Arabidopsis thaliana Columbia-0 was used as the wild-type (WT) plant. The pRPS5A:H2B-tdTomato pACA3:Lyn24-mNeonGreen double marker has been previously described (Motomura et al., 2021). The lorelei-4 (lre-4) mutant (SAIL_8_D07) was provided by Dr. Palanivelu (Univ. Arizona) (Tsukamoto et al., 2010). The double mutant of anxur1-1 (anx1-1) and anx2-2 was provided by Dr. Miyazaki through Dr. Mizuta (Nagoya Univ.) (anx1-1: SALK_016179, anx2-2: SALK_133057). Seeds were germinated on Murashige-Skoog medium and transferred to soil. Plants were grown at 22°C under continuous lighting conditions.

Mature pollen grains and lre-4 ovules were mounted in 5% sucrose solution. Samples were observed using a confocal laser-scanning microscope equipped with a 63× objective lens (Leica TCS SP8; Leica, Wetzlar, Germany). The mNeonGreen was excited at 506 nm laser irradiation of the white light laser, and the fluorescence was detected using a Hybrid Detector (HyD) with a range of 511–553 nm. The tdTomato was excited at 554 or 558 nm laser irradiation of the white light laser, and the fluorescence was detected using an HyD with a range of 559–670 nm. Time-lapse images of the semi-in vivo fertilization assay was acquired every minute using the G/R split mode of an inverted microscope (IX-73; Olympus, Tokyo, Japan) equipped with a 60× objective lens, a spinning disk confocal scanning unit (CSU-W1; Yokogawa, Tokyo, Japan), and an sCMOS camera (Zyla 4.2; Andor, Belfast, Northern Ireland).

The pollen grains were placed on pollen tube growth medium (0.01% boric acid, 5 mM CaCl₂, 5 mM KCl, 1 mM MgSO₄, 10% sucrose, 10 μM epibrassinolide, 1.5% NuSieve GTG agarose, pH7.5) solidified on a glass bottom dish (D11130H; Matsunami Glass, Osaka, Japan). Pollen tube rupture was observed using the spinning disk confocal microscopy during incubation at 22°C.

Semi-in vivo fertilization assays were performed as previously described with minor modifications (Susaki et al., 2017). Emasculated WT pistils were cut beneath the style using a 27-gauged needle. The pistils were placed on pollen tube growth medium (0.001% boric acid, 1.27 mM Ca (NO₃)₂, 0.4 mM MgSO₄, 14% sucrose, 10 μM epibrassinolide, 1.5% NuSieve GTG agarose, pH7.0) solidified on a glass bottom dish (D11130H; Matsunami Glass, Osaka, Japan). Afterward, the stigma was pollinated with pollen from the pRPS5A:H2B-tdTomato pACA3:Lyn24-mNeonGreen plant. WT ovules were dissected from the emasculated pistils and aligned in front of the pollinated style. After 3 h of incubation at 22°C, pollen tubes reaching the ovules were observed using spinning disk confocal microscopy.

IVPM is supposed to dissociate from sperm cells to allow quick gamete associations after discharge; although, membrane dynamics remain elusive (Sprunck, 2020). Through in vitro and semi-in vivo assays, we observed immediate IVPM fragmentation after sperm cell release. IVPM breakdown displayed good contrast to IVPM robustness before pollen tube discharge. These observations highlighted the rapid transition in IVPM stability.

WT pollen tubes maintained an eyeglass-like IVPM morphology throughout the growth phase ( Figures 1G ). In the pollen tube, the MGU is continuously exposed to mechanical stresses: stretching, twisting, and moving back and forth in intense protoplasmic flow (Schattner et al., 2020). The MGU structure, particularly the tail-like connection, becomes fragile after reducing cellular stiffness by disrupting the coordination of motility between a pair of sperm cells and the vegetative nucleus (Zhou and Meier, 2014; Motomura et al., 2021). We assume that IVPM has sophisticated homeostasis systems that maintain membrane integrity to resist excess mechanical forces and spontaneous membrane breaks. For example, the ESCRT-III complex is a good candidate for controlling IVPM integrity because it promotes membrane repair in a variety of organelles (Gatta and Carlton, 2019; Isono, 2021). The coordination of lipid biosynthesis and proper membrane trafficking would also contribute to IVPM integrity by maintaining the tight membrane association between IVPM and sperm plasma membrane.

Released sperm cells immediately showed IVPM fragmentation and separated from each other ( Figures 1C, D ). Such a finding was reported in earlier studies, where sperm cells were isolated from mature pollen grains or pollen tubes (Russell, 1986; Zhang et al., 1992; Borges et al., 2008; Russell et al., 2017). It is still unclear whether these in vitro sperm cell separations reflect native IVPM breakdown because mechanistic damage by homogenization or osmotic shock could induce artificial IVPM removal during the isolation procedures. By contrast, we directly observed IVPM breakdown using systems closer to the in vivo conditions, which clarified the timing of IVPM breakdown unanswered by the analyses of pollinated pistil in Plumbago and maize (Russell, 1983; Gilles et al., 2021). According to the original study of Arabidopsis imaging of double fertilization, gamete mergers occur within 7.4 ± 3.3 min after pollen tube discharge (Hamamura et al., 2011). During a short period, sperm cells must complete various events, including fine adjustment of sperm cell position (Huang et al., 2015) and consecutive interactions of male and female gametes, including adhesion, recognition, and fusion (Igawa et al., 2013; Sprunck, 2020). Furthermore, recent studies have depicted a scenario of sperm cell activation prior to gamete fusion, in which egg cell-secreted EC1 family peptides trigger the relocation of the bona fide fusogenic factor GCS1-HAP2 on the surface of the sperm plasma membrane in a tetra-spanning membrane proteins DMP8 and DMP9 dependent manner (Sprunck et al., 2012; Cyprys et al., 2019; Wang et al., 2022). Rapid IVPM breakdown could be the initial step in sperm cell activation, which exposes the sperm cell surface and allows EC1-mediated activation and subsequent gamete interactions. Thus, analyzing the two opposing states of IVPM ( Figure 2 )—stable and vulnerable—and understanding the transition mechanisms of IVPM stability will become important issues for future research on double fertilization.

Plants must control the timing of IVPM breakdown because even a slight delay or precocious IVPM disappearance from sperm cells could compromise double fertilization and reduce reproductive fitness. Abiotic stimuli upon pollen tube discharge, including osmotic shock or abrasion on cell wall surface, may explain such a timing control. Intact sperm cell pairs were obtained from various species by employing moderate isolation protocols; some studies detected IVPM capsule (Taylor et al., 1991; Southworth and Kwiatkowski, 1996; Sprunck et al., 2012; Gilles et al., 2021). IVPM behavior in anx1/2 mutant was also consistent with the abiotic stimulation hypothesis ( Figures 1C ). However, these ex vivo releases would provide non-native physiological conditions to the sperm cells. Moreover, this hypothesis does not explain well the distinctive selectivity of IVPM breakdown discussed below. Hence, current data cannot exclude the relevance of other signaling cues at peak during intimate male–female interaction.

Putative signals that trigger IVPM breakdown should be temporarily elevated at the time of pollen tube rupture. Pollen tube cell wall integrity is regulated using an autocrine system, in which CrRLK1L-type receptors, ANX1, ANX2, BUDDHA’S PAPER SEAL 1 (BUPS1), and BUPS2 receive RAPID ALKALINIZATION FACTOR 4 (RALF4) and RALF19 peptides at the pollen tube tips (Boisson-Dernier et al., 2009; Miyazaki et al., 2009; Ge et al., 2017; Mecchia et al., 2017; Zhou et al., 2021). A competitive RALF34 ligand expressed in the ovules disrupts cell wall integrity before tube discharge. However, there is little evidence that the maximum signal of RALF34 can accurately determine the timing of IVPM breakdown via the ANX/BUPS pathway. Because anx1/2 mutant did not show IVPM fragmentation before MGU release, the ANX/BUPS pathway unlikely determines the timing of IVPM breakdown ( Figures 1C , Supplementary Figure 1 ).

Inorganic environmental changes surrounding the MGU could possibly trigger IVPM breakdown. Reactive oxygen species (ROS), most likely nitric oxide, are produced at high concentrations in synergid cells; lre mutants exhibit lower ROS levels than that of the WT (Duan et al., 2014; Duan et al., 2020). The characteristic pollen tube overgrowth phenotype in lre mutant ovules can be phenocopied via scavenger treatment in WT pistils, suggesting an important role of ROS in pollen tube rupture (Duan et al., 2014). However, these observations are insufficient to predict the ROS-dependent IVPM breakdown mechanism, as dynamic changes in ROS levels during fertilization remain to be analyzed. If synergid cells exhibited a clear oxidative burst upon pollen tube reception, highly membrane-permeable ROS would be an ideal intercellular signal controlling the delicate spatiotemporal activation of sperm cells and/or IVPM breakdown.

Compared with other signaling molecules, the dynamics of signal change were best characterized in cytoplasmic Ca²⁺ during double fertilization. In the time-lapse imaging of semi-in vivo fertilization assay using genetically encoded Ca²⁺ sensors, two synergid cells began to show Ca²⁺ oscillation immediately after pollen tube arrival (Ngo et al., 2014). The synergid cells and pollen tube produce a steep cytoplasmic Ca²⁺ elevation upon pollen tube rupture (Iwano et al., 2012; Ngo et al., 2014). Moreover, Ca²⁺ is considered as a putative gamete activator because Ca²⁺ spikes were also observed in the egg and central cells upon pollen tube discharge (Denninger et al., 2014; Hamamura et al., 2014). However, the inhibition of Ca²⁺ influx severely affects pollen tube growth and discharge (Duan et al., 2014). Simple pharmacological and genetic approaches are not applicable to semi-in vivo fertilization assays. Thus, decoding the physiological role of Ca²⁺ spikes in IVPM breakdown would be challenging without considerable technical advances in the future.

One of the most striking features of IVPM breakdown is its high membrane selectivity: IVPM breakdown must occur without damaging the sperm plasma membrane closely apposed with IVPM until pollen tube rupture. This selectivity may be because of its unique phospholipid composition: enriched phosphatidylserine and phosphatidylinositol 4,5-bisphosphate (Gilles et al., 2021). Anionic phospholipids are also enriched in the tips of pollen tube plasma membrane and play an important role in pollen tube growth (Helling et al., 2006; Zhou et al., 2020). It is interesting if membrane-degenerating factors are recruited to the anionic phospholipid–rich membrane domains and induce selective and simultaneous membrane rupture at the pollen tube tip and IVPM. Among the limited number of IVPM proteins reported recently, Rho of Plant (ROP) proteins and MTL/ZmPLA1/NLD exhibit peripheral membrane localization via post-translational protein modifications such as palmitoylation or myristoylation (Li et al., 2013; Gilles et al., 2021). MTL/ZmPLA1/NLD also contains positively charged amino acids around the acyl modification motif, which reinforce the association with negatively charged phospholipids in IVPM. Understanding the basis of the protein-localization mechanism of IVPM would be an important clue to narrowing down the candidates for putative membrane-degenerating factors.

s the correct subcellular localization of the MTL/ZmPLA1/NLD, an agronomically important haploid inducer protein, has been identified as IVPM, the unique sperm-enclosing membrane has come into the spotlight (Gilles et al., 2021). Although haploid induction in the mtl/Zmpla1/nld mutant is likely to be caused by sperm cell damage due to elevated ROS production, the importance of rapid and selective elimination of IVPM is evident in sexual reproduction of flowering plants (Jiang et al., 2022). Our current knowledge of the unique phospholipid composition and specifically localized proteins is indispensable for developing a novel experimental system to analyze IVPM breakdown. For example, IVPM markers are necessary to establish an efficient fractionation protocol to isolate intact IVPM with the entire MGU structure. Intact MGU can be used in an in vitro assay for IVPM breakdown and identification of novel IVPM-specific proteins through mass spectrometry analysis. This progress will lead to more substantial studies on the selective IVPM breakdown mechanism and its triggering signal that culminates in pollen tube rupture.