DDI-4, but neither Pron nor ProK, triggers temperature-sensitive MOF in C. elegans spermatids

In a recent study, we screened the Core Library provided by the Drug Discovery Initiative (DDI; University of Tokyo) using an in vitro C. elegans spermiogenesis assay [13]. This screening identified one positive compound, DDI-4, which exhibited higher activity than the quinolinol analog DDI-6, previously reported as a dual activator of MOF and ASR [13]. Thus, we further examined how DDI-4 triggers MOF in C. elegans spermatids. Although C. elegans spermiogenesis is characterized by pseudopod extension from spermatids accompanied by exocytotic remodeling, MOF was assessed exclusively based on FM1-43 staining rather than sperm morphology.

DDI-4 is a benzylamine analog (1-(4-n-pentylphenyl)ethan-1-amine; Fig 1A) with previously unknown biological activity. As shown in Fig 1B and 1C, under optimized assay conditions, MOF induction by DDI-4 reached a plateau at 100 µM and after 15 min of incubation, respectively. Then, we compared MOF-inducing activities of DDI-4 and its analogs (DDI-4A, 1-(4-ethylphenyl)ethan-1-amine; DDI-4B, (S)-(-)-1-phenylethylamine; and DDI-4C, (R)-(-)-1-phenylethylamine). Only DDI-4 triggered MOF in spermatids, suggesting that a specific alkyl chain length linked to the phenyl group is required for the MOF-inducing activity (Fig 1D).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. DDI-4 is a dual activator of nematode MOF and mouse ASR.

(A) Chemical structures of DDI-4 and its analogs. (B) Concentration-dependent effect of DDI-4 on MOF. ΔMOF-positive ratios were calculated by subtracting the baseline value obtained with 0.5% dimethyl sulfoxide (DMSO) from those obtained with each DDI-4 concentration in 0.5% DMSO. Net ratios are shown as histograms with the mean ± standard error of the mean (SEM; n = 3). (C) Incubation period–dependent effect of DDI-4 on MOF. The nematode MOF assay was performed using 0.5% DMSO or 100 μM DDI-4 as SAFs. At each time point, ΔMOF-positive ratios were calculated as described in panel (B) and are shown as histograms indicating the mean ± SEM (n = 3). (D) Comparison of DDI-4 and its analogs (DDI-4A, DDI-4B, and DDI-4C) in MOF-inducing activity. The nematode MOF assay was performed using 0.5% DMSO, 100 μM DDI-4, or 100 μM DDI-4 analogs. ΔMOF-positive ratios were calculated as described in panel (B) and are shown as histograms indicating the mean ± SEM (n = 3). (E) Temperature-sensitive MOF triggered by DDI-4. The nematode MOF assay was performed using 1x Sperm Medium (1x SM), 200 μg/mL Pronase (Pron), 200 μg/mL Proteinase K (ProK), 0.5% DMSO, or 100 μM DDI-4. 1 × SM was used as the solvent control for Pron and ProK, and DMSO was used as the solvent control for DDI-4. At each rearing temperature, based on the micrographs shown in S2 Fig, ΔMOF-positive ratios were calculated as described in panel (B) and are shown as histograms indicating the mean ± SEM (n = 3). P-values for differences in DDI-4–triggered ΔMOF-positive ratios between spermatids raised at 20°C and 25°C are indicated by asterisks. ***, P < 0.005. (F) Images of acrosome (AS)-intact and AS-reacted mouse spermatozoa after staining with Coomassie Brilliant Blue G-250. A red arrow indicates the AS region. Scale bar = 10 µm. (G) Concentration-dependent effect of DDI-4 on ASR. Data are shown as histograms indicating the mean ± SEM (n = 3). *, P < 0.05; ***, P < 0.005. (H) Incubation period–dependent effect of DDI-4 on ASR. The mouse ASR assay was performed using 0.1% DMSO or 200 μM DDI-4 as ASR inducers. *, P < 0.05; ***, P < 0.005. (I) Comparison of A23187 and DDI-4 in ASR-inducing activity. The mouse ASR assay was performed using 0.1% DMSO, 10 μM A23187, or 200 μM DDI-4. Data are shown as histograms indicating the mean ± SEM (n = 3). **, P < 0.01; ***, P < 0.005.

https://doi.org/10.1371/journal.pgen.1012275.g001

When spermatids from males raised at high temperature (25°C) were examined, the ΔMOF-positive ratio induced by DDI-4—calculated by subtracting the MOF-positive ratio of the control sample from that of the DDI-4–treated sample—was markedly reduced to 17.9%, compared with 96.7% in spermatids raised at 20°C (Figs S2 and 1E). In contrast, ΔMOF-positive ratios by Pron or ProK were unaffected by the tested temperatures (Figs S2 and 1E). These results indicate that DDI-4, but neither Pron nor ProK, triggers temperature-sensitive MOF.

DDI-4 activates the ASR and induces IZUMO1 relocation in mouse spermatozoa

We next tested whether DDI-4 activates the ASR in mouse spermatozoa (Fig 1F–1I). Acrosome (AS)-reacted spermatozoa were distinguished from AS-intact spermatozoa by staining acrosomal contents with Coomassie Brilliant Blue G-250 (Fig 1F). As in Fig 1B and 1C, the effects of DDI-4 concentration and incubation period on the ASR were examined; ASR-positive ratios approached a plateau at 200 µM DDI-4 (Fig 1G) or after 30 min of incubation (Fig 1H). Under these optimized conditions, ASR-positive ratios induced by DDI-4 were comparable to those induced by the calcium ionophore A23187, used as a positive control (Fig 1I).

We also examined IZUMO1 relocation during DDI-4–induced ASR by immunofluorescence (IF) using a rabbit anti-mouse IZUMO1 polyclonal antibody (α-IZUMO1) [14] (Fig 2, column 4). The same samples were counterstained with 4′,6-diamidino-2-phenylindole (DAPI) (Fig 2, column 2) and fluorescein isothiocyanate–conjugated peanut agglutinin (FITC-PNA) (Fig 2, column 3) to visualize the nucleus and the AS, respectively.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. DDI-4 triggers IZUMO1 translocation during the ASR in mouse spermatozoa.

Mouse spermatozoa were released from the cauda epididymis, incubated with 200 µM DDI-4, and stained with 5 µg/mL rabbit anti-mouse IZUMO1 antibody (α-IZUMO1). The sperm samples were also counterstained with 5 µg/mL 4′,6-diamidino-2-phenylindole (DAPI) and 5 µg/mL fluorescein isothiocyanate–labeled peanut agglutinin (FITC-PNA) to label the nucleus and the AS, respectively. We also used 3% bovine serum albumin (BSA) as a positive ASR inducer. PBS, phosphate-buffered saline; control IgG, normal rabbit IgG. Scale bar = 10 µm.

https://doi.org/10.1371/journal.pgen.1012275.g002

When AS-intact spermatozoa were mock-treated with phosphate-buffered saline (PBS) as a solvent control, FITC-PNA stained the AS region of the sperm head (Fig 2, panels a3 and b3), whereas α-IZUMO1 immunoreactivity was not detected (Fig 2, panels a4 and b4), indicating that neither the ASR nor IZUMO1 relocation occurred. In the presence of bovine serum albumin (BSA), a conventional ASR inducer, the FITC-derived signals were diminished or absent in many spermatozoa (Fig 2, panels c3 and d3), indicating ASR induction. Concomitantly, α-IZUMO1 immunoreactive signals were detected in a region of the sperm head corresponding to the equatorial segment (Fig 2, panels c4 and d4), indicating IZUMO1 relocation during the ASR.

IF patterns observed with dimethyl sulfoxide (DMSO) (Fig 2, panels e1–f4) and DDI-4 (Fig 2, panels g1–h4) were similar to those obtained with PBS (Fig 2, panels a1–b4) and BSA (Fig 2, panels c1–d4), respectively. DDI-4–treated spermatozoa underwent the ASR (Fig 2, panels g3 and h3) and exhibited IZUMO1 relocation to the equatorial segment (Fig 2, panels g4 and h4). Thus, DDI-4 activates the ASR, demonstrating that DDI-4 targets are present in mouse spermatozoa as in nematode spermatids.

DDI-4 activates an exocytotic sperm remodeling pathway distinguishable from canonical Pron- and ProK-mediated pathways

To elucidate the molecular basis of DDI-4–induced exocytotic sperm remodeling, we focused on the C. elegans system. Given that several genes involved in Pron- and/or ProK-mediated activation have been identified, we tested whether DDI-4 could activate spermatids carrying mutations in these genes. As shown in Fig 3A, previous studies have indicated that snf-10 [10,15] and zipt-7.1 [10,16] function in both the Pron and ProK pathways, whereas spe-8 class genes (spe-8 [9,10,17], spe-19 [10,18], spe-27 [10,19], and spe-29 [10,20]) act primarily in the Pron pathway. Upon treatment with Pron or ProK, these mutant spermatids fail to undergo normal pseudopod extension and to exhibit the wild-type pattern of FM1-43 staining characteristic of MOF [10]. Our results (Fig 3B and S1 Table) were largely consistent with these profiles across the tested temperatures, except for the zipt-7.1(ok971); him-5(e1490) males: spermatids raised at 25°C exhibited higher ΔMOF-positive ratios than those raised at 20°C.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 3. The DDI-4 pathway is distinct from the canonical exocytotic sperm remodeling pathways.

(A) Canonical pathways activated by Pron or ProK. (B) Comparison of Pron-, ProK-, and DDI-4–induced MOF in spermatids from males lacking genes involved in the Pron and/or ProK pathways, raised at 20°C or 25°C. ΔMOF-positive ratios for each tested activator were calculated as described in Fig 1 and are shown as color gradients. The values of each ΔMOF-positive ratio are shown in S1 Table.

https://doi.org/10.1371/journal.pgen.1012275.g003

DDI-4 induced MOF in spe-8(hc40), spe-19(eb52), and spe-27(it110) spermatids raised at 20°C but failed to activate the spe-27 mutant and significantly diminished activation of the spe-8 and spe-19 mutants when raised at 25°C (Fig 3B and S1 Table). him-8(e1489); snf-10(hc194) and spe-29(it127) spermatids exhibited moderate and minimal activation by DDI-4 at 20°C and 25°C, respectively (Fig 3B and S1 Table). In contrast, DDI-4–induced MOF was significantly reduced in zipt-7.1 mutant spermatids at both temperatures (Fig 3B and S1 Table). These results suggest that zipt-7.1 plays a significant role in DDI-4–induced MOF downstream of the DDI-4 target, whereas snf-10 and spe-29 likely contribute to lesser extents.

Furthermore, genetic comparisons of the activation profiles induced by Pron, ProK, and DDI-4 suggest that the spe-8 class genes act downstream of the Pron target but upstream of the DDI-4 target. In the ProK-activated pathway, snf-10 appears to function downstream of the ProK target but upstream of the DDI-4 target. Because Pron- and ProK-induced MOF was temperature-insensitive, unlike DDI-4–induced MOF (Figs S2 and 1E), the DDI-4–triggered pathway is likely at least partially distinct from those activated by Pron and ProK.

C. elegans tRNA methyltransferase NSUN-2 is required for both meiosis and DDI-4–induced MOF

To identify components involved in the DDI-4 pathway, we isolated the nyg20 mutant after introducing random mutations into the him-8(tm611) genome using ethyl methanesulfonate (EMS). As shown in Fig 4A, him-8(tm611); nyg20 males raised at 20°C produced spermatids that were Pron or ProK activated, but not by DDI-4, suggesting that this strain carries a mutation(s) affecting DDI-4–triggered MOF.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 4. DDI-4–induced MOF are dependent on nsun-2.

(A) MOF-inducing activity of DDI-4 in nyg20 spermatids. ΔMOF-positive ratios were calculated as described in Fig 1 and are shown as histograms indicating the mean ± SEM (n = 3). Differences in ΔMOF-positive ratios between him-8(tm611) (control) and him-8(tm611); nyg20 worms are indicated by asterisks. ***, P < 0.005. (B) Self-fertility of control and the nyg20 and nsun-2(nyg21) hermaphrodites raised at 20°C or 25°C. Brood sizes of each strain are shown as histograms indicating the mean ± SEM (n = 3). ***, P < 0.005. (C) Complementation analysis between the nyg20 and nsun-2(nyg21) worms. ΔMOF-positive ratios induced by DDI-4 were determined as described in Fig 1 and are shown as histograms indicating the mean ± SEM (n = 3). ***, P < 0.005. (D) Comparison of Pron-, ProK-, and DDI-4–induced MOF in spermatids from him-8(tm611) or nsun-2(nyg21); him-8(tm611) males raised at 20°C or 25°C. As described in Fig 1, ΔMOF-positive ratios were calculated, and statistical differences between net ratios are indicated by asterisks. **, P < 0.01; ***, P < 0.005.

https://doi.org/10.1371/journal.pgen.1012275.g004

Our single-nucleotide polymorphism (SNP) mapping (S3 Fig and S2–S3 Tables) localized the nyg20 mutation to a region near −6 cM on chromosome I and/or a region near −8 cM on chromosome X. During the SNP-mapping procedure, the nyg20 mutant was crossed with the Hawaiian strain, and progeny homozygous for both him-8(tm611) on chromosome IV and the nyg20 mutation were selected. Therefore, SNPs on chromosome IV were expected to be homozygous for the N2 allele in many of the analyzed clones. As part of our effort to identify the gene affected in nyg20, we focused on nsun-2, one of the genes located within this candidate interval, because it had previously been implicated in temperature-sensitive reproductive phenotypes. In C. elegans, NSUN-2 produced approximately 88% of total 5-methylcytosines (m⁵Cs) as a tRNA methyltransferase [21], and demethylation of these m⁵Cs by simultaneous disruption of all the nsun genes resulted in self-sterility at 27°C but not at 20°C [21]. These findings suggest that C. elegans nsun-2 contributes to temperature-sensitive reproduction. Thus, we hypothesized that a subset of spermatid proteins translationally regulated by NSUN-2 functions during temperature-sensitive, DDI-4–induced MOF.

Using CRISPR/Cas9, we generated a predicted loss-of-function strain of nsun-2 (nsun-2(nyg21); him-8(tm611)), carrying a premature stop codon in exon 2. As shown in Fig 4B, brood sizes of him-8(tm611), him-8(tm611); nyg20, and nsun-2(nyg21); him-8(tm611) hermaphrodites raised at 20°C were comparable (~190, ~ 170, and ~200 F₁ self-progeny per hermaphrodite, respectively). However, the nyg20 and nsun-2 mutants raised at 25°C exhibited markedly reduced brood sizes (~13 and ~34 F₁ self-progeny) compared with control worms (~77 F₁ self-progeny). This indicates that the nyg20 and nsun-2 mutants become severely sub-fertile at 25°C, but not at 20°C.

To determine whether nsun-2 is affected in nyg20, complementation analysis was performed using the nyg20 and nsun-2 mutants raised at 20°C (Fig 4C). Male spermatids from self- or outcross progeny were subjected to the nematode MOF assay using DDI-4 as a SAF. Male F₁ progeny from N2 (wild type), him-8(tm611) (control), him-8(tm611); nyg20, and nsun-2(nyg21); him-8(tm611) hermaphrodites produced spermatids exhibiting ΔMOF-positive ratios of 85.4, 92.1, 7.7, and 54.1%, respectively. Male outcross progeny heterozygous for the nyg20 or nsun-2 mutations displayed ΔMOF-positive ratios comparable to those of N2 and him-8(tm611) spermatids, respectively, indicating that both the nyg20 and nsun-2 mutations are recessive. Moreover, ΔMOF-positive ratios in male spermatids of outcross progeny from nyg20/nsun-2 worms were comparable to those of the nsun-2 mutant. Collectively, these results suggest that NSUN-2 function during DDI-4–induced MOF may be compromised in the nyg20 mutant, although no mutations were identified in the nsun-2 coding, intronic, and flanking regions by next-generation sequencing (NGS) or Sanger sequencing analyses (S4 Table).

In Nsun2-null mice [22], tRNA cleavage was increased, leading to reduced protein translation rates and activation of stress pathways. Because spermatogenesis is arrested at the spermatocyte stage in the Nsun2 mutant mice, we also examined meiosis in the gonads of nsun-2(nyg21); him-8(tm611) male worms (S4 Fig). In C. elegans males, meiosis proceeds from the distal to proximal ends of the gonad (S4A Fig; him-8(tm611) (control) at 20°C). The distal mitotic region harbors germline stem cells, whereas the adjacent transition zone marks the onset of meiosis, where chromosome condensation produces crescent-shaped nuclei characteristic of the leptotene/zygotene stages. In control males raised at 20°C, crescent-shaped nuclei were detected in the transition zone by DAPI staining (S4A Fig). In contrast, in the nsun-2 mutant, cells with crescent-shaped nuclei were mislocalized to the mitotic zone. Similar abnormalities were observed in the mitotic zone of control males raised at 25°C (S4A Fig). These findings suggest that male gonad meiosis is partially temperature-sensitive and nsun-2–dependent.

We also quantified the number of spermatids released from male gonads (S4B Fig). In single control males raised at 20°C, ~ 2,400 spermatids were obtained, whereas spermatid numbers were reduced by 42.2% and 32.9% in control males raised at 25°C and in nsun-2 mutant males raised at 20°C, respectively. In the nsun-2 mutant, rearing at 25°C further compromised spermatid production, with counts decreasing from ~1,600 cells at 20°C to ~1,000 cells. These findings suggest that meiosis is partially impaired by either elevated temperature or loss of NSUN-2 function. Intriguingly, the meiotic defects observed in the nsun-2 mutant were prominent in males (S4 Fig), whereas self-fertility was largely unaffected in hermaphrodites (Fig 4B). Although the basis for this apparent sexual dimorphism remains unclear, one possible interpretation is that spermatogenesis occurs only during the fourth larval (L4) stage in hermaphrodites [2,3], whereas we examined the male germlines derived from adult males undergoing continuous spermatogenesis. Thus, the requirement for NSUN-2 may vary with the developmental stage or physiological context of the germline.

When spermatids from control and nsun-2 mutant males were compared (Fig 4D), Pron or ProK treatment did not alter ΔMOF-positive ratios, regardless of genotype or rearing temperature. In contrast, the ratios of DDI-4–induced MOF-positive cells were significantly reduced by either the nsun-2 mutation or elevated rearing temperature. These results indicate that nsun-2 is required for temperature-sensitive MOF activation.

DDI-4 is predicted to target sphingosine kinases (SPHKs)

To understand how DDI-4 triggers MOF, it is essential to identify DDI-4 targets. In silico analysis using the Similarity Ensemble Approach (SEA; https://sea.bkslab.org/) [23] predicted that DDI-4 binds SPHKs (S5 Table). In vertebrates, SPHKs exist as two subtypes—SPHK1 and SPHK2 [24–27]. In contrast, C. elegans possesses a single SPHK homolog, SPHK-1. We conducted a phylogenetic analysis to infer which vertebrate subtype the C. elegans SPHK-1 is most closely related to. Fig 5A shows a phylogenetic tree of SPHK1 and SPHK2 proteins from various species, revealing two major vertebrate clades corresponding to SPHK1 and SPHK2. In contrast, invertebrate SPHKs (nematode SPHK-1 and fruit fly SPHK1/SPHK2) branched outside these clades, forming a distinct basal group that likely represents an ancestral SPHK lineage predating vertebrate gene duplication. Thus, this phylogenetic tree does not clearly classify nematode SPHK-1 as either vertebrate SPHK1 or SPHK2.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 5. sphk-1, but not spin-4, is involved in DDI-4–induced MOF.

(A) Phylogenetic analysis of sphingosine kinase (SPHK) homologs. A molecular phylogenetic tree was constructed using the neighbor-joining method implemented in MEGA11 (https://www.megasoftware.net/). The analysis was based on amino acid sequences of SPHKs retrieved from UniProt (https://www.uniprot.org/) and identified by their accession IDs: human SPHK1 (human 1; Q9NYA1); human SPHK2 (human 2; Q9NRA0); mouse SPHK1 (mouse 2; Q8CI15); mouse SPHK2 (mouse 2; Q9JIA7), chicken SPHK1 (chicken 1; A0A1D5P1F1), frog SPHK1 (frog 1; F7CBM1); frog SPHK2 (frog 2; A0A803JMT9); fruit fly SPHK1 (fruit fly 1; Q9VYY8); fruit fly SPHK2 (fruit fly 2; Q9VZW0); nematode SPHK-1 (nematode 1; Q18425). (B) Docking models of human SPHK1 (HuSPHK1) and nematode SPHK-1 (NeSPHK-1) with ATP and SPH or DDI-4. Note that Phe173 is obscured in the docking models of HuSPHK1, although this residue was predicted to participate in interactions with both SPH and DDI-4 (S6 Table). Phe173 is not visible in panel (B) because it is located behind the displayed structure. (C) Self-fertility of wild-type (N2), sphk-1(ok1097), and spin-4(knu1099) hermaphrodites raised at 20°C or 25°C. Brood sizes of each strain are shown as histograms indicating the mean ± SEM (n = 3). ***, P < 0.005. (D) Comparison of Pron-, ProK-, and DDI-4–induced MOF in spermatids from N2, sphk-1(ok1097), or spin-4(knu1099) males raised at 20°C or 25°C. As described in Fig 1, ΔMOF-positive ratios were calculated, and statistical differences in ΔMOF-positive ratios are indicated by asterisks. *, P < 0.05; ***, P < 0.005.

https://doi.org/10.1371/journal.pgen.1012275.g005

Next, based on X-ray crystallographic data of human SPHK1 in complex with ATP and sphingosine (SPH) [28,29], we constructed docking models of ATP and DDI-4 with human SPHK1 or nematode SPHK-1 (Fig 5B). These models revealed that the binding mode of DDI-4 to nematode SPHK-1 closely resembles that to human SPHK1. We predicted amino acid residues in human SPHK1 and nematode SPHK-1 that may mediate conserved interactions with both SPH and DDI-4 by the Protein-Ligand Interaction Profiler (PLIP) [30]. As shown in S6 Table, Phe173, Ile174, Val177, and Phe303 in human SPHK1 and Phe266 and Ser241 in nematode SPHK-1 appear to be involved in interactions with both SPH and DDI-4. Additionally, these models also predict that the alkyl chain length of DDI-4 is insufficient to properly fit the SPHK binding pocket, which may explain the relatively high DDI-4 concentration required to trigger MOF (Fig 1B).

We further examined whether the C. elegans sphk-1 gene functions in the DDI-4–induced exocytotic sperm remodeling pathway. spin-4(knu1099) worms were also tested, as spin-4 is a germline-specific gene encoding a transmembrane transporter for sphingosine-1-phosphate (S1P), which has been reported to act in spermiogenesis occurring in hermaphrodites [31].

When we compared the brood sizes of N2 (wild-type) and sphk-1(ok1097) hermaphrodites raised at 20°C or 25°C, those of the sphk-1 mutant were severely reduced at 25°C but not at 20°C (Fig 5C). In contrast, spin-4(knu1099) hermaphrodites produced brood sizes comparable to those of wild type at 25°C, although their brood sizes at 20°C were higher than those of wild type (Fig 5C). We next compared Pron-, ProK-, and DDI-4–induced MOF in spermatids from wild-type, sphk-1, and spin-4 mutant males raised at 20°C or 25°C. As observed in nsun-2(nyg21) males (Fig 4D), ΔMOF-positive ratios by Pron or ProK were unaffected by genotype or rearing temperature (Fig 5D). In contrast, when DDI-4 was used as a SAF, spermatids from sphk-1 mutant males, but not from spin-4 mutant males, raised at 20°C showed dramatically reduced MOF (Fig 5D). At 25°C, neither wild-type nor mutant spermatids responded to DDI-4 (Fig 5D). These results suggest that sphk-1, but not spin-4, is required for DDI-4–triggered exocytotic sperm remodeling.

To confirm the involvement of SPH in the DDI-4–activated pathway, SPH and its analog FTY720 (Fig 6A) were tested for their ability to activate the DDI-4 pathway. MOF induced by SPH (Fig 6B) or FTY720 (Fig 6C) was concentration-dependent. In spermatids from nsun-2(nyg21); him-8(tm611) males raised at 20°C or 25°C, ΔMOF-positive ratios induced by SPH were significantly reduced compared to those in him-8(tm611) spermatids (Fig 6D). In contrast, FTY720-induced activation was unaffected by the nsun-2 mutation or rearing temperature (Fig 6D). These results suggest that both SPH and FTY720 possess SAF activity, but only SPH-induced activation was nsun-2–dependent and temperature-sensitive.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 6. DDI-4 and sphingosine may induce MOF via the same pathway.

(A) Chemical structures of sphingosine (SPH) and its analog, FTY720. (B) and (C) Spermatids from control (him-8(tm611)) males raised at 20°C were incubated at room temperature for 15 min with various concentrations of SPH (B) or FTY720 (C). MOF-positive ratios were then measured. In panel (C), FTY720 concentrations above 30 µM could not be evaluated because of precipitation. MOF-positive ratios in panels (B) and (C) were calculated from at least three independent experiments. ND, not determined. (D) nsun-2–dependence of MOF activation by SPH or FTY720. Spermatids from him-8(tm611) or nsun-2(nyg21); him-8(tm611) males raised at 20°C or 25°C were activated by 15 µM SPH or 25 µM FTY720. ΔMOF-positive ratios for each tested SAF were then calculated as described in Fig 1 (n ≥ 3). *, P < 0.05; **, P < 0.01; ***, P < 0.005. (E) Spermatids from N2, sphk-1(ok1097), or spin-4(knu1099) males raised at 20°C or 25 °C were activated by 15 µM SPH or 25 µM FTY720. ΔMOF-positive ratios for each tested SAF were then calculated as described in Fig 1 (n ≥ 3). *, P < 0.05; ***, P < 0.005.

https://doi.org/10.1371/journal.pgen.1012275.g006

Although SPH stimulated spermatids from sphk-1(ok1097) or spin-4(knu1099) males, the sphk-1 mutation markedly reduced the proportion of MOF-positive sperm at both 20°C and 25°C, whereas spin-4 mutant spermatids exhibited similar activation profiles to those of N2 spermatids regardless of rearing temperature (Fig 6E). The activation profile of FTY720 differed from that of SPH; FTY720 failed to activate spermatids regardless of genotype or rearing temperature (Fig 6E). Thus, SPH-triggered activation was sphk-1–dependent but spin-4–independent, similar to DDI-4. In contrast, FTY720 activated spermatids in a manner dependent on both sphk-1 and spin-4, suggesting that FTY720 is mechanistically distinct from DDI-4 and SPH.

In a widely proposed model of SPH-mediated signaling, S1P receptors are present on the PM, and intracellularly generated S1P is exported via S1P transporters into the extracellular space, where it binds these receptors [32–37]. However, it remains unclear whether S1P receptors are encoded in the C. elegans genome. We therefore examined the subcellular localization of SPIN-4 in spermatids of the COP2416 strain, a CRISPR/Cas9-generated knock-in strain that is otherwise wild type and carries an mCherry tag inserted immediately upstream of the endogenous spin-4 stop codon. As shown in S5 Fig, SPIN-4::mCherry signals co-localized with Alexa Fluor 488–labeled wheat germ agglutinin (Alexa 488-WGA), which labels MOs [38], suggesting that SPIN-4 localizes to the MO membrane. Thus, FTY720 appears to activate MOF through an organelle-intrinsic signaling pathway mediated by MO-localized SPIN-4.

nsun-2 and sphk-1 may function genetically upstream of the DDI-4 pathway

We next examined whether spermatids from him-8(e1489); snf-10(hc194), zipt-7.1(ok971); him-5(e1490), nsun-2(nyg21); him-8(tm611), and sphk-1(ok1097) males raised at 20°C or 25°C could be activated by alternative in vitro SAFs, including the weak base triethanolamine (TEA) [38], the cation ionophore monensin (MON) [9,39], and the chloride channel blocker 4,4′-diisothiocyanostilbene-2,2′-disulfonic acid (DIDS) [40] (S6A Fig).

S6B Fig and S7 Table show the activation profiles of spermatids from each mutant strain treated with TEA, MON, or DIDS. snf-10 functions in both Pron- and ProK-induced activation pathways (Fig 3B and S1 Table). snf-10 mutant spermatids showed a dramatically reduced response to TEA at both 20°C and 25°C. In contrast, MON or DIDS induced MOF in snf-10 mutant spermatids raised at 20°C or 25°C, although MON-induced ΔMOF-positive ratios were reduced at 25°C. The zipt-7.1 mutant, which likely function in both the DDI-4 pathway and canonical pathways (Fig 3B), similarly showed involvement in TEA responsiveness at both temperatures. Because DDI-4–induced MOF depends on nsun-2 (Fig 4D) and sphk-1 (Fig 5D), these genes may function upstream of the TEA target in the DDI-4–induced pathway. MON-induced activation was markedly attenuated in spermatids from zipt-7.1 mutant males raised at 20°C and 25°C, whereas other tested mutations did not substantially affect MON-induced MOF at either rearing temperature. Only the zipt-7.1 mutation also impaired DIDS-induced MOF in spermatids from males raised at 25°C. Thus, the MON and DIDS targets may not be essential for the DDI-4 pathway. Moreover, zipt-7.1 presumably plays a crucial role in both canonical (Pron- and ProK-induced) and non-canonical (DDI-4–induced) exocytotic sperm remodeling pathways.

NSUN-2 may be a key regulator of temperature-sensitive MOF

We isolated the nyg20 mutant, in which spermatids fail to respond to DDI-4–induced activation (Fig 4A). Complementation analysis suggested that impaired nsun-2 function contributes to the nyg20 phenotype (Fig 4C), although no mutations were detected in the nsun-2 coding, intronic, or flanking regions in the nyg20 genome (S4 Table). Consistent with this observation, SNP mapping localized the causative mutation(s) in nyg20 to chromosome I and/or chromosome X (S3 Fig and S2–S3 Tables). Because nsun-2 is located on chromosome I, one possible interpretation is that nyg20 affects an X-linked gene functionally related to nsun-2. Such a gene would be expected to escape X-chromosome silencing during spermatogenesis, unlike most X-linked genes [41]. Intriguingly, DDI-4–induced MOF was more severely impaired in nyg20 than in the nsun-2 single mutant (Fig 4C), raising the additional possibility that, besides affecting NSUN-2 function, nyg20 may also compromise one or more genes that participate in DDI-4–induced MOF. Identification of the causative mutation(s) in nyg20 will therefore be essential for distinguishing among these possibilities.

The male germline–acting genes translationally regulated by NSUN-2 remain unknown. However, our findings suggest that these genes are involved in meiosis (S4 Fig) and exocytotic sperm remodeling (Fig 4D). In Nsun2-null mice, male meiosis is arrested at the pachytene stage [22]. Owing to the severe meiotic defects in Nsun2-deficient male mice, which lead to markedly reduced spermatozoon production, it is challenging to determine whether Nsun2-null spermatozoa can undergo the ASR in response to DDI-4. Therefore, identifying C. elegans genes translationally regulated by NSUN-2 may provide insight into conserved mechanisms underlying meiosis and exocytotic sperm remodeling in nematodes and mice.

MOF may be activated via at least three distinct pathways in C. elegans spermatids

Because DDI-4 is water-insoluble, it is likely that one or more intracellular proteins in C. elegans spermatids bind DDI-4 and transmit signals that induce temperature-sensitive, nsun-2–dependent MOF. This is consistent with the in silico analysis predicting intracellular SPHK-1 as a DDI-4 target (S5 Table) and with the genetic analysis exhibiting the involvement of sphk-1 in DDI-4–induced MOF (Fig 5D).

In contrast, Pron and ProK are water-soluble SAFs, suggesting that these proteases generate extracellular signals that are transmitted to intracellular molecules to trigger MOF. Pron- or ProK-induced MOF was affected by mutations in genes encoding intracellular proteins, including spe-8, spe-27, and zipt-7.1 (Fig 3B). However, Pron- or ProK-triggered exocytotic sperm remodeling was neither temperature-sensitive nor nsun-2–dependent (Fig 4D), genetically suggesting that DDI-4 activated a pathway at least partially distinct from those activated by Pron or ProK. Because zipt-7.1 mutant spermatids failed to be activated by Pron, ProK, or DDI-4 (Fig 3B), these pathways may converge genetically at or near zipt-7.1. At any rate, the identification of physiological SAFs is essential for elucidating how each MOF-activation pathway is regulated.

The Pron and ProK pathways are considered to be activated by hermaphrodite- and male-produced SAFs in vivo, respectively, thereby suggesting that these two are canonical and essentially required for reproduction in C. elegans. In contrast, the DDI-4 pathway appears dispensable for nematode reproduction under standard rearing conditions. The biological significance of the DDI-4 pathway remains to be elucidated. However, one compelling hypothesis posits that this pathway emerged incidentally from the canonical Pron or ProK pathways and subsequently evolved as a non-canonical mechanism contributing to the ASR in mammalian spermatozoa. This proposition is noteworthy for identifying conserved molecular frameworks underlying exocytotic sperm remodeling in nematodes and mammals. Comparative analyses of the Pron/ProK and DDI-4 pathways will be important for delineating their respective contributions to ASR induction.

SPH-mediated signaling is involved in temperature-sensitive MOF

Activation profiles of additional in vitro SAFs provided insight into the DDI-4–induced pathway. For instance, the weak base TEA, which increases intracellular pH [38], induced temperature-sensitive MOF by acting at a site genetically downstream of the DDI-4 target (S6 Fig and S7 Table). Because increased intracellular pH elevates intracellular Ca² ⁺ concentration ([Ca²⁺]_i) in mammalian spermatozoa [42–46], SPH-mediated signaling may regulate [Ca²⁺]_i to promote MOF.

In a canonical model, SPH is phosphorylated by SPHKs to generate S1P, which is exported via S1P transporters and binds to G protein–coupled S1P receptors, initiating signaling cascades that regulate a variety of biological processes [32–37]. In C. elegans, loss of the S1P transporter gene spin-4 appears to disrupt spermiogenesis in hermaphrodites, resulting in defective pseudopod extension [31]. However, in hermaphrodites, the SPIN-4 localization was not examined in spermatids, whereas SPIN-4 was detected on the self-sperm PM [31]. Because the MO membrane fuses with the spermatid PM during exocytotic sperm remodeling (S1 Fig), our observation that SPIN-4 localized to the MO membrane in male spermatids is not conflicted with the previous findings in hermaphrodites. Although spin-4 involvement in MOF remains unclear, our findings indicate that spin-4 is not essential for MOF induced by Pron, ProK, DDI-4, or SPH (Figs 5D and 6E). In contrast, FTY720 activated MOF in a SPIN-4–dependent manner (Fig 6E). SPHK2 preferentially phosphorylates FTY720, and the phosphorylated SPH analog binds S1P receptors [47,48]. However, it remains unclear whether FTY720 is a better substrate for C. elegans SPHK-1 in spermatids than SPH. Notably, SPH may activate MOF independently of phosphorylation, consistent with the observation that DDI-4, which lacks phosphorylation sites (Fig 1A), triggered MOF in an SPHK-1–dependent but SPIN-4–independent manner (Fig 5D), similar to SPH (Fig 6E). One possible interpretation is that C. elegans SPHK-1 functions as a scaffold for downstream signaling rather than solely as a catalytic kinase, when DDI-4 or SPH serves as a SAF.

Mitochondrial localization and the ability to phosphorylate FTY720 are characteristic of mammalian SPHK2 [27]. C. elegans SPHK-1 localizes to mitochondria [35] and likely phosphorylates FTY720 (Fig 6E), suggesting functional similarity to SPHK2. In general, it remains unclear whether FTY720 is a preferred substrate of SPHK2 relative to SPH. However, if nematode SPHK-1 preferentially phosphorylates FTY720 rather than SPH, this could explain why DDI-4– and SPH-induced MOF occurred independently of SPIN-4 (Figs 5D and 6E), whereas FTY720-induced MOF required SPIN-4 (Fig 6E).

Previous studies [49,50] have suggested that mammalian SPHK2 functions as a catalytic enzyme that phosphorylates SPH and FTY720 and as a scaffold that organizes signaling complexes on intracellular membranes. For instance, mitochondrial SPHK2 generates S1P, which binds Prohibitin 2 to regulate assembly of respiratory complex IV and mitochondrial respiration [49]. SPHK2 also activates autophagy by forming a complex with B-cell lymphoma 2 (BCL2), thereby dissociating the Beclin-1/BCL2 complex [50]. These findings suggest that SPHK2 acts as a molecular platform that spatially coordinates lipid signaling with protein–protein interactions, linking sphingolipid metabolism to mitochondrial function. Consistent with this notion, mitochondria-derived vesicles, termed mitophers, are expelled from developing C. elegans spermatids [51]. This process, known as mitopherogenesis, is a mitochondria-specific form of ectocytosis that regulates the mitochondrial number within cells. By modulating mitochondrial abundance, mitopherogenesis fine-tunes sperm motility and fertilization competence.

Fig 7 shows a proposed model for MOF activation by DDI-4, SPH, or FTY720, based on our findings. If SPHK-1 in C. elegans spermatids functions as a scaffold on the outer membrane of mitochondria, this could reconcile its apparent signaling activity with the absence of S1P production during MOF activation by DDI-4 or possibly SPH. Thus, the scaffold hypothesis provides a unifying explanation for the dual enzymatic and non-enzymatic functions of SPHKs across species. Once S1P is produced within spermatids, it may also induce MOF via SPIN-4 localized on the MO membrane.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 7. Model for the DDI-4–triggered exocytotic sperm remodeling pathway.

SPM, spermatid plasma membrane; MOM, membranous organelle membrane; OMM, outer membrane of mitochondria; FTY720-P, phosphorylated FTY720. Created in BioRender. Nishimura, H. (2026) https://BioRender.com/7ramet3.

https://doi.org/10.1371/journal.pgen.1012275.g007

In mammals, SPH-mediated signaling is presumably involved in the ASR [52]. S1P may promote the ASR by mobilizing [Ca²⁺]_i [53], implicating SPHKs in this process. Moreover, in mouse and capacitated human spermatozoa, SPHK1 localizes to the acrosomal region [53,54]. Although mice lacking either Sphk1 [55] or Sphk2 [56] remain fertile, double knockout of both genes results in embryonic lethality [57]. Hence, the genetic involvement of SPHK1 and SPHK2 in the ASR remains unclear. Our genetic analyses place NSUN-2 upstream of DDI-4–dependent MOF, but the molecular link between NSUN-2–mediated RNA methylation and sphingolipid signaling remains unknown. Clarifying this connection will be an important subject for future studies.

Ethics statement

All nematode and mouse experiments were performed according to protocols approved by the Institutional Animal Care and Use Committee (IACUC) at Setsunan University (Approval Number K25-35).

Animals

We maintained and cultured worms at 20°C unless otherwise specified, following previously described protocols [66]. For heat-stressed conditions, worms at the L4 stage were cultured at 25°C for 48 h. Most strains used in this study were obtained from the Caenorhabditis Genetics Center (CGC; University of Minnesota) or the National BioResource Project (NBRP; Tokyo Women’s Medical University). The nyg20 mutant was isolated in this study by screening for genes involved in DDI-4–triggered MOF after introducing random mutations into the him-8(tm611) background by EMS mutagenesis (for details, see “EMS mutagenesis”). The him-8 mutant hermaphrodites are otherwise normal but produce ~37% males among their self-progeny [67]. The nyg20 mutant was outcrossed five times with the him-8 mutant worms before use, to minimize unrelated background mutations. The nsun-2(nyg21) strain was generated by co-injecting synthetic single guide RNA, recombinant Cas9 nuclease, and a synthetic single-stranded DNA repair template targeting the nsun-2 locus into the germline of him-8(tm611) hermaphrodites (for details, see ”CRISPR/Cas9”). Worms carrying the expected mutation in exon 2 of nsun-2, which introduces a premature stop codon and an EcoRI-cleavage site, were identified by single-worm PCR (SWP)–restriction fragment length polymorphism (RFLP) analysis (for details, see “SWP and SWP-RFLP analyses for genotyping”). A complete list of C. elegans strains is shown in S8 Table. On the other hand, to evaluate ASR-positive ratios in mouse spermatozoa, we used ~10-week-old Slc:ICR males (Japan Slc).

Microscopy

Most worm handling procedures, such as picking and dissection, were performed under SZ61 or SZX10 stereomicroscopes (Evident). Digital differential interference contrast (DIC) and fluorescent images were captured using either a BX53 microscope equipped with a DP72 CCD camera and cellSens software (Evident) or an LSM 710 confocal microscope with ZEN 2010 software (Carl Zeiss). Co-localization of the SPIN-4::mCherry and Alexa 488–WGA signals was also observed and analyzed using the confocal microscope. The number of spermatids released from male gonads was quantified, following DAPI staining, using a BZ-X810 microscope equipped with BZ-X Analyzer software and BZ-H4A/BZ-H4C modules (Keyence).

Nematode MOF assay

This assay was carried out essentially as previously described [10,13,68], and all steps were performed at room temperature (~22°C). Males at the L4 stage were cultured for 48 h at either 20°C or 25°C and then dissected to release round spermatids in 10 μL of 1x Sperm Medium (1x SM; 50 mM Hepes-NaOH, pH 7.4, containing 45 mM NaCl, 5 mM KCl, 1 mM MgSO₄, 5 mM CaCl₂, and 10 mM glucose). After 5-min incubation, 10 μL of SAF Solution (1x SM or 1x SM containing 400 μg/mL Pron, 400 μg/mL ProK, or 100 mM TEA; 1x SM containing 1% DMSO, 40–400 μM DDI-4, 200 μM DDI-4 analogs, or 1 mM DIDS; 1x SM containing 1% ethanol or 2 μM MON; 1x SM containing 1% dimethyl formamide, 10–100 μM SPH, or 20–100 μM FTY720) was added to the cells, and they were incubated for 5–15 min to induce spermiogenesis. To stain cellular membranes, 5 μL of FM1-43 Solution (1x SM containing 25 μg/mL FM1-43) was added to the 20-μL preparation. After a further 5-min incubation, digital DIC and fluorescent (FM1-43) images were acquired as described in “Microscopy”. The number of MOF-positive cells was counted from fluorescent images obtained from each experimental group.

Mouse ASR assay

This assay was based on the previous report [69]. Spermatozoa were released from the cauda epididymis of ICR males into 200-μL drops of human tubal fluid omitting BSA medium (HTF(-) medium) and allowed to stand at 37°C for 30 min in 5% CO₂/95% air. Then, 1.0 x 10⁷/mL epididymal spermatozoa were incubated with either of 0.1% DMSO (solvent control as a mock), 200 μM DDI-4, or 10 μM A23187 in 50 μL of HTF(-) medium at 37°C for 30 min in 5% CO₂/95% air. To quench the ASR by fixing sperm, 100 μL of 1x PBS containing 4% paraformaldehyde (PFA) was added to the sperm suspension. After 20-min incubation at room temperature (~22°C), cells were washed three times by centrifugation at 2,200 x g for 5 min at room temperature in 200 μL of 100 mM ammonium acetate, pH 9.0, and then re-suspended in 100 μL of the same buffer. A 5-μL aliquot from the sperm suspension was put onto a glass slide and then air-dried at room temperature. To stain the AS in the air-dried spermatozoa with Coomassie Brilliant Blue G-250 (CBB), 50 μL of Staining Solution (0.22% CBB contained in 50% methanol and 10% acetic acid) was added to the air-dried cells. After 10 min of incubation at room temperature, cells were washed three times with ultrapure water in a chamber to remove excess CBB. The stained cells were mounted with 10 μL of 1x PBS and observed under a BX53 microscope, as described in “Microscopy”. To perform IF for IZUMO1, epididymal spermatozoa (1.0 x 10⁷/mL) were incubated with 1x PBS, 3% BSA, or 200 μM DDI-4 to induce the ASR. After fixation, washing, and air-drying as described above, the resulting sperm spots were washed three times with 50 μL of 1x PBS, and incubated with 50 μL of Blocking Solution (1x PBS containing 3% BSA and 0.05% Tween-20) for 30 min at room temperature. Then, the sperm spots were incubated with 50 μL of Primary Antibody Solution (Blocking Solution containing 5 µg/mL normal rabbit IgG or 5 µg/mL α-IZUMO1) for 60 min at room temperature, washed twice with Washing Solution (1x PBS containing 0.05% Tween-20), and incubated again with Secondary Antibody Solution (Blocking Solution containing 5 µg/mL Alexa Fluor 568–labeled goat anti-rabbit IgG antibody, 5 µg/mL FITC-PNA, and 5 µg/mL DAPI) for 30 min at room temperature. After washing three times with 50 μL of Washing Solution, one drop of ProLong antifade reagent was placed onto the sperm spots. Mouse spermatozoa were then observed under a BX53 microscope, as described in “Microscopy”.

SWP and SWP-RFLP analyses for genotyping

For him-8 or nsun-2 genotyping, single worms were digested with 1 mg/mL ProK at 60°C for 60 min in 20 μL of 1x Phusion HF Buffer, followed by incubation at 100°C for 15 min to inactivate the enzyme. One microliter of each digest was used as a template in a 10-μL PCR reaction containing Phusion DNA polymerase and the primer pair STNHN326/STNHN327 for him-8 or STNHN520/STNHN521 for nsun-2, according to the manufacturer’s instructions. The expected PCR fragment sizes for him-8 were 2,105 bp for the wild-type allele and 1,692 bp for the tm611 allele. To determine the nsun-2 genotype in each worm, RFLP analysis of the PCR products was performed using EcoRI; the restriction digestion yielded a 314-bp fragment from the wild-type allele, and 241- and 82-bp fragments from the nyg21 allele generated by CRISPR/Cas9 (see “CRISPR/Cas9”). For SNP mapping, DraI was used as a restriction enzyme for SWP-RFLP analysis (see “SNP mapping”).

EMS mutagenesis

C. elegans hermaphrodites were mutagenized with EMS as described previously [10]. L4 him-8(tm611) hermaphrodites were incubated with 50 mM EMS at room temperature (~22°C) for ~4 h. Following mutagenesis, 10 P₀ worms were placed on each 60-mm nematode growth medium (NGM) agar plate seeded with E. coli OP50 (20 plates; 200 P₀ worms in total). After ~100 fertilized F₁ eggs had been deposited on each plate, the P₀ worms were removed and discarded. Twenty-five F₁ hermaphrodites from each plate were individually transferred to fresh NGM plates (500 F₁ worms in total). From each plate, ~ 20 F₂ males were isolated and examined individually by the nematode MOF assay using DDI-4 to detect spermatids unresponsive to DDI-4. In total, approximately 10,000 F₂ males were screened. When a positive F₂ male was identified, additional males from the same pedigree were assayed until hermaphrodites producing exclusively DDI-4–resistant sons were established. From this screen, a single mutant line, nyg20, was isolated. Male spermatids from nyg20 were unresponsive to DDI-4, whereas mutant hermaphrodites remained fertile at 20°C.

SNP mapping

Many of the SNPs examined in this study were previously described [70], and SNP mapping was performed based on the methods reported in that study [70]. Briefly, him-8(tm611); nyg20 hermaphrodites (P₀) were outcrossed with males of the Hawaiian strain (CB4856). Approximately 20 L4 hermaphrodite progeny (F₁) were individually transferred to 35-mm NGM plates (F₁ plates). After confirming that each F₁ plate contained approximately 100 fertilized eggs (F₂), the F₁ hermaphrodites were individually collected and subjected to SWP analysis for genotyping of him-8 (see “SWP and SWP-RFLP analyses for genotyping”). From F₁ plates containing F₁ worms heterozygous for him-8(tm611), a total of 200 L4-stage F₂ hermaphrodites were individually transferred to NGM plates (F₂ plates). From each F₂ plate, approximately five F₃ males were individually examined using the nematode MOF assay with DDI-4 as a SAF (see “Nematode MOF assay”). Finally, ten F₂ plates in which all tested F₃ males produced spermatids unresponsive to DDI-4 were selected as mapping plates. Males from each mapping plate were then subjected to SWP-RFLP analysis using DraI to genotype the SNPs listed in S2 Table. PCR primers used to amplify genomic regions containing these SNPs are listed in S8 Table.

CRISPR/Cas9

A mixture containing 17.5 μM nsun-2 crRNA, 17.5 μM tracrRNA, 17 μM Cas9 nuclease, and 1 μM STNHN519 was prepared according to the manufacturer’s instructions. To introduce the desired mutation into the nsun-2 gene, the mixture was microinjected into the distal gonad arms of him-8(tm611) hermaphrodites at the young adult or adult stage. F₁ self-progeny from the injected P₀ worms were then subjected to SWP-RFLP analysis to identify animals carrying the mutation (see “SWP and SWP-RFLP analyses for genotyping”). Through subsequent self-fertilization, we established the nsun-2(nyg21); him-8(tm611) strain, which was used for further experiments.

Complementation analysis

L4 hermaphrodites of C. elegans strains—wild-type (N2), him-8(tm611), him-8(tm611); nyg20, and nsun-2(nyg21); him-8(tm611)—were cultured at 20°C for 48 h. F₁ males at the L4 stage were individually picked from each NGM plate containing either of the strains, cultured at 20°C for 48 h, and subjected to the nematode MOF assay using DDI-4 as a SAF. In addition, nyg20 and nsun-2 mutant hermaphrodites at the L4 stage were outcrossed with adult N2 and him-8(tm611) males, respectively, at 20°C for 48 h. F₁ male progeny obtained from these crosses were individually cultured at 20°C for 48 h and dissected to release spermatids for the nematode MOF assay using DDI-4. Male carcasses were subsequently collected for genotyping. F₁ males derived from the outcrosses of N2/nyg20 and him-8(tm611)/nsun-2 were analyzed by SWP for the tm611 allele and by SWP-RFLP for the nyg21 allele, respectively.

Evaluation of meiosis in the male gonad

Nuclei of male germ cells in the distal gonad were observed under an LSM 710 confocal microscope to assess meiotic progression. Briefly, to isolate gonads, males cultured for 48 h after the L4 stage at 20°C or 25°C were dissected in 5 μL of 1x M9 (22 mM KH₂PO₄, 42 mM Na₂HPO₄, 86 mM NaCl, and 1 mM MgSO₄) containing 0.1% Tween-20 and 12 mM levamisole. After a 5-min incubation in 10 μL of 1x  M9 containing 1% PFA, the gonads were permeabilized by freeze-cracking [71] and incubated at room temperature for 1 min in 5 μL of 100% ethanol. Following three washes with 100 μL of 1x  PBS, the gonads were incubated at room temperature for 5 min in 5 μL of 1x  PBS containing 0.5% Triton X–100 and 1 mM EDTA. Subsequently, 5 μL of 1x  M9 containing 1 μg/mL DAPI was added, and the samples were incubated for an additional 5 min at room temperature. DAPI fluorescence was then observed under an LSM 710 confocal microscope. In addition, the number of spermatids was counted after release from the male gonad. L4 males were cultured for 48 h at 20°C or 25°C and then individually dissected in a 20-μL drop of 1x SM containing 5 μg/mL DAPI and 0.05% Tween-20 on a slide glass to release spermatids from the gonad. The number of cells was counted as described in “Microscopy.”

Construction of docking models for SPHKs bound to ATP and either DDI-4 or SPH

The human SPHK1 (HuSPHK1)/ADP/SPH complex structure was modeled by superimposing the HuSPHK1/SPH complex (3VZB) and the HuSPHK1/ADP complex (3VZD). The ADP was manually modeled into the ATP-binding site to obtain the HuSPHK1/ATP/SPH complex structure. The initial nematode SPHK-1 (NeSPHK-1)/ATP/SPH complex structure was obtained by superimposing ATP and SPH of the HuSPHK1/ATP/SPH complex onto the NeSPHK-1 structure from the AlphaFold database [72]. The NeSPHK-1/ATP/SPH complex used as the docking template was obtained by minimizing the structure around the binding site of the initial complex. Docking of DDI-4 was performed using AutoDock Vina 1.2.5 [73]. The binding site center was defined as the center of mass of SPH, and a 30 Å cubic box was used to generate 10 binding poses. The final docking model was selected visually from the obtained binding poses. As shown in S6 Table, the amino acids involved in ligand interactions were identified using the Protein-Ligand Interaction Profiler (PLIP) [30]

Evaluation of SPIN-4 localization in spermatids

In COP2416 worms, the mCherry gene was inserted in the N2 genetic background before the termination codon of the endogenous spin-4 gene to express the SPIN-4::mCherry protein. L4 males of the N2 or the COP2416 strains were cultured for 48 h at 20°C and then dissected to release round spermatids in 10 μL of 1x SM. After 5-min incubation, spermatids were further incubated for 5 min in 10 μL of 1x SM containing 2% PFA to fix the cells. Following permeabilization by 5-min incubation in 10 μL of 1x SM containing 0.05% Triton X–100, the cells were stained by 5-min incubation in 10 μL of 1x SM containing 1 μg/mL DAPI and 5 μg/mL Alexa 488–WGA to visualize nuclei and MOs, respectively. Fluorescent images were acquired and analyzed as described in “Microscopy”.

Quantification and statistical analysis

Statistical analyses were performed using Microsoft Excel. Percentage data were subjected to arcsine square-root transformation prior to analysis using Welch’s t-test. For all tests, n indicates the number of independent experiments (at least three), and at least 100 cells were analyzed per experiment. Statistical significance was defined as P < 0.05. All P-values are indicated in the figure legends by asterisks.