Testis expressed 50 is essential for maintaining sperm acrosome integrity during epididymal transit

Significance

Mammalian spermatozoa form in the testis and gain the ability to fertilize during migration from the testis through the epididymis. We reported that sperm malformations are often observed during epididymal transit. The molecular mechanisms for epididymal spermatozoa integrity and survival must be investigated. Here, we identified the testis-specific gene, Testis expressed 50 (Tex50), as essential for male fertility, and TEX50’s absence causes sperm malformations only in the epididymis. There were no differences in spermatogenesis between wild-type and Tex50 mutant mice. Moreover, mCherry-fused TEX50 signals move from the acrosome of round spermatids to the posterior side of the sperm heads during spermiogenesis. We found that sperm membrane protein TEX50 is necessary for sperm integrity in the epididymis and male fertility.

Abstract

In mammals, sperm formation is completed in the seminiferous tubules within the testis, and sperm maturation occurs during the epididymal transit of the spermatozoa. Sperm morphology drastically changes when abnormal spermatozoa migrate from the testis to the epididymis. Detailed molecular mechanisms for sperm survival in the epididymis have not been determined yet. Globozoospermia is a cause of male infertility and is characterized by round-headed spermatozoa without acrosomes, an abnormal sperm nuclear membrane, and sperm midpiece defects. Testis expressed 50 (Tex50) is a testis-enriched gene that is expressed in mice and humans. Using CRISPR-Cas9, we generated Tex50 knockout (KO) mice and found that the KO males were sterile due to epididymal sperm malformations and impaired sperm motility. Surprisingly, electron microscopy, sperm morphology, and globozoospermia-related protein expression and localization in the KO testis were all normal. To understand this phenotype in more detail, we created TEX50-mCherry knockin mice to determine the localization of the TEX50 protein during spermatogenesis. The mCherry signals detected a ring-shaped structure surrounding the sperm acrosome and migrated to the posterior region of the sperm head. After the acrosome reaction, most of the spermatozoa retained mCherry signals. These results indicate that the Tex50 KO globozoospermia phenotype occurs after the migration of spermatozoa from the testis to the epididymis. We found that sperm membrane protein TEX50 is a putative key molecule to survive against globozoospermia-like malformations in the epididymis. It is essential for complete sperm formation and male fertility in mice.

Mammalian reproduction is a complex, highly regulated, and conserved process that needs a specific spatiotemporal set of events to occur. In males, spermatogenesis is the first stage, where germ cells undergo a highly regulated process leading to the differentiation of structurally mature spermatozoa. However, these testicular sperm cannot initially fertilize oocytes and must transit through the epididymis, where they undergo crucial biochemical modifications that allow them to gain the ability to fertilize. In the epididymis, sperm gain the ability to swim through the uterus to the oviduct, where fertilization takes place. Several key events must occur, including zona pellucida binding, the acrosome reaction, zona penetration, and then fusion with the oocyte (1–4). In summary, for mammalian sperm to effectively fertilize, they need to have the correct morphology, the minimum necessary motility, and the correct localization of proteins that interact with the epididymis environment and the oocyte. We have been actively identifying factors that govern these essential steps.

As of April 2023, the World Health Organization has found that about 1 in 6 people experience infertility, steadily rising and becoming an issue around the world (https://www.who.int/publications/i/item/978920068315). Historically, infertility was often misattributed solely to women, but recent studies suggest that 50% of infertility cases are due to issues related to male reproduction (5). When spermatogenesis goes awry, it can lead to infertility due to physical imperfections. Globozoospermia is a known cause of infertility due to improper acrosomal membrane formation, abnormal nuclear shape, and impaired arrangement of the sperm mitochondria, leading to round-headed spermatozoa (6). Without the acrosome, the spermatozoa are unable to penetrate the zona pellucida. This acrosome is a highly conserved, specialized structure that contains a multitude of hydrolyzing enzymes located anteriorly in a thick cap-like structure (7). Many studies have been done on disease model mice for globozoospermia, and all have found a lack of acrosome (8). How the disease is linked to the molecular mechanisms of spermatogenesis is becoming clearer with the identification of essential factors.

It is well understood that sperm morphology is completed in the testis. The distinct sperm structure, including the headpiece and flagella, is created via spermatogenesis (4). In the past decade, numerous proteins have been important for proper sperm development, and at least 12 have been directly linked to globozoospermia. These include DPY19L2, FAM209, FAM71F1, FAM71F2, GBA2, GOPC, PDCL2, PICK1, SPACA1, SSMEM1, TEX46, and ZPBP1 (9–19). In these reports, abnormal acrosome formation in testicular sperm leads to abnormal morphology of the flagella. As a result, knockout (KO) mice with globozoospermia lose the three abilities required for fertilization–proper morphology, motility, and protein localization–and become infertile. The formation of the sperm acrosome progresses when acrosomal granules secreted from the Golgi apparatus and ER are supplied onto the nuclear membrane at the round spermatid stage (20). However, to date, no factors other than SPACA1 have been found that function as sperm membrane proteins localized to the acrosome (12). Based on the above, elucidating the factors that act on the sperm nucleus side will lead to elucidating the molecular mechanism of morphogenesis of the sperm head, including the acrosome, and to identifying the causative gene for male infertility. Thus, we focused on sperm membrane proteins.

The epididymis is composed of three parts [caput (head), corpus (body), and cauda (tail)] and functions in sperm transit from the testis, sperm maturation, and sperm storage (4). The environment of the epididymis and the molecular mechanism by which spermatozoa acquire their fertilization capacity have become clear, but little is known about the mechanism by which sperm maintain the correct morphology (21–23). We recently reported that Tex46 KO mice are infertile because they show abnormal sperm head tip morphology and cannot pass through the zona pellucida of the oocyte. We found that more than 50% of the sperm exhibited sperm head bending when the Tex46 null spermatozoa migrated from the testis to the epididymis (11). Thus, the epididymis is not only a place that positively confers the fertilization ability to sperm but may also have an aspect of providing a survival environment due to lumicrine signaling that does not allow abnormal spermatozoa to proceed to fertilization (24). It has been suggested that it has a similar function not only in mice but also in other animals, such as rabbits (25).

In this paper, we examined Testis expressed 50 (TEX50), a protein that shares an integral membrane domain according to RefSeq (https://www.ncbi.nlm.nih.gov/gene/?term=Tex50). UniProt shows a helical transmembrane domain of mouse TEX50 at positions 78 to 100. We show that mouse TEX50 is a testicular germ cell (TGC)-specific protein expressed during the late stages of spermiogenesis. We characterized a coding sequence deletion of mouse Tex50 generated via the CRISPR-Cas9 system and found that TEX50 is essential for male fertility. Surprisingly, electron microscopy, sperm morphology, and globozoospermia-related sperm proteins in the KO testis were all normal. This suggests that Tex50 KO sperm formation including acrosome biogenesis is completely normal and TEX50 is not necessarily related to globozoospermia. Instead, we found that TEX50 is a putative key sperm protein to protect against sperm malformations in the epididymis.

Results

Mouse Tex50 Is a Conserved and Testis-Enriched Gene.

Mouse Tex50 is a testis-enriched gene on chromosome 1 encoding a 189 amino acid protein. Like other Tex family genes, we first wanted to understand its expression throughout the mouse and human organ systems. RT-PCR analysis in both human and mouse tissues showed strong enrichment of Tex50 in the adult testis (Fig. 1 A and B). We performed RT-PCR using mouse testes at different postnatal stages to determine the stage of spermatogenesis at which Tex50 is expressed. Mouse Tex50 was detected at postnatal week 3 (Fig. 1C), indicating that the gene is expressed during the haploid stage of spermatogenesis. Sequence alignment of TEX50 showed that human and mouse orthologs share a 65.5% (116/177 amino acids) identity in protein sequence. A 23 amino acid signal peptide domain, highlighted by the red box, and a 23 amino acid transmembrane domain, highlighted by the blue box, are conserved across various species, including humans, chimpanzees, rhesus macaques, cows, pigs, and rodents (Fig. 1D). We have also included a 3D image of mouse TEX50 protein as predicted by AlphaFold (Fig. 1E, ID: AF-Q9D5B7-F1).

Fig. 1.

Characterization of TEX50 and generation of Tex50^−/− mice via the CRISPR-Cas9 system. (A) RT-PCR examined the expression of human TEX50 in various organs. TEX50 is testis-specific. Human GAPDH was used as an expression control. Li, liver; Ki, kidney; Br, brain; In, intestine; Te, testis; Ut, uterus; Ep, epididymis; H, head; B, body; T, tail. (B) The expression of mouse Tex50 in various organs was examined by RT-PCR analysis. Tex50 is a testis-specific gene. Hprt was used as an expression control. He, heart; Li, liver; Sp, spleen; Lu, lung; Ki, kidney; Br, brain; St, stomach; In, intestine; Te, testis; Ov, ovary; Ut, uterus. (C) The expression of mouse Tex50 on indicated postnatal weeks in the testis was examined by RT-PCR. Tex50 and Spaca1 begin expression at postnatal week 3, though Zpbp1 expression starts from postnatal week 2. Hprt was used as an expression control. (D) Sequence similarity of TEX50 proteins in various organisms: humans, chimpanzees, rhesus macaques, cows, pigs, rats, and mice. Black indicates a match in four species (>50% match). The red and blue boxes indicate signal peptide and transmembrane regions, respectively. TEX50 is a single-pass transmembrane protein. (E) 3D structure of mouse TEX50 from AlphaFold prediction. The AlphaFold ID is AF-Q9D5B7-F1, and the following link is https://alphafold.ebi.ac.uk/entry/Q9D5B7. (F) Schematic representation of the mouse Tex50 locus on Chromosome 1 and CRISPR-Cas9 targeting scheme. Arrowheads indicate guide RNAs #a/b for Cas9 cleaving. Arrows represent primers #1/2 for genotyping. (G) Genotyping Tex50^−/− mice via PCR amplification using primers #1 and #2. Two PCR bands are shown at 2,740 bp for the wt allele and 425 bp for the mutant (em: enzymatic mutation) allele. (H) The average testicular weight per body weight. n = 4 males each for wt and Tex50^−/− (KO) were used. There was no significant difference (n.s.) between WT and KO mice. n = 4 males.

Generation of Tex50 KO Mice.

To understand the roles of TEX50 in fertility, we generated a Tex50 KO mouse line using the CRISPR-Cas9 system and guide RNAs (gRNA) #a (5’ of exon 1) and #b (downstream of the Tex50 termination codon) (Fig. 1F). A 2,315 base pair deletion of the Tex50 gene, including the 5’ UTR, all of the protein-coding sequences in exons 1 and 2, and the Tex50 intron, was confirmed via PCR using primers (Pr) #1 and #2 to detect the wild-type (wt) and the mutant (em: enzymatic mutation) alleles (Fig. 1G) and via sequence analysis (SI Appendix, Fig. S1A). Homozygous mutant mice (Tex50^−/−) had no overt abnormalities, and their testicular weights and spermatogenesis were not significantly different (Fig. 1H; WT 6.2 ± 0.6 × 10³ vs. KO 5.2 ± 1.2 × 10³, SI Appendix, Fig. S2A).

Tex50 Is Required for Male Fertility, Sperm Head Formation, and Motility.

To test the fertility of the mutant male mice, Tex50^−/− males and wt controls were mated with wt females for about 3 mo. While the three control males had 33 plugs with an average of 100% (33/33 plugs) delivery per plug, the three Tex50^−/− males were sterile despite forming 25 copulatory plugs. To understand why Tex50^−/− males were infertile, we checked the cauda epididymal sperm motility. Using three wt and four Tex50^−/− (KO) mice, we saw that after 10 and 120 min of sperm incubation, none of the KO spermatozoa were motile (Fig. 2A; WT 71.6 ± 8.4% for 10 min and 80.2 ± 6.0% for 120 min incubation). Using light microscopy, we compared the morphology of cauda epididymal spermatozoa (EpiS) of WT and KO mice and found that all Tex50 null spermatozoa had abnormally shaped heads, missing the signature hook shape, and displayed irregular flagellar bending (Fig. 2B). Globozoospermia was confirmed using scanning electron microscopy (SEM) of cauda EpiS, revealing globular, deformed heads (Fig. 2C). Transmission electron microscopy (TEM) revealed that the EpiS exhibited abnormal acrosome (arrow in Fig. 2D), with coiling of the sperm tail (asterisks in Fig. 2D) around the head and malformation of the tail structure (Fig. 2E). To elucidate why the acrosome is underdeveloped, we examined the spermatozoa populations with and without the acrosome in the KO cauda epididymis. We used transgenic mice expressing mitochondria-targeted DsRed2 and acrosome-targeted enhanced green fluorescent protein (EGFP) to track how the mitochondria and acrosome change in WT and KO spermatozoa. There were infrequent acrosome-intact spermatozoa in the cauda epididymis of the KO males (Fig. 2F; wt 89.9 ± 6.4% vs. KO 5.7 ± 1.3%, n = 3 males). Next, we wanted to track acrosomal integrity as the spermatozoa move throughout the reproductive system, given that the acrosome was no longer present in KO EpiS. Interestingly, testicular spermatozoa had no visual abnormalities of the Tex50 null sperm acrosome and flagella vs. the WT sperm (Fig. 2G). However, cauda EpiS did not have EGFP-tagged acrosome signals nor proper flagella formation, as represented by a lack of proper DsRed2 signaling. These results indicate that globozoospermia phenotype (loss of the acrosome and sperm midpiece defects, etc.) in Tex50^−/− mice occurs between the migration of spermatozoa from the testes to the epididymis.

Fig. 2.

Phenotypic analysis of Tex50 null spermatozoa. (A) Percent of motile spermatozoa after capacitation in TYH medium for 10 min and 2 h. Tex50 null (KO) EpiS showed no motility. **P < 0.001, Student’s t test. (B) Phase-contrast microscopy of spermatozoa from the cauda epididymis. Tex50 null (KO) spermatozoa showed malformations of sperm head and tail similar to globozoospermia. n = 3 males. (Scale bar: 10 μm.) (C) SEM observation of mature spermatozoa extracted from the cauda epididymis. Tex50 null (KO) spermatozoa lacked sperm acrosomes and showed a round head and coiled tail. n = 3 males. (Scale bar: 1 μm.) (D) TEM observation of mature spermatozoa in the cauda epididymis. Tex50 null (KO) spermatozoa lacked the acrosome and were characterized by an abnormally bent nucleus (indicated by an arrow) and coiled tail (indicated by asterisks) around the sperm head, consistent with the SEM observations (Fig. 2C). n = 3 males. (Scale bar: 1 μm.) (E) TEM analysis of the tail structure in Tex50 null EpiS. Tex50^−/− (KO) mice display abnormal sperm tail morphology. n = 3 males. (Scale bar: 0.4 μm.) (F) Percent of average acrosome-intact spermatozoa in the cauda epididymis. Most of the Tex50 null (KO) spermatozoa showed acrosomal disruption (WT 89.9 ± 6.4% vs. KO 5.7 ± 1.3%). n = 3 males. **P < 0.001, Student’s t test. (G) Observation of the acrosome and flagella using EGFP expressed under the Acrosin promoter (Acr-EGFP) and mitochondria-targeted DsRed2 (Su9-DsRed2) in the principal piece in wt and Tex50 null (KO) spermatozoa. Tex50 null (KO) spermatozoa showed normal sperm head morphology, including the acrosome and sperm tail in the testis. Malformations of Tex50 null spermatozoa occurred in the epididymis. TS: testicular spermatozoa, EpiS: cauda epididymal spermatozoa. n = 3 males. (Scale bar: 10 μm.)

Observation of Spermiogenesis and the Relationship of Globozoospermia-Related Proteins in Tex50 KO Mice.

In KOs of many other globozoospermia-related proteins, spermatogenesis is often abnormal. For example, in Spaca1 KO mice, the formation of the acrosome deviates during step 6 of spermatogenesis, leading to the degeneration and eventual disappearance of the acrosome (12). To study acrosome biogenesis in the testis of Tex50 KO mice, we used an Acr-EGFP transgenic mouse line (26) that was mated to Tex50 mutant mice and subsequently produced Tex50^−/−, Acr-EGFP transgenic mice. Acrosome formation could, therefore, be examined in the testis via immunofluorescence. In normal development, the acrosome begins its formation in step 3 of spermiogenesis and is completed in step 15. In both WT and Tex50 null spermatozoa, there are no issues in acrosome formation, which is distinctly different from other globozoospermia-related mutant mice (Fig. 3A). To confirm these results, we observed the testis using TEM and found that in both the round spermatid and testicular spermatozoa stages, there were no structural differences in spermiogenesis between WT and Tex50 KO mice (Fig. 3B). Given our previous immunofluorescence results that testicular spermatozoa had no visual abnormalities, we then compared the percentage of abnormal spermatozoa between the testis and cauda epididymis in WT and Tex50 KO mice. There were no abnormalities (0%) in the Tex50 KO testicular spermatozoa but massive abnormalities in the KO EpiS (WT 0.2 ± 0.3% vs. KO 96.8 ± 0.5%) (Fig. 3C). These results confirmed that the globozoospermia phenotype occurs after the testicular spermatozoa migrate from the testis to the caput epididymis (Fig. 3D and SI Appendix, Fig. S2B). As mentioned previously, there are many genes that, when mutated, cause globozoospermia. However, the pathophysiology of these genes differs based on where the protein is expressed. To uncover the relationship between TEX50 and these other proteins, we performed immunoblot analyses of WT and Tex50 null TGCs and EpiS and compared protein expression levels of five other globozoospermia-related proteins (DPY19L2, GOPC, PDCL2, SPACA1, ZPBP1) and two sperm membrane proteins (IZUMO1, SPACA4). Briefly, DPY19L2 is a testis-specific transmembrane protein found in the inner nuclear membrane of mouse spermatids (16, 27). GOPC is a protein localized to the Golgi that supplies acrosomal vesicles (18). SPACA1 is a membrane protein that localizes in the equatorial segment of spermatozoa (12). ZPBP1 is a protein essential for proper acrosome compaction and zona pellucida binding (13). IZUMO1 and SPACA4 are both acrosomal membrane proteins required for fertility (10, 28). Intriguingly, there were no changes in protein levels in the TGCs of WT vs. Tex50 null TGCs (Fig. 3E). In contrast, we see that acrosomal proteins ZPBP1, SPACA1, and IZUMO1 in the Tex50 KO EpiS completely disappear, while DPY19L2 has very faint expression (Fig. 3E). We also find that the levels of GOPC are unchanged, which suggests that TEX50 is not localized to the Golgi. PDCL2 is not present in WT and Tex50 KO EpiS, as this protein is linked to the endoplasmic reticulum which is shed in WT EpiS. SPACA4, a sperm membrane protein crucial for zona pellucida (ZP) penetration (28). Although SPACA4 is present at lower levels in EpiS, the KO spermatozoa are immotile and incapable of reaching the ZP of oocytes. Taken together, TEX50 is a sperm membrane protein required for stability of the sperm acrosome after its formation in the testis.

Fig. 3.

Relationship between spermiogenesis and globozoospermia-related proteins in the Tex50 mutant testis. (A) Immunofluorescence analysis of TGCs in Tex50 null (KO) and WT mice mated to Acr-EGFP transgenic mice. No significant differences between wt and KO mice were observed in steps 3, 12, or 15 of spermiogenesis. Acr-EGFP signals were used as a marker of acrosome biogenesis. n = 3 males. (Scale bar: 10 μm.) (B) TEM observation of round spermatid and testicular spermatozoa stages in Tex50 null (KO) mice. As shown in the immunofluorescence of Acr-EGFP and Hoechst (A), the round spermatid and spermatozoa in the KO testis showed no difference compared to wt germ cells. n = 3 males. (Scale bar: 1 μm.) (C) Percent of average abnormal spermatozoa in the testis and epididymis. Most of the Tex50 null (KO) EpiS were abnormally shaped (WT 0.2 ± 0.3% vs. KO 96.8 ± 0.5%), but no abnormal testicular spermatozoa were found in wt and Tex50 KO mice. n = 3 males. **P < 0.001, Student’s t test. (D) Percent of average abnormal spermatozoa in Tex50^−/− (KO) epididymis. Most of the caput and corpus EpiS are abnormally shaped in the KO mice (Caput 90.5 ± 3.7% and Corpus 97.1 ± 1.4%). n = 3 males. (E) Immunoblot analysis of proteins related to acrosome biogenesis in wt and Tex50 null (KO) TGCs and cauda EpiS: DPY19L2, GOPC, PDCL2, SPACA1, ZPBP1, IZUMO1, and SPACA4. PDCL2 is a TGC-specific protein. GAPDH was used as a loading control.

TEX50-mCherry Localization Pattern in Spermiogenesis.

Our next goal was to understand the localization pattern of TEX50 during spermatogenesis. Because we could not obtain TEX50 antibodies, we instead created mCherry knockin (KI) mice using the CRISPR-Cas9 system (Fig. 4A and SI Appendix, Figs. S1 B and C and S3 A and B). The mCherry signal was incorporated into the C-terminus of TEX50. In cauda EpiS, the TEX50-mCherry signal fluoresces inferior to the acrosomal region, near the sperm head–flagellar junction (Fig. 4B). Next, we tracked the TEX50-mCherry signal spatiotemporally in the testis across 5 different steps of spermiogenesis. There is no TEX50-mCherry signal in the spermatid at the early round spermatid stage of spermatogenesis (Fig. 4C, step 3). As spermiogenesis progresses, the TEX50-mCherry signals are expressed inferior to the acrosome in a collapsed ring-shaped structure (Fig. 4C, step 6). Then, the ring opens up and surrounds the circumference of the sperm heads as they are squeezed into shape (Fig. 4C, step 10). The TEX50-mCherry ring constricts, transitioning from an annular to a flattened conformation (Fig. 4C, step 12). As spermatogenesis continues, the protein continues to flatten and travels away from the acrosome (Fig. 4C, step 14). When spermatozoa are formed, TEX50-mCherry was found near the sperm head–flagellar junction and far from the acrosome. Confocal microscopy confirms the dynamics (steps 7 to 16) of the TEX50-mCherry signals and the protein accumulation at the posterior part of the sperm head as a ring-shaped structure in step 16 (SI Appendix, Fig. S3 C and D and Movies S1–S4). Finally, we wanted to determine TEX50 localization after capacitation, when the acrosome releases its enzymes. The acrosome reaction was induced after 2 h of incubation in a capacitation medium. In acrosome-reacted spermatozoa, Acr-EGFP signals disappear, but TEX50-mCherry signals remain (Fig. 4D). After 10 min of sperm incubation, 60.2 ± 17.0% of EpiS (total 382 spermatozoa analyzed, n = 3 males) are shown to have an intact acrosome (i.e., are Acr-EGFP positive). While 98.5 ± 1.4% of Acr-EGFP-positive spermatozoa had TEX50-mCherry signals before the acrosome reaction, 75.9 ± 8.7% of Acr-EGFP-negative (acrosome-reacted) spermatozoa retained TEX50-mCherry after 2 h of sperm incubation in the capacitation medium (Fig. 4E). Thus, this confirms that TEX50 is not localized to the acrosome of cauda EpiS.

Fig. 4.

Localization of TEX50 in Tex50-mCherry knockin mice. (A) Genotyping Tex50-mCherry KI mice via PCR amplification using primers #1 and #2. A mCherry coding sequence was inserted before the terminal codon of the second exon of the Tex50 gene. A PCR band is shown at 2,740 bp for the wt allele and a PCR band of 3,511 bp for the mCherry-KI allele. (B) Immunofluorescence observation of EpiS in Tex50-mCherry KI mice. TEX50-mCherry signals localized to the postacrosomal region of the sperm head. Acr-EGFP signals indicate sperm acrosomes. n = 3 males. (Scale bar: 10 μm.) (C) Immunofluorescence analysis of TGCs in Tex50-mCherry KI mice. TEX50-mCherry signals are not detected in round spermatids during step 3 of spermiogenesis. mCherry signals are observed around the acrosome of the sperm heads from steps 6 to 10 of spermiogenesis, during which spermatids undergo elongation. Subsequently, TEX50 dissociates from the sperm acrosome (marked by EGFP signals) and relocates to the postacrosomal region of the sperm head during steps 12 and 14 of spermiogenesis, coinciding with the formation of mature spermatozoa. Representative illustrations of each step are indicated in the Right panels (created in https://BioRender.com). n = 3 males. (Scale bar: 10 μm.) (D) Localization of TEX50-mCherry in acrosome-reacted spermatozoa. TEX50-mCherry signals are detected on most acrosome-reacted spermatozoa. Asterisks indicate spermatozoa after an acrosome reaction. n = 3 males. (Scale bar: 10 μm.) (E) Percent of TEX50-mCherry positive spermatozoa during acrosome reaction. After 2 h of sperm incubation, 75.9 ± 8.7% of Acr-EGFP-negative (acrosome-reacted) spermatozoa retained TEX50-mCherry, while 98.5 ± 1.4% of Acr-EGFP-positive (before acrosome reaction) spermatozoa retained TEX50-mCherry. n = 3 males. *P < 0.05, Student’s t test.

Discussion

Tex50 is a testis-enriched gene expressed in the postnatal week 3 testis and encodes a transmembrane protein with high homology across various mammals (Fig. 1 C and D). Using CRISPR-Cas9-mediated KO mice, we demonstrate that TEX50 is critical for proper sperm head formation and male fertility. Sperm motility analysis reveals a complete lack of flagellar movement, resulting in infertility (Fig. 2A). Electron microscopy analysis revealed globular sperm heads and coiled sperm tails (Fig. 2 C–E). Notably, spermatogenesis and sperm morphology appear normal in the testis (Figs. 2G and 3 and SI Appendix, Fig. S2A). However, upon transit through the epididymis, the majority of spermatozoa exhibit characteristics of globozoospermia and become immotile due to abnormal sperm tail formation, ultimately leading to infertility.

Globozoospermia is a cause of male infertility characterized by rounded sperm heads, loss of the acrosome, and sperm midpiece malformation (6). In this study, we examined five proteins—DPY19L2, GOPC, PDCL2, SPACA1, and ZPBP1- previously identified as contributors to globozoospermia and investigated their relationship to TEX50. DPY19L2, SPACA1, and ZPBP1 were absent in the Tex50 null spermatozoa, while the Golgi-localized protein GOPC was unchanged. These findings indicate that TEX50 is a sperm membrane protein, based on several observations: the spermiogenesis expression of Tex50 (Fig. 1C), the unchanged testis weight in Tex50 KO mice (Fig. 1H), the absence of acrosomal membrane protein IZUMO1 (Fig. 3E), and the localization of TEX50-mCherry around the acrosome after step 6 of spermiogenesis (Fig. 4C). A key distinction from previous reports on globozoospermia-related proteins is that the Tex50 KO testis did not exhibit the globozoospermia phenotype (Figs. 2G and 3). Instead, Tex50 null spermatozoa displayed acrosome loss, abnormal head and tail structures, and loss of motility only after migrating from the testis to the epididymis (Figs. 2 and 3 and SI Appendix, Fig. S2). This is a report of a KO mouse model demonstrating this unique phenomenon.

Unfortunately, we were not able to obtain an antibody against TEX50, preventing direct visualization of its localization and interactions with other globozoospermia- or acrosome-related proteins. However, by generating TEX50-mCherry KI mice, we were able to observe its localization (Fig. 4 and SI Appendix, Figs. S1 B and C and S3). During the early round spermatid stage, we saw no TEX50-mCherry signals in the spermatid (Fig. 4C). This correlates to postnatal week 3, coinciding with the onset of Tex50 mRNA expression (Fig. 1C). As spermiogenesis progresses and the sperm head begins to form, the TEX50-mCherry signals become detectable around the acrosome (Fig. 4C). Eventually, the TEX50-mCherry signal localizes to the base of the sperm head (Fig. 4B). We confirmed this localization pattern in mature EpiS. TEX50-mCherry was initially localized around the acrosome in the round spermatids but later shifted to the opposite side of the acrosome by the end of spermatogenesis and remained present after the acrosome reaction (Fig. 4). While TEX50-mCherry signals disappeared in approximately 25% of acrosome-reacted spermatozoa, this percentage included abnormal spermatozoa that lacked an acrosome from the beginning. Therefore, it is likely that few spermatozoa lost TEX50-mCherry signals solely as a result of the acrosome reaction (Fig. 4E). The significance of TEX50-mCherry being initially localized around the acrosome in the testis but later repositioned to a distant location in spermatozoa remains unclear and requires further investigation.

Since spermatogenesis itself proceeds normally in the absence of TEX50, this protein is likely required for maintaining spermatozoa morphology rather than sperm acrosome formation. TEX50 may play a role in stabilizing sperm head shaping during epididymal transit. Its dynamic localization, transitioning from a ring around the acrosome to the head–flagellar junction, suggests that TEX50 is initially involved in acrosome development and later contributes to the stabilization of the head–tail junction. Our findings identify TEX50 as a sperm membrane protein with a previously unrecognized role in spermatogenesis and sperm morphology. The induced globozoospermia-like phenotype in Tex50 null spermatozoa suggests the existence of an additional regulatory mechanism governing sperm survival in the epididymis (25). Notably, spermatozoa in the Tex50 KO testis initially appear normal before migration to the caput epididymis (Figs. 2G and 3 C and D and SI Appendix, Fig. S2B), followed by significant structural changes likely due to defects in sperm acrosome integrity that were not detectable at earlier stages. In addition to its established role in sperm maturation (24), our studies suggest that the epididymis senses the presence or absence of TEX50 and functions as a selective barrier to eliminate abnormal spermatozoa, identifying and eliminating defective spermatozoa that lack fertilization potential. This system reveals an important aspect of the process of sperm maturation in the epididymis. Further investigations are required to identify TEX50-interacting proteins and develop methods for detecting TEX50-deficient spermatozoa.

In summary, we report that TEX50 is essential for proper sperm head and tail formation, motility, and male fertility in mice. Identification of TEX50 as a key regulator of sperm head formation provides valuable insights into the role of acrosome formation and stabilization during epididymal transit. We describe an induced sperm malformation phenotype in which structural abnormalities arise during the migration of testicular spermatozoa (Figs. 2G and 3 C and D and SI Appendix, Fig. S2B). Future investigations are necessary to determine whether the TEX50, a gene conserved from reptiles to mammals, shares a common functional mechanism across species and whether the presence of TEX50 in the sperm serves as a criterion for the fate of the sperm in the epididymis. Given that human TEX50 exhibits testis-enriched expression (Fig. 1A), its role in sperm function may also be conserved in humans. In the urology clinic, human TEX50 may be a diagnostic marker for the regulation of sperm morphology. If a functional inhibitor against TEX50 is identified, it may also be a nonhormonal contraceptive. Our findings highlight the potential of TEX50 as a diagnostic and therapeutic target for male infertility and developing male-specific contraceptives.

Methods

Animals and Human Tissues.

As previously described (10), all animal procedures were approved by the Institutional Animal Care and Use Committees of the National Cerebral and Cardiovascular Center Research Institute and the Research Institute for Microbial Diseases, Osaka University. Human testis tissue was obtained from the Human Tissue Acquisition & Pathology Core at Baylor College of Medicine (BCM). The BCM Institutional Review Board, under Protocol H-14435, approved the collection of these tissues from consenting donors. All samples were provided to the researchers in a deidentified format. Mice were housed on a 12-h light/dark cycle. We generated Tex50 mutant mice—null (B6D2-Tex50<em1Osb>) and mCherry knock-in [B6D2-Tex50<em1(mCherry)Ncvc>]—and crossed them with two transgenic lines, B6D2-Tg(CAG/Su9-DsRed2, Acr3-EGFP)RBGS002Osb (26) and C57BL/6-Tg(CAG/Acr-EGFP)C3-N01-FJ002Osb (29), to obtain Tex50^−/−; Acr-EGFP/Su9-DsRed2 and Tex50-mCherry KI; CAG/Acr-EGFP compound transgenics. These mouse lines will be deposited with the RIKEN BioResource Research Center and the Center for Animal Resources and Development, Kumamoto University (http://card.medic.kumamoto-u.ac.jp/card/english/).

Generation of Tex50 KO and Tex50-mCherry Knockin Mice with CRISPR-Cas9.

CRISPR-Cas9 mediated mutant mice were generated as described previously (30). Briefly, Tex50^−/− mice were generated by electroporating gRNA/CAS9 RNP complex into fertilized eggs (B6D2 background) (NEPA21, Nepagene, Chiba, Japan). CRISPRdirect software was used to search for gRNA and off-target sequences (http://crispr.dbcls.jp/). PCR and direct sequencing were used to screen for mutant mice. The gRNAs and primers used were 5’-TATGACTGACGATGTAACAA-3’ for around the first exon of Tex50 (gRNA#a in Fig. 1F), 5’-AAGCTCGTGCAAATGTAGCT-3’ for the second exon of Tex50 (gRNA#b in Fig. 1F), and 5’-GCTACATTTGCACGAGCTTA-3’ for Tex50-mCherry KI (Integrated DNA Technologies, IA). 1,022-bp single-strand oligonucleotide (Integrated DNA Technologies, IA) encoded mCherry was coelectroporated with the gRNA. The genotyping primers (Pr#1/2 in Fig. 1F) used were 5’-ACAGCCAATCTTTCCGTTGG-3’ and 5’-AGCCACGTTAGAAAGAGTGG-3’ for KO and KI alleles.

Statistical Analysis.

Statistical analyses were performed using Student’s t test in Microsoft Excel after testing the data for normality of distribution. All data were indicated as mean ± SD. Statistical differences were considered significant at *P < 0.05 and **P < 0.001.

Supplementary Material

Appendix 01 (PDF)

Movie S1.

Movie S2.

Movie S3.

Movie S4.

Data, Materials, and Software Availability

All study data are included in the article and/or supporting information.