2023 Volume 88 Issue 1 Pages 35-39
Sexual reproduction in angiosperms is a complex and precise process of regulation, which includes the pollen tube guidance, double fertilization, and the seed development process. The previous report identified that for seed development, there is a period of preparation for fertilization (between pollen tube guidance and double fertilization) that is called pollen-tube-dependent ovule enlargement morphology (POEM). It has been shown that the pollen tube content (PTC) plays a crucial role in the enlargement of ovules and the initiation of seed coat formation. However, we did not investigate the potential of endosperm proliferation in autonomous mutants at a later stage. Here, we investigated this phenomenon using vanillin staining and transparent experiments to examine the manner, in which the PTC affects the potential of endosperm formation. Interestingly, the PTC increased the number of endosperm nuclei without fertilization equally and synchronously in mea and fis2 ovules. This finding might help improve the study of apomixes and our understanding of how the molecular mechanisms that regulate this phenomenon will contribute to plant reproductive science in the future.
Plant fertilization is an extremely important phenomenon because it is strongly associated with genetics and seed development in plants (Nawaschin 1898, Guignard 1899). In addition, the foundations of the reproductive system have direct implications on plant breeding and agriculture, as it has been connected directly to normal seed formation.
We reported previously an important plant phenomenon, which was termed pollen-tube-dependent ovule enlargement morphology (POEM) (Kasahara et al. 2016, 2017). POEM ensured that, even in the absence of fertilization, the release of the pollen tube content (PTC) into the ovule leads to the enlargement of the ovule and the formation of the seed coat. Moreover, we briefly reported that the PTC plays a crucial role in increasing autonomous endosperm without fertilization when combined with autonomous endosperm mutants (Kasahara et al. 2016, 2017). That study showed that, before double fertilization, ovules were stimulated by PTC to perform a series of preparatory activities, thus creating an enabling environment for the development of the endosperm nuclei and zygotes. However, because of the double fertilization defect in the gcs1 mutant (Mori et al. 2006, Nagahara et al. 2015, Fédry et al. 2017), ovules cannot continue to develop. Several genes that encode regulators of early seed development are also required for the repression of fertilization-independent seed development. The Medea (MEA) (Grossniklaus et al. 1998, Pien and Grossniklaus 2007), fertilization-independent endosperm (FIE), and fertilization-independent seed-2 (FIS2) genes (Chaudhury et al. 1997, Luo et al. 1999), as well as the MSI1 gene (Köhler et al. 2003, Guitton and Berger 2005), are required for the repression of seed development in the absence of fertilization. These four mutant ovules undergo initiation of endosperm development in the absence of fertilization. Furthermore, these mutants show a gametophytic maternal effect: the endosperm/seed development phenotype can only be observed when the female gametophytes possess the mutations.
We reported previously that PTC, as the first signal, promotes ovule enlargement and mentioned briefly that the PTC can facilitate endosperm proliferation when the ovules of autonomous endosperm mutants are crossed with the gcs1/gcs1 mutant (Kasahara et al. 2016). However, the results of our previous report are not sufficient to understand the whole underlying mechanism deeply, as we have not observed the potential of endosperm proliferation in later days in autonomous endosperm mutants. Therefore, in this report, we performed a crossing experiment to understand the function of PTC both at the early stage and late stages and found that PTC release can increase the potential for endosperm proliferation without fertilization.
Arabidopsis thaliana ecotype Columbia (Col-0) plants were used as the wild-type (WT) plants. Test cross-experiments were conducted in gcs1/gcs1 (Nagahara et al. 2015, Kasahara et al. 2016), mea/+ (Grossniklaus et al. 1998), fis2/+ (Chaudhury et al. 1997), and WT plants. Seeds were sterilized with Cl2 gas (add 5 mL of hydrochloric acid to 45 mL of sodium hypochlorite solution to produce Cl2 gas) for 20 min, blown dry in a clean bench, and germinated on 1/2 MS plates in a growth chamber at 21.5°C under 24 h of light after cold treatment (4°C) for 2 days. Subsequently, seedlings were transferred to soil and grown at 21.5°C under 24 h of light.
Phenotypic analysisFor staining of the silique tissue, the WT flowers were emasculated at stage 12c and pollinated with gcs1/gcs1, mea/+, and fis2/+ pollen grains, as well as no pollination. For mea and fis2 experiments, the mea and fis2 mutant flowers were emasculated at stage 12c and pollinated with WT and gcs1/gcs1 pollen grains, as well as no pollination. The siliques were collected at 3 days after pollination (3DAP); in the non-pollinated group (mea and fis2), siliques were collected at 7 days after emasculation (7 days with no pollen tubes, 7DNPT).
Vanillin stainingSiliques were collected from each mutagenized plant 2–3 days after pollination and stained with 1% 4-hydroxy-3-methoxy benzaldehyde (vanillin) in 6 M HCl (Sigma) without dissection. Vanillin-staining images were collected using a Nikon Ni-U camera system.
Transparent analysisOvules were examined by dissection on glass slides in a chloral hydrate : glycerol : water mixture (8 : 1 : 2, w/v/v). Slides were incubated at room temperature for 2 h, and ovules were analyzed by differential interference contrast microscopy.
To investigate how the PTC affects the ovule phenotype, we crossed WT, mea/+, fis2/+, and gcs1/gcs1 mutants and performed vanillin staining (Fig. 1). According to a previous report (Kasahara et al. 2016, Liu et al. 2019, 2020), when the WT ovules accepted no pollen tubes, the WT ovules were not stained by vanillin as they did not receive PTC (Fig. 1A). In turn, when the WT ovules accepted WT pollen tubes, the ovules were fully stained by vanillin (Fig. 1B). Finally, when the WT ovules accepted gcs1/gcs1 pollen tubes, the ovules were enlarged and produced a partial seed coat (Fig. 1C). As described by Nagahara et al. (2015), gcs1/gcs1 pollen produces embryo-alone phenotype or endosperm-alone phenotype at an extremely low frequency since the mutant sperm cells can fertilize egg cell or central cell by chance. In this case, when the sperm cell single-fertilized to the egg cell-producing embryo, the ovule was partially stained by vanillin but when the sperm cell single-fertilized to the central cell-producing endosperm, the ovule was fully stained by vanillin. Next, we observed mea/+ ovules without pollination and found that the ovules had no staining (Fig. 1D), indicating that the mea/+ mutant has no potential for endosperm proliferation at 3 days with no pollen tube inserted (3DNPT). However, at 7DNPT, mea/+ mutant ovules exhibited endosperm proliferation at a very low ratio (Fig. 1E), indicating that 7 days are required for endosperm proliferation in the mea/+ mutant in the absence of pollination. We observed mea/+ ovules pollinated by WT pollen and the ovules were fully stained by vanillin (Fig. 1F) since all ovules were fertilized to contribute to endosperm formation. However, when the mea/+ ovules were crossed using gcs1/gcs1 mutant pollen, we observed a significant number of vanillin-positive ovules at 3DAP, indicating that the PTC has the potential to proliferate the endosperm at 3DAP, which is not sufficient for endosperm proliferation in the mea/+ ovules without pollination (Fig. 1G). Similarly, the PTC was sufficient to proliferate the endosperm at 3DAP in fis2/+ ovules (Fig. 1H–K).
Next, we observed transparent ovules crossed using a variety of pollens (Fig. 2). WT ovules without pollination exhibited intact egg and central cells (3DNTP, Fig. 2A). WT ovules crossed with WT pollen showed an embryo and proliferated endosperm at 3DAP (Fig. 2B). Most WT ovules crossed with gcs1/gcs1 pollen showed enlargement, albeit in the absence of an embryo and endosperm formation, which corresponded to the POEM phenotype (Fig. 2C); however, a few ovules showed an embryo-alone phenotype (Fig. 2D) or an endosperm-alone phenotype (Fig. 2E) at very low frequencies. Most mea/+ ovules crossed with WT pollen showed a WT-fertilized phenotype (Fig. 2F), whereas most mea/+ ovules without pollination exhibited a WT-NPT phenotype (Fig. 2G). Moreover, some mea/+ (Fig. 2H) ovules crossed with gcs1/gcs1 pollen displayed just-enlarged phenotypes. A subgroup of fis2/+ ovules without pollination showed a no-endosperm-proliferation phenotype at 7DNTP (Fig. 2I). Figure 2J and K depict the phenotype of proliferated endosperm observed in fis2/+ ovules without pollination at 7DNTP. Some fis2/+ ovules crossed with gcs1/gcs1 pollen exhibited a proliferated endosperm phenotype at 3DAP (Fig. 2L), which was never observed in fis2/+ ovules without pollination at 3DNTP (Fig. 2I). Figure 2M and N depict the endosperm proliferation phenotype displayed by fis2/+ ovules crossed with gcs1/gcs1 pollen at 5DAP. Figure 2O and P depict the endosperm proliferation phenotype displayed by fis2/+ ovules that were crossed with gcs1/gcs1 pollen at 7DAP. These results indicate that 3 days are sufficient for the autonomous proliferation of the endosperm when the PTC is released inside the female gametophyte in these mutants without fertilization.
Fig. 3 shows the statistical analysis of the ratios of vanillin-stained ovules. The crossing of WT ovules with gcs1/gcs1 pollen at 3DAP led to a proliferated endosperm in 3.7% of the ovules (FSC-B), whereas WT ovules without pollination at 3DNTP produced no endosperm. The crossing of mea/+ ovules with gcs1/gcs1 pollen at 3DAP led to a proliferated endosperm in 8.3% of the ovules, whereas mea/+ ovules without pollination at 3DNTP produced no endosperm. Moreover, mea/+ ovules without pollination at 7DNTP produced an endosperm in 6.7% of cases, which was similar to the ratio detected in the ovules crossed with gcs1/gcs1 pollen at 3DAP. The crossing of fis2/+ ovules with gcs1/gcs1 pollen at 3DAP led to endosperm proliferation in 30.8% of the ovules, whereas fis2/+ ovules without pollination at 3DNTP produced endosperm in 0.7% of cases. Finally, fis2/+ ovules without pollination at 7DNTP produced an endosperm in 4.8% of cases. These results indicate that the PTC enhances and facilitates autonomous endosperm development both in WT and autonomous endosperm mutants.
Fig. 4 shows the statistical analysis of the ratios of the ovule phenotypes based on transparent experiments. The crossing of WT ovules with gcs1/gcs1 pollen at 3DAP led to double fertilization in 2.3% of the ovules, an embryo-only phenotype in 4.9% of the ovules, and an endosperm-only phenotype in 0.7% of the ovules, similar to the results of Nagahara et al. (2015); in contrast, WT ovules without crossing at 3DNTP exhibited a no-POEM phenotype exclusively (no phenotypic change). The crossing of mea/+ ovules with gcs1/gcs1 pollen at 3DAP led to an endosperm-only phenotype in 20.7% of the ovules, whereas mea/+ ovules without crossing at 3DNTP showed the no-POEM phenotype exclusively. Moreover, mea/+ ovules without crossing at 7DNTP displayed a 2.0% endosperm/proliferation ratio. In fis2/+ ovules, a similar endosperm trend was observed. These results suggest that the frequency of endosperm proliferation is dramatically increased by the PTC and that the PTC enhances the potential for endosperm proliferation in autonomous endosperm mutants.
We showed previously that the PTC can increase the number of endosperm nuclei (Kasahara et al. 2016); however, our observation was not sufficient because we did not confirm the terminal phenotypes of the autonomous endosperm mutants to investigate the potential ability of endosperm proliferation. However, in this report, we investigated this potential by testing and comparing 3DAP, 3DNTP, and 7DNTP for each crossing experiment. The crossing of mutant ovules with gcs1/gcs1 pollen at 3DAP led to an increase in the ratio of endosperm proliferation, similar to the ovules with no pollination at 7DNTP, indicating that the potential for endosperm proliferation is similar at 3DAP and 7DNTP. We found that the PTC increased not only the number of nuclei in autonomous endosperm but also the potential for endosperm proliferation, i.e., the PTC can shorten the time of endosperm proliferation. Previously, we found that the vanillin-stained states represent the phenotypes of the ovule; for example, partial staining indicates the POEM state, and full staining indicates endosperm proliferation (Ohad et al. 1996, Chaudhury et al. 1997). In this report, we added other probes by testing for full seed coat staining and big ovules (FSC-B) and full seed coat staining and small ovules (FSC-S). We concluded that FSC-B ovules must have endosperm proliferation and might have an embryo inside, whereas FSC-S ovules might exhibit endosperm proliferation and potentially have an embryo inside. Here, we found that the PTC increases and facilitates endosperm proliferation and identified new probes, which can distinguish the phenotype of ovules. Further research is required to understand the molecular mechanism underlying this phenomenon.
We thank Xiaoyan Wu, Shaowei Zhu, Chen Huang, Jiale He, Aidi Xu, and Wei Xia for their technical assistance. This work was supported by start-up funds from the School of Life Sciences, Fujian Agriculture and Forestry University (Grant#: 114-712018008 to R.D.K.) and the FAFU-UCR Joint Center, Haixia Institute of Science and Technology, Fujian Agriculture and Forestry University (Grant#: 102-118990010 to R.D.K.). This work was also supported by the Chinese NSFC fund (Grant#: 31970809).
