viable endosperm-containing compatible genome sets and facilitate hybrid embryo formation via fertilization recovery by second pollen tube attraction. (B) Surrogate seed development: dual pollination is performed in pistils from haploid inducer plants that cause maternal chromosome elimination during embryogenesis. When polyspermy occurs in the egg cell, the triploid zygote finally develops into a diploid embryo through maternal genome elimination. The resulting seed inherits nuclear genomes from different fathers and maternal cytoplasmic genomes.
Isolation of a single fertilization inducer is a keystone of heterofertilization-mediated triploid block bypassing. Single fertilization inducers have different fertilization targets. The gex2 and kokopelli mutants do not show a clear preference for single fertilization (Ron et al. 2010, Maruyama et al. 2013, Mori et al. 2014). Mutant pollen grains of cdka;1 contain single sperm cells and have long been considered to cause non-preferential single fertilization (Nowack et al. 2006, Aw et al. 2010, Maruyama et al. 2013). However, a carefully performed restricted pollination experiment showed that the cdka;1 mutant single sperm cell predominantly fertilizes the egg cell (Li et al. 2022a). These single fertilization inducers are less useful in causing heterofertilization by bypassing the triploid block. In contrast, loss-of-function mutants of the sperm cell–specific four-span transmembrane proteins DOMAIN OF UNKNOWN FUNCTION 679 MEMBRANE PROTEIN 8 (DMP8) and DMP9 preferentially fertilize central cells, which would be a reliable single fertilization inducer for the heterofertilization approach (Takahashi et al. 2018, Cyprys et al. 2019). Recently, genome editing approach targeting DMP family genes have been reported in several organisms, including maize, tomato, A. thaliana, Brassica napus, Medicago truncatula and Nicotiana tabacum (Zhong et al. 2019b, 2020, 2022b, Wang et al. 2022, Li et al. 2022b, Zhang et al. 2022). Although most dmp mutants were intended to produce haploid inducers, genetic tools can also be used for bypassing triploid blocks via heterofertilization.
Cytoplasm Swapping of Surrogate Seed Technology
Polyspermy is another tool that could potentially expand breeding options. Tekleyohans and Gross-Hardt (2019) proposed three significant aspects of polyspermy in plant breeding: (i) bypass of the triploid block, as described earlier; (ii) improvement in productivity of the plants of interest by increasing ploidy levels, similar to other polyploid crop species cultivated today, and (III) integration of a single maternal genome and two paternal genomes in the F1 generation. The concept of tri-parental seed generation is new, and its significance in breeding science remains unclear. We propose the practical use of polyspermy, a surrogate seed technology that produces diploid offspring from two fathers. In this hypothetical method, researchers would perform dual pollination of pistils from haploid inducer lines that cause maternal chromosome elimination during embryogenesis (Thondehaalmath et al. 2021). Therefore, the resulting polyspermy seeds are expected to be ‘surrogate seeds’ containing diploid embryos comprising two different paternal genomes (Fig. 2B). Theoretically, surrogate seeds possess hybrid nuclear genomes that can be reproduced by normal sexual crosses. However, surrogate seeds would inherit their mitochondrial and chloroplast (plastid) genomes from the haploid inducer, whereas the normal cross-products inherit the cytoplasmic genomes concomitantly with the maternal nuclear genome (Ravi and Chan 2010, Ravi et al. 2014). In other words, surrogate seed technology can accelerate cytoplasm-swapping breeding due to the one-step replacement of the cytoplasm genome in the F1 hybrid.
Surrogate seed technology is in the conceptual stage, and the low efficiencies of haploid induction and strong polyspermy blocking activity in the egg cells are major technical barriers. Uniparental chromosome elimination was shown to be induced by modifications of the centromeric histone H3 (CENH3). In A. thaliana, the original CENH3-based haploidization study used a cenh3 null mutant complemented with a construct, GFP-tailswap CENH3, which had the N-terminal tail replaced with another H3 variant and tagged with GFP; the haploid inducing rate (HIR) was 34% when used as a female parent (Ravi and Chan 2010). Further studies revealed that point mutations in CENH3 also induce uniparental chromosome elimination, among which the cenh3 E89K mutant had the highest HIR (44.1%) in A. thaliana (Kuppu et al. 2015, 2020, Thondehaalmath et al. 2021). The GFP-tailswap was also tested in maize, but the HIR was low (0.08% females) (Kelliher et al. 2016). However, CRISPR/Cas9-mediated cenh3 heterozygous maize displayed approximately 5% HIR, indicating the broad conservation of haploid-inducible activity using the CENH3-targeting approach (Wang et al. 2021a). As Wang et al. mentioned, the cenh3 heterozygous maize can be improved through optimization of cross-conditions, similar to the case of other haploid inducers (Uliana Trentin et al. 2020). Like haploid inducer studies, extremely low levels of the present polyspermy rate can be improved by changing the number of pollen grains and the physiological conditions of the male and female gametophytes during dual pollination. A recently deposited preprint reported that double mutants of two egg cell–specific aspartic endopeptidase genes, ECS1 and ECS2, exhibited a polyspermy frequency three times higher (0.15%) than the wild-type in A. thaliana (Mao et al. 2022). The isolation of useful polyspermy inducers, as well as the development of efficient polyspermy detection systems, could be the next key breakthrough for polyspermy applications.
Summary and Future Direction of Polytubey and Polyspermy Biology
In the past two decades, genetic and live-imaging approaches have uncovered communications between male and female gametophytes, including pollen tube guidance, pollen tube reception and gametic fusions. These studies have elucidated a state-of-the-art sexual reproduction system with outstanding flexibility and robustness. The highly established system often does not allow polytubey and polyspermy as irregular fertilization events. However, recent findings in polytubey biology have begun to undo the rigid framework of double fertilization in A. thaliana. As polytubey and polyspermy induction rates improve, polytubey biology will spread to other model plants and then to non-model plants. Beyond these advances, polytubey and polyspermy will become breakthrough technologies in plant breeding. Heterofertilization-mediated triploid blocking avoidance and surrogate seed technology are early examples of polytubey and polyspermy applications. Currently, various male- or female-gametophytic mutants containing altered numbers of gametes are available (e.g. tes mutant pollen with supernumerary sperm cells and cdka;1 mutant pollen with single sperm cells). These genetic tools can contribute to further applications in polytubey and polyspermy.
Data Availability
The data of this article will be provided by the authors without undue reservation.
Funding
Japan Society for the Promotion of Science (JSPS) KAKENHI (JP20H03280, JP20K21432, JP20H05778, JP20H05781); Takeda Science Foundation (Life Science Research Grants); Yokohama City University [grant for 2020–2022 academic research, grant for 2020–2022 Research Development Fund, grant for 2021–2024 Strategic Research Promotion (SK201903)].
Disclosures
The authors have no conflicts of interest to declare.
