Evolutionarily conserved sperm factors, DCST1 and DCST2, are required for gamete fusion

To trigger gamete fusion, spermatozoa need to activate the molecular machinery in which sperm IZUMO1 and oocyte JUNO (IZUMO1R) interaction plays a critical role in mammals. Although a set of factors involved in this process has recently been identified, no common factor that can function in both vertebrates and invertebrates has yet been reported. Here, we first demonstrate that the evolutionarily conserved factors dendrocyte expressed seven transmembrane protein domain-containing 1 (DCST1) and dendrocyte expressed seven transmembrane protein domain-containing 2 (DCST2) are essential for sperm–egg fusion in mice, as proven by gene disruption and complementation experiments. We also found that the protein stability of another gamete fusion-related sperm factor, SPACA6, is differently regulated by DCST1/2 and IZUMO1. Thus, we suggest that spermatozoa ensure proper fertilization in mammals by integrating various molecular pathways, including an evolutionarily conserved system that has developed as a result of nearly one billion years of evolution.

Here, we show that an osteoclast fusion-related factor, dendrocyte expressed seven transmembrane protein (DC-STAMP) (Kukita et al., 2004;Yagi et al., 2005), homologue DC-STAMP domaincontaining 1 (DCST1), and its paralogue DCST2, which show testis and haploid-specific expression ( Figure 1D), are indispensable for fertilization in mice. Mouse DCST1 is considered to be an orthologue of Caenorhabditis elegans spermatogenesis-defective 49 (SPE-49) (Wilson et al., 2018) and Drosophila SNEAKY (Wilson et al., 2006), whereas mouse DCST2 is considered to be an orthologue of C. elegans SPE-42 (Kroft et al., 2005) and Drosophila DCST2, although the numbers of putative transmembrane regions appear different ( Figure 1A,B). Previously, these factors have been shown to be essential for fertilization. In fact, phylogeny analysis revealed that DCST1 and DCST2 belong  human  to different clades ( Figure 1C). This supports the finding that DCST2 appears to be closer to ancestor molecules than DCST1, because DCST1 is considered to have arisen after branching from DCST2 ( Figure 1C). For the first time, we here provide evidence that a common set of related factors between vertebrates and invertebrates actively participates in fertilization and appears to promote gamete merging in mammals.

Results and discussion
Since Dcst1 and Dcst2 are transcribed in different directions so as to face each other (Figure 2A), we first produced Dcst1/2 doubly-disrupted mouse lines using the homologous recombination system in embryonic stem cells, replacing Dcst1 exon 1-3 and Dcst2 exon 1-4, including the promoter region with the neomycin-resistance gene (Neo r ) (Figure 2A,B). The Dcst1/2 -/male mice, but not female, became completely infertile ( Figure 3E). When eggs were collected from the oviducts of the female mice 8 hr after coitus with a Dcst1/2 -/male, many zona-penetrated spermatozoa that had not been fertilized, and possessed a normal IZUMO1 localization upon acrosome reaction, were seen in the perivitelline space of oocytes ( Figure 3A). There was no difference in sperm motility between the Dcst1/2 -/and wild-type mice (motility was measured 120 min after incubation by computer-assisted sperm motility analysis in CEROS; mean ± s.e.m = 97.0 ± 0.73% in wild-type [1522 spermatozoa, six individuals] and 96.8 ± 0.87% in Dcst1/2 -/-[1425 spermatozoa, six individuals]). Therefore, it is reasonable to assume that there are no disturbances in sperm migration into the oviduct, acrosome reaction, and zona penetration ( Figure 3A). More precisely, gamete fusion assay showed that Dcst1/2 -/spermatozoa were capable of binding to the plasma membranes of oocytes, but no Dcst1/2 -/spermatozoa had fused with the oocytes ( Figure 3B). Since the syngamy was restored by bypassing with intracytoplasmic sperm injection (ICSI) using Dcst1/2 -/spermatozoa (numbers of pups born from Dcst1/2 +/and Dcst1/2 -/spermatozoa per survived transplanted 2 cell embryos were 8/27 [30%] and 17/48 [35%], respectively), DCST1/2 should be an essential part of gamete fusion machinery. Furthermore, the oocyte fusion-related factors JUNO and CD9 were found to be concentrated in regions, at which acrosome-reacted (AR) spermatozoa from Dcst1/2-deficient mice labeled with a-IZUMO1 antibody were bound, same as wild-type spermatozoa ( Figure 3C, Inoue et al., 2020), suggesting that Dcst1/2 -/spermatozoa are capable of recruiting the oocyte factors at the contact site; however, these spermatozoa fail to proceed to membrane fusion with oocytes.
In order to know which factor plays a more crucial role in fertilization, we next produced solo Dcst1 -/and Dcst2 -/mouse lines using CRISPR/Cas9-mediated gene disruption, and assessed male fertilization ability simultaneously with performed transgenic rescued experiments (Figure 2A,B). In in vitro fertilization (IVF), as expected, all of the Dcst1/2 -/-, Dcst1 -/and Dcst2 -/spermatozoa failed to fertilize the eggs, where the average fertilization rates of Dcst1/2 -/-, Dcst12 -/-, Dcst2 -/spermatozoa were 0%, 3.3%, and 0.8%, respectively ( Figure 3D). The impaired fertilization step undoubtedly followed zona penetration, because motile Dcst1/2-deficient spermatozoa penetrated the zona pellucida and accumulated in the perivitelline space of the oocytes (Video 1). It is noteworthy that the spermatozoa from the single gene transgenic rescued mice (Dcst1-TG or Dcst2-TG) with Dcst1/2 knockout genetic backgrounds were unsuccessful in fertilization; however, the spermatozoa from the double transgenic mice (Dcst1/2 -/-Dcst1/2-TG) had normal fertility compared to that of the Dcst1/ 2 +/mice ( Figure 3D). Regarding the numbers of offspring, which is the final outcome of fertility, all Dcst1/2 -/-, Dcst1 -/-, and Dcst2 -/males were completely sterile despite normal mating behavior with ejaculation and vaginal plug formation, whereas the double transgenic rescued males (Dcst1/

S p a c a 6 -/ -W i l d -t y p e I z u m o 1 -/ -D c s t 1 -/ -D c s t 2 -/ -D c s t 1 / 2 -/ -I z u m o 1 -/ -I z u m o 1 -T G D c s t 1 / 2 -/ -D c s t 1 / 2 -T G S p a c a 6 -/ -W i l d -t y p e I z u m o 1 -/ -D c s t 1 -/ -D c s t 2 -/ -D c s t 1 / 2 -/ -I z u m o 1 -/ -I z u m o 1 -T G D c s t 1 / 2 -/ -D c s t 1 / 2 -T G Izumo1
Spaca6 Dcst1 Dcst2 PA-tag B Figure 2. Strategy for Dcst1 and Dcst2 genes disruption and its validation. (A) Targeted disruption of the Dcst1/2, Dcst1, Dcst2, and Spaca6 genes, and constructs for Dcst1 and Dcst2 transgenic mice. Complete structures of the wild-type mouse Dcst1/2 alleles are shown. Exons and introns are represented by boxes and horizontal lines, respectively. An open reading frame for Dcst1 and Dcst2 genes is shown in black. For the targeted disruption of mouse Dcst1/2, the neomycin-resistance gene (Neo r ) was inserted between Dcst1 exon 3 and Dcst2 exon 4. A herpes simplex virus thymidine kinase gene was introduced into the targeting construct for negative selection. For the gene complementation experiments, transgenic mouse lines expressing dendrocyte expressed seven transmembrane protein domain-containing 1 (DCST1), in which full-length genome DNA fragment including all promoter region, exons, and introns was incorporated, and dendrocyte expressed seven transmembrane protein domain-containing 2 (DCST2), in which full-length cDNA with a PA-tag sequence was inserted between a testis-specific (pachytene spermatocyte to spermatid stage) Figure 2 continued on next page 2 -/-Dcst1/2-TG) showed normal litter sizes, consistent with IVF ( Figure 3D,E). From these results, we conclude that both DCST1 and DCST2 are required for establishing the gamete fusion leading to embryogenesis.
Finally, we investigated a protein profile of sperm factors such as IZUMO1 and SPACA6, which are essential factors for gamete recognition and fusion (Figure 2A,C; Barbaux et al., 2020;Inoue et al., 2005;Lorenzetti et al., 2014;Noda et al., 2020). As a result, in Izumo1, Spaca6, Dcst1/2, Dcst1, and Dcst2-deficient spermatozoa, SPACA6 typically disappears from the mature spermatozoa, whereas the IZUMO1 protein remains, except in Izumo1-null spermatozoa ( Figure 3F). However, loss of SPACA6 was recovered by Izumo1 complementation (Izumo1 -/-Izumo1-TG) ( Figure 3F). Thus, IZUMO1 and SPACA6 are likely to be cooperative factors. Unexpectedly, although Dcst1/2-deficient male fertility could be restored in the Dcst1/2 transgenic rescued mice ( Figure 3D,E), retrieval of SPACA6 was not observed ( Figure 3F). While we do not have the data to explain this, the result implies that SPACA6 seems to be dispensable for fertilization in Dcst1/ 2 -/-Dcst1/2-TG spermatozoa. In this context, it is interesting that DCST1 has a ubiquitin ligase activity (Nair et al., 2016). It is conceivable that precise temporal expression of putative substrates negatively regulated by ubiquitin-mediated degradation by DCST1/2 during spermatogenesis has to be required for SPACA6 protein stability, which could have been impaired in the Dcst1/2 complementation.
It should be emphasized that DCST1/2 are the first identified well-conserved factors from human to nematode or fruit fly that are actively involved in gamete recognition and fusion. Although the detailed molecular mechanisms underlining gamete recognition and fusion are still unknown, new insight into molecular behavior and interaction through DCST1 and DCST2 involvement would greatly advance our understanding of elaborate common molecular mechanisms in mysterious fertilization in a sexually reproducing organism. Continued on next page Calmegin promoter and rabbit b-globin polyadenylation signal, were produced. For genome-editing of Dcst1, Dcst2, and Spaca6 using the CRISPR/ Cas9 system, the red and blue letters show the target and protospacer adjacent motif (PAM) sequence, respectively (left panels). A DNA sequence chromatogram shows that Dcst1 -/-, Dcst2 -/-, and Spaca6 -/had 5-base, 5-base, and 28-base deletions, respectively. The deleted sequences are boxed in red (right panels). (B) Validation of gene disruption by reverse transcription polymerase chain reaction (RT-PCR). All primer sets employed wild-type mRNA-specific oligonucleotide sequences except for PA-tag. As a result, all gene disruptions and transgenes were confirmed appropriately. b-actin mRNA was used as an internal control. (C) In vitro fertilization analysis using Spaca6 +/and Spaca6 -/spermatozoa (n = 4 and 4, respectively). The fertilization rate was evaluated at the two-cell stage embryo. The numbers in parentheses indicate the numbers of oocytes used. The error bars represent standard error of the mean (s.e.m.). The online version of this article includes the following source data for figure 2: Mice C57BL/6, B6D2F1, and ICR mice (Mus musculus) were purchased from Japan SLC Inc, and IZUMO1 knockout and transgenic mice (Inoue et al., 2005) were kindly provided by Osaka University. All animal studies were approved by the Animal Care and Use Committee of Fukushima Medical University, Japan, and performed under the guidelines and regulations of Fukushima Medical University.

Generation of Dcst1/2 double knockout mice
A targeting vector was constructed using modified pNT1.1 containing neomycin-resistance gene as a positive selection marker and herpes simplex virus thymidine kinase gene as a negative selection marker (http://www.ncbi.nlm.nih.gov/nuccore/JN935771). A 1.7 kb Not I-Xho I fragment and a 6.8 kb Pac I-Mfe I fragment were inserted as short and long arms, respectively. Embryonic stem cells derived from 129/Sv were electroporated with Cla I digested linearized DNA. Of the 192 G418-resistant clones, five had undergone homologous recombination correctly. Two targeted cell lines were injected into C57BL/6 mice derived blastocysts, resulting in the birth of male chimeric mice. These mice were then crossed with C57BL/6 to obtain heterozygous mutants. The mice used in the present study were the offspring of crosses between F1 and/or F2 generations. The PCR primers used for genotyping were as follows: 5'-A TTCTTCCTCTCCCTTTCGGACATC-3' and 5'-GC TTGCCGAATATCATGGTGGAAAATGGCC-3' for the short arm side of the mutant allele; 5'-ATTC TTCCTCTCCCTTTCGGACATC-3' and 5'-CCAGGGGATCCAACACCCTT-3' for the short arm side of the wild-type allele; 5'-TCTGTTG TGCCCAGTCATAGCCGAATAGCC-3' and 5'-C TGGAAGTGTCTTCCGAGTGCCACTCCACA-3' for the long arm side of the mutant allele; and 5'-CCCTAGGCTGTGAAGTGTTGATGAG-3' and 5'-CTGGAAGTGTCTTCCGAGTGCCACTCCACA-3' for the long arm side of the wild-type allele. The Dcst1/2-disrupted mouse line was submitted to RIKEN BioResource Research Center (https:// web.brc.riken.jp/en/), and is available to the scientific community (accession number: RBRC05733).

Production of DCST1 and DCST2 transgenic mice
Dcst1 gene fragments were amplified by PCR using a bacterial artificial chromosome (BAC) DNA clone (RP24-185C5) as template DNA. The DNA fragments were ligated into StrataClone PCR Vector pSC-B-amp/kan (Agilent Technologies). Until injection, a 17.5 kb DNA fragment that included whole Dcst1 gene was digested with Not I and Sal I (Figure 2A). For DCST2, we designed PAtagged Dsct2, to be inserted between the Calmegin promoter (testis specific promoter) and a rabbit b-globin polyadenylation signal (Figure 2A). The transgenic mouse lines were produced by injecting Dcst1 gene or Dsct2-PA DNA fragments into the pronuclei of fertilized eggs. For the detection of transgene, PCR was performed on tail-tip DNA using following primer sets: 5'-GGCCGCTCTAGAAC TAGTGGAT-3' and 5'-GGTGTTTGTGTTACTGAAGTCACTG-3' for DCST1-TG specific amplification, Video 1. 6 hr after insemination of Dcst1/2-disrupted spermatozoa. The first and second half of the video show Dcst1/2 heterozygous and homozygous KO spermatozoa insemination, respectively. https://elifesciences.org/articles/66313#video1 and 5'-CCTTCCTGCGGCTTGTTCTCT-3' and 5'-GGCTATGTGTTTCACGGCAT-3' for DCST2-TG specific amplification.

Gamete fusion assay
Superovulation was induced in each B6D2F1 female mouse (>8 weeks old) by injecting 7.5 IU of equine chorionic gonadotropin (eCG) and 7.5 IU of human chorionic gonadotropin (hCG) at 48 hr intervals. Oocytes were collected from the oviduct 16 hr after hCG injection, and placed in 50 ml of Toyoda, Yokoyama and Hoshi (TYH) medium. The zona pellucida was removed from the oocytes via treatment with 1.0 mg ml À1 of collagenase (FUJIFILM Wako Pure Chemical Corporation) (Yamatoya et al., 2011). They were preloaded with 1 mg ml À1 Hoechst 33342 (Thermo Fisher Scientific) in TYH medium for 10 min, and were washed prior to the addition of the spermatozoon. Dcst1/ 2 -/spermatozoa were collected from the cauda epididymis, and capacitated in vitro for 2 hr in a 200 ml drop of TYH medium. Zona-free oocytes were co-incubated with 2 Â 10 5 mouse spermatozoa ml À1 for 30 min at 37˚C in 5% CO 2 . After 30 min of incubation, the eggs were observed under a fluorescence microscope after being fixed with 0.25% glutaraldehyde in flushing holding medium (FHM) medium. This procedure enabled staining of fused sperm nuclei only, by transferring the dye into spermatozoa after membrane fusion.

Fluorescence imaging
Zona-free oocytes were prepared as described above. They were incubated with 0.5 mg ml À1 MZ-FITC and 0.25 mg ml À1 TH6-Alexa647 for 1 hr at 37˚C in TYH medium. Dcst1/2 -/spermatozoa were collected from the cauda epididymis, and capacitated in vitro for 2 hr in a 200 ml drop of TYH medium with 0.5 mg ml À1 Mab125-Alexa546 that was covered with mineral oil (Merck). Then, 2 Â 10 5 spermatozoa ml À1 was inseminated in the oocytes for 30 min at 37˚C with 5% CO 2 . After the eggs were washed three times in TYH medium by transferring spots of oocytes containing media, the eggs were fixed with 4% paraformaldehyde and 0.5% polyvinylpyrrolidone (PVP) in phosphate buffered saline (PBS) for 30 min at room temperature, then observed in FHM medium on the glass-bottom dishes (No. 0, MatTek). A 100Â oil-immersion objective (numeric aperture 1.49) was used to capture confocal images with an A1R microscope (Nikon). The pinhole was set at 3.0 airy unit. For 3D reconstruction, 100-120 fluorescent images were taken at 1 mm intervals on the Z axis, and then 3D images were reconstructed using the built-in software NIS-Elements ver. 4.3 (Nikon).

In vitro fertilization
Mouse spermatozoa were collected from the cauda epididymis, and were capacitated in vitro for 2 hr in a 200 ml drop of TYH medium that was covered with mineral oil. Superovulation was induced in female mouse, as described above. Oocytes were collected from the oviduct placed in 100 ml of TYH medium, and incubated with 2 Â 10 5 spermatozoa ml À1 at 37˚C with 5% CO 2 . After 5 hr of insemination, the oocytes were washed with TYH medium and the fertilization rate was evaluated at the two-cell stage embryo.