Testis-specific proteins, TSNAXIP1 and 1700010I14RIK, are important for sperm motility and male fertility in mice

Abstract

Background:

TSN (translin), also called testis brain RNA-binding protein, binds to TSNAX (translin-associated factor X) and is suggested to play diverse roles, such as RNA metabolism and DNA damage response. TSNAXIP1 (Translin-associated factor X-interacting protein 1) was identified as a TSNAX-interacting protein using a yeast two-hybrid system, but its function in vivo was unknown.

Objective:

To reveal the function of TSNAXIP1 in vivo in mice.

Materials and methods:

We generated Tsnaxip1 knockout mice using the CRISPR/Cas9 system and analyzed their fertility and sperm motility. Further, we generated 1700010I14Rik knockout mice, because 1700010I14RIK is also predominantly expressed in testes and contains the same Pfam (protein families) domain as TSNAXIP1.

Results:

Reduced male fertility and impaired sperm motility with asymmetric flagellar waveforms were observed in not only Tsnaxip1 but also 1700010I14Rik knockout mice. Unlike Tsn knockout mice, no abnormalities were found in testicular sections of either Tsnaxip1 or 1700010I14Rik knockout mice. Furthermore, TSNAXIP1 was detected in the sperm tail and fractionated with axonemal proteins.

Discussion and conclusion:

Unlike the TSN-TSNAX complex, whose disruption causes abnormal vacuoles in mouse testes, TSNAXIP1 and 1700010I14RIK may play roles in regulating sperm flagellar beating patterns.

1 ∣. INTRODUCTION

Spermatogenesis is a specialized process by which spermatogonia differentiate into elongated spermatozoa by undergoing meiosis and subsequent spermiogenesis. During spermiogenesis, round spermatids undergo drastic morphological changes such as nuclear compaction and flagellar elongation, leading to the formation of spermatozoa. Spermatozoa are highly unique cells consisting of two distinct parts, heads and flagella. Sperm heads are composed of a nucleus and an acrosome, a specialized organelle containing enzymes that facilitate fertilization. Sperm flagella serve as a motility apparatus and can be divided into three parts, midpiece, principal piece, and end piece.^1,2 The midpiece possesses mitochondria that surround the axoneme, a 9+2 microtubular arrangement, and are thought to produce energy for sperm motility. The principal piece contains a rigid accessory structure called the fibrous sheath that serves not only as a structural component but also as a scaffold for several glycolytic enzymes that also produce energy for sperm motility. No accessory structures are found in the end piece, the tip of sperm flagella. Defects in sperm flagellar structures and/or functions often lead to impaired sperm motility and reduced fertility in men.

TSNAXIP1 (translin-associated factor X-interacting protein 1) was first identified as an interacting protein of TSNAX (translin-associated factor X) using yeast two-hybrid analysis.³ TSNAX forms a heterodimer with TSN (translin) that is also called TB-RBP (testis brain RNA-binding protein).⁴ The interaction between TSN and TSNAX was confirmed in male germ cells and brains.⁵ The TSN-TSNAX complex is proposed to be involved in a variety of biological processes, such as regulation of RNA transport/translation/degradation and DNA damage response.^6,7 Knockout (KO) of Tsn in mice caused the loss of TSNAX in the testis and showed abnormal spermatogenesis, especially in aged (8-month old) mice with vacuoles observed in the seminiferous tubules, although Tsn KO mice did not exhibit impaired male fertility.⁸ In the same study, no significant differences in sperm motility were found in Tsn KO mice compared to the control wild-type (WT) mice.⁸ In contrast, functions of TSNAXIP1 in vivo are still unknown.

In this study, we disrupted TSNAXIP1 in mice using the CRISPR/Cas9 system and found that Tsnaxip1 KO mice exhibited male subfertility. Further examination revealed that Tsnaxip1 KO spermatozoa exhibited impaired motility. We also disrupted 1700010I14RIK, which has the same Pfam (protein families) domain as TSNAXIP1, and found that 1700010I14Rik KO mice showed similar phenotypes to those of Tsnaxip1 KO mice.

2 ∣. MATERIALS AND METHODS

2.1 ∣. Animals

All animal experiments performed in this study were approved by the Institutional Animal Care and Use Committees of Osaka University (Osaka, Japan) in compliance with the guidelines and regulations for animal experiments (approval code: H30–01-0; approval date: July 4, 2018). Mice were purchased from CLEA Japan, Inc. (Tokyo, Japan) or Japan SLC, Inc. (Shizuoka, Japan). WT or heterozygous mice were used for controls. Gene-modified mice generated in this study will be made available through either the RIKEN BioResource Research Center (ID number: RBRC11473 for Tsnaxip1 KO mice; RBRC11494 for 1700010I14Rik KO mice) or the Center for Animal Resources and Development (CARD), Kumamoto University (ID number: 3117 for Tsnaxip1 KO mice; 3138 for 1700010I14Rik KO mice).

2.2 ∣. Pfam domain analysis of amino acid sequences

Amino acid sequences of TSNAXIP1 and 1700010I14RIK were obtained from the CCDS database (https://www.ncbi.nlm.nih.gov/projects/CCDS/CcdsBrowse.cgi). The sequence IDs of each amino acid sequence used were as follows: TSNAXIP1 (CCDS 22614.1) and 1700010I14RIK (CCDS28385.2). Pfam domain detection was performed using Simple Modular Architecture Research Tool (SMART) (http://smart.embl-heidelberg.de/).⁹

2.3 ∣. Isolation of RNA and RT-PCR

RNA was isolated and purified from multiple adult tissues of C57BL/6N mice or testes from 1–5-week-old male mice using TRIzol (Thermo Fisher Scientific, MA, USA). Subsequently, reverse transcription was performed using the purified RNA and SuperScript IV Reverse Transcriptase (Thermo Fisher Scientific). The primers used for PCR of each gene are listed in Table S1.

2.4 ∣. Generation of Tsnaxip1 and 1700010I14Rik KO mice using the CRISPR/Cas9 system

We designed gRNAs using CRISPRdirect software to reduce off-target possibilities.¹⁰ Electroporation was performed as described previously.¹¹ Two gRNAs listed in Table S1 and tracrRNA (Merck, Darmstadt, Germany) were mixed with Cas9 protein (Thermo Fisher Scientific) and Opti-MEM (Thermo Fisher Scientific). This solution was incubated at 37°C to make the gRNA/Cas9 ribonucleoprotein complex, and the obtained complex was electroporated into fertilized oocytes (B6D2F1 X B6D2F1) using NEPA21 Super Electroporator (Nepagene, Chiba, Japan). The treated embryos reaching the two-cell stage were transplanted into the oviducts of ICR pseudo-pregnant females at 0.5 days after mating with vasectomized males. Pups were obtained by natural or caesarean section, and subsequent crosses with B6D2F1 WT mice or siblings were performed to obtain homozygous KO mice. Genotyping was performed using primers listed in Table S1. Tsnaxip1 and 1700010I14Rik double-KO mice were obtained by breeding of each KO mouse line.

2.5 ∣. Fertility analysis of KO lines

Sexually mature B6D2F1 WT or KO male mice were housed individually with three 6-week-old female B6D2F1 mice for 3 months. The numbers of pups and copulation plugs were counted every weekday morning.

2.6 ∣. Morphological and histological analysis of testis and epididymis

After euthanasia, testes and epididymides were dissected. After measuring the testicular weight, testes and cauda epididymides were fixed in Bouin’s fluid (Polysciences, Inc., PA, USA), embedded in paraffin, sectioned, rehydrated, and treated with 1% periodic acid for 10 min, followed by treatment with Schiff’s reagent (FUJIFILM Wako, Osaka, Japan) for 20 min. The sections were stained with Mayer’s hematoxylin solution (FUJIFILM Wako) and observed using an Olympus BX53 microscope equipped with an Olympus DP74 color camera (Olympus, Tokyo, Japan).

2.7 ∣. Analysis of sperm morphology and motility

Cauda epididymal spermatozoa were suspended in a drop of TYH medium.¹² For morphological analyses, spermatozoa were placed on MAS-coated glass slides (Matsunami Glass, Osaka, Japan) and observed using an Olympus BX53 microscope. Sperm motility was analyzed as described previously.¹³ Spermatozoa obtained from the top of the TYH drop were analyzed using the CEROS II sperm analysis system (software version 1.5; Hamilton Thorne Biosciences, MA, USA) after 10 and 120 min of incubation. Spermatozoa were considered progressively motile when VSL/VAP ≥ 0.8 and VAP ≥ 50 μm/s. For sperm motility movies, waveform tracing, and maximum bending angles, spermatozoa were observed with an Olympus BX-53 microscope equipped with a high-speed camera (HAS-L1, Ditect, Tokyo, Japan). Waveforms of sperm flagellum were analyzed using the sperm motion analyzing software (BohBohsoft, Tokyo, Japan).¹⁴ Maximum bending angles were measured using ImageJ (NIH, MD, USA).

2.8 ∣. Immunoblot analysis

Testis and cauda spermatozoa were homogenized in lysis buffer containing 6 M urea, 2 M thiourea, and 2% sodium deoxycholate and incubate overnight at 4°C with rotation. The supernatant was collected after centrifuging at 15,000 × g for 15 min. Samples were subjected to SDS-PAGE under reducing condition using 5% 2-mercaptoethanol, followed by immunoblotting. After blocking with 10% skim milk in TBS containing 0.1% Tween-20, blots were incubated with primary antibodies at 4°C, and then incubated with secondary antibodies conjugated to horseradish peroxidase (1:5000, goat anti-mouse IgG [H+L] #115-036-062, goat anti-rat IgG [H+L] #112-035-167, or goat anti-rabbit IgG [H+L] #111-036-045, Jackson ImmunoResearch, West Grove, PA, USA) for 120 min at room temperature. Primary antibodies used: rabbit anti-TSNAXIP1 1:1000 (#17730-1-AP, Proteintech, IL, USA); rabbit anti-TSN antibody (a kind gift from Dr. Shin-ichi Kashiwabara, University of Tsukuba, Ibaraki, Japan)¹⁵; anti-SLC2A3 monoclonal antibody 1:1000, KS64-10¹⁶; anti-IZUMO1 monoclonal antibody 1:1000, KS64-125¹⁷; mouse anti-acetylated tubulin 1:5000 (#T7451, Sigma-Aldrich, MO, USA); rabbit anti-ODF2 1:1000 (#12058-1-AP, Proteintech); rat anti-PA tag 1:1000 (016-25861, FUJIFILM Wako); mouse anti-FLAG tag (M2) 1:1000 (F1804, Sigma-Aldrich); and mouse anti-α-tubulin (B-5-1-2) 1:5000 (T5168, Sigma-Aldrich). Immunoreactive proteins were detected using Chemi-Lumi One Super (#02230, Nacalai Tesque, Kyoto, Japan) or Chemi-Lumi One Ultra (#11644, Nacalai Tesque).

2.9 ∣. Separation of sperm heads and tails

Sperm head-tail separation was performed as previously described¹⁸ with some minor modifications. Spermatozoa obtained from the cauda epididymis were suspended in 1 ml PBS and sonicated to separate tails from heads on ice (Sonifier SLPe, Branson Ultrasonics, CT, USA). The sample was centrifuged at 15,000 × g for 15 min. The pellet was then resuspended with 200 μl PBS and mixed with 1.8 ml of 90% Percoll solution (GE Healthcare, IL, USA) in PBS. After centrifugation at 15,000 × g for 15 min, sperm heads were at the bottom of the tube and the tails were in the top layer of the solution. The separated sample was diluted five-fold with PBS and centrifuged at 10,000 × g for 15 min. The pellets were washed twice with PBS and dissolved in lysis buffer containing 6 M urea, 2 M thiourea, and 2% sodium deoxycholate.

2.10 ∣. Fractionation of spermatozoa protein

Fractionation of spermatozoa obtained from the cauda epididymis was performed as described previously^19,20 with some minor modifications. After sperm proteins were solubilized with 1% SDS lysis buffer (75 mM NaCl, 24 mM EDTA, pH 6.0), the samples were centrifuged at 15,000 × g for 10 min and the pellet was dissolved in lysis buffer containing 6 M urea, 2 M thiourea, and 2% sodium deoxycholate.

2.11 ∣. Construction of expression plasmids

RNA obtained from C57BL/6N mouse testes was used to amplify cDNAs encoding Tsnaxip1 and 1700010I14Rik. Tsnaxip1 cDNA and 1700010I14Rik cDNA were cloned into a C-terminal FLAG-tagged pCAG vector²¹ and a C-terminal PA-tagged pCAG vector, respectively. Primers used to construct the plasmid vectors are listed in Table S1.

2.12 ∣. Cell culture and transfection

HEK293T cells²² were cultured in DMEM (#11995–065, Thermo Fisher Scientific) supplemented with 10% fetal bovine serum (#S1560, Biowest, MO, USA) and 1% penicillin-streptomycin-glutamine (#10378–016, Thermo Fisher Scientific) at 37°C under 5% CO₂. The calcium phosphate transfection method was utilized to transiently transfect the plasmid DNA,²² and cells were cultured for 24 h before harvesting.

2.13 ∣. Immunoprecipitation

Transfected HEK293T cells were suspended in lysis buffer (1% Triton X-100, 20 mM Tris-HCl pH 7.4, 50 mM NaCl) containing protease inhibitor cocktail (#25955, Nacalai Tesque) and incubated for 1 h at 4°C. The lysates were centrifuged at 15,300 g for 5 min at 4°C. The obtained supernatants were incubated for 1 h at 4°C with anti-PA antibody-conjugated Dynabeads (#10009D, Thermo Fisher Scientific). After washing three times with wash buffer (40 mM Tris-HCl pH 7.5, 150 mM NaCl, 0.1% Triton X-100 and 10% glycerol), the immune complexes were eluted with elution buffer (132 mM Tris HCl pH 7.4, 4% SDS, 20% glycerol, and 0.01% Bromophenol Blue). The eluents were subjected to SDS-PAGE for immunoblotting as described above.

2.14 ∣. Ultrastructural analysis of sperm flagellum using TEM

Cauda epididymis was fixed with 4% PFA in PBS at 4°C. After the fixation, the samples were prepared as previously described,²³ and the ultrastructure was observed using a JEM-1400 plus electron microscope (JEOL, Tokyo, Japan) at 80 kV with a CCD Veleta 2K×2K camera (Olympus).

2.15 ∣. Statistical analyses

Statistical analyses were performed using a two-tailed Welch’s t-test by Microsoft Office Excel (Microsoft Corporation, WA, USA). p-Values lower than 0.05, 0.01, and 0.001 were considered significant (*), (**), and (***), respectively. Error bars in the graphs represent standard deviation.

3 ∣. RESULTS

3.1 ∣. TSNAXIP1 is important for male fertility

RT-PCR using RNA obtained from multiple mouse tissues showed that Tsnaxip1 was specifically expressed in the testis while Tsnax was expressed ubiquitously in multiple tissues, with strong signals detected in the testis and ovary (Figure 1A), suggesting that Tsnaxip1 may play specific roles in the testis. To determine which stage Tsnaxip1 starts to express over the course of spermatogenesis, we performed RT-PCR using RNA obtained from mouse postnatal testes. Unlike Tsnax, Tsnaxip1 started to express postnatally from 2 weeks when primary spermatocytes were seen (Figure 1B).²⁴ We then generated Tsnaxip1 KO mice using the CRISPR/Cas9 system to examine the function of TSNAXIP1 in vivo. Two gRNAs were designed to largely delete the open reading frame (ORF) of Tsnaxip1 (Figure 1C). Of the 58 fertilized oocytes that were electroporated, 40 two-cell embryos were transplanted into the oviducts of two pseudo-pregnant female mice, and 15 pups were born. Nine of the 15 pups had a large deletion, and subsequent crosses were performed to obtain homozygous KO mice. Tsnaxip1 KO mice did not show any obvious abnormalities in development or behavior. Genotyping of obtained mice was performed by genomic PCR (Figure 1D) using the listed primers (Table S1), and subsequent Sanger sequencing revealed that obtained Tsnaxip1 KO mice exhibited a 5,717 bp deletion (Figure 1E). We performed immunoblot analysis to confirm the depletion of TSNAXIP1 in Tsnaxip1 KO testis and spermatozoa (Figure 1F). To examine if Tsnaxip1 plays a role in male fertility, we caged individual Tsnaxip1 KO males with three females for 3 months. Mating tests revealed that Tsnaxip1 KO male mice exhibited impaired fertility compared to the WT control mice (Figure 1G), suggesting that Tsnaxip1 is important for male fertility.

FIGURE 1.

TSNAXIP1 is important for male fertility. (A) Expression patterns of Tsnaxip1 in mouse multiple tissues. Actb was used as a loading control. Br: brain, Th: thymus, Lu: lung, He: heart, Li: liver, Sp: spleen, Ki: kidney, Te: testis, Ut: uterus, Ov: ovary, and NC: negative control (water). (B) Expression of Tsnaxip1 in testes at various postnatal days. Actb was used as a loading control. NC: negative control (water). (C) CRISPR/Cas9 targeting scheme of mouse Tsnaxip1. White boxes indicate untranslated regions while black shaded boxes indicate protein coding regions. Primers for genotyping are indicated in blue. (D) Genotyping of Tsnaxip1 mutant mice. Fw #1-Rv #1 (for KO) and Fw #2-Rv #2 (for WT) primers in (C) were used. NC: negative control (water). (E) Wave pattern sequence of Tsnaxip1 KO mice. Tsnaxip1 KO allele exhibits a 5,717 bp deletion. (F) Confirmation of Tsnaxip1 KO by immunoblotting. Acetylated tubulin was used as a loading control. Asterisk indicates nonspecific bands. (G) The number of pups born per plug detected. Three WT females were caged with each WT or Tsnaxip1 KO male for 3 months. p = 3.37 × 10⁻¹⁹ (Welch’s t-test).

3.2 ∣. Tsnaxip1 KO mice showed reduced sperm motility with abnormal flagellar waveforms

To identify the cause of impaired male fertility found in Tsnaxip1 KO mice, we initially examined spermatogenesis. No abnormalities were observed in testis gross morphology and weights between the control and Tsnaxip1 KO mice (Figure 2A,B). PAS staining of both testis and cauda epididymis showed no obvious differences between the two genotypes as well (Figure 2C and Figure S1A). Because it was reported that Tsn KO males showed increase in the number of vacuoles in seminiferous tubules as they got older,⁸ we also observed testicular sections of aged Tsnaxip1 KO mice (8-month old) (Figure S1B). However, no abnormalities were found even in aged Tsnaxip1 KO testes, suggesting that the function of TSNAXIP1 may be different from that of TSN. Furthermore, when we obtained mature spermatozoa from the cauda epididymis, no obvious morphological defects were found in Tsnaxip1 KO spermatozoa by phase contrast microscopy (Figure 2D).

FIGURE 2.

Histological analyses and sperm morphology of Tsnaxip1 KO mice. (A) Gross morphology of WT and Tsnaxip1 KO testes. (B) Average weights of control and Tsnaxip1 KO testes. p = 0.916 (Welch’s t-test). (C) PAS staining of testis and cauda epididymis sections. (D) Observation of spermatozoa obtained from the cauda epididymis using phase contrast microscopy.

We then assessed fertilizing ability of spermatozoa in vitro to elucidate the function of TSNAXIP1. In vitro fertilization (IVF) analysis using cumulus-intact oocytes revealed a significant reduction of fertilization rates in Tsnaxip1 KO mice (Figure 3A) consistent with the result of mating tests (Figure 1G). However, the decrease in fertilization rates was rescued when the zona pellucida (ZP), an extracellular matrix surrounding the oocyte, was removed (Figure 3B). These results suggest that Tsnaxip1 KO spermatozoa can undergo the acrosome reaction and fuse with eggs. We then performed computer-assisted sperm analysis (CASA) to evaluate sperm motility (Figure 3C). Although Tsnaxip1 KO mice did not exhibit significant reduction in the percentage of motile spermatozoa compared to the control spermatozoa (Figure S2A), the percentage of progressively motile spermatozoa was significantly lower in Tsnaxip1 KO mice than that of the control mice at 120-min incubation (Figure 3D), likely due to their circular motion (Figure 3C). Intriguingly, among velocity parameters such as VAP (average path velocity), VSL (straight-line velocity), and VCL (curvilinear velocity), there was a significant difference between the control and Tsnaxip1 KO mice only in the VCL at 10-min incubation (Figure S2B), suggesting that impaired progressive motility is caused by the cumulative effect of changes in velocity parameters.

FIGURE 3.

TSNAXIP1 is important for sperm motility. (A) IVF analyses using cumulus-intact oocytes, p = 0.0037 (Welch’s t-test). (B) IVF analyses using zona pellucida (ZP)-free oocytes, p = 0.153 (Welch’s t-test). (C) Trajectories of moving spermatozoa analyzed by CASA (computer assisted sperm analysis) after 120-min incubation in TYH medium. Light blue tracks highlight progressively motile spermatozoa; green for motile; magenta for late (slow moving). Red points indicate immotile spermatozoa. (D) Percentages of progressively motile spermatozoa, p = 0.737 for 10 min and p = 0.018 for 120 min (Welch’s t-test). (E) Sperm flagellar beating patterns were analyzed after 10- and 120-min incubation. Moving spermatozoa were videotaped at 200 frames per second, and one cycle of beating was indicated by superimposing videotaped images.

To further examine the defects in sperm motility, we traced the flagellar waveform (Figure 3E Movies S1 and S2). Flagellar beating pattern of control spermatozoa showed a symmetrical waveform, whereas Tsnaxip1 KO spermatozoa showed an asymmetric waveform. This asymmetric waveform might result in circular motion and reduced progressive motility in Tsnaxip1 KO mice (Figure 3C,D). In addition, we quantified maximum bending angles of the midpiece in primary anti-hook curvature after 120-min incubation in capacitating medium, which is used to quantitatively analyze hyperactivation.²⁵ In contrast to the control, Tsnaxip1 KO spermatozoa exhibited low maximum bending angles, suggesting that Tsnaxip1 is important for hyperactivation (Figure S2C).

3.3 ∣. 1700010I14RIK contains the same TSNAXIP_N domain as TSNAXIP1 and its absence caused male subfertility with impaired sperm motility

Our laboratory has been analyzing testis-enriched genes by generating KO mice with the CRISPR/Cas9 system.^26,27 Among the testis-enriched genes, we found that 1700010I14RIK had the same Pfam domain called TSNAXIP_N as TSNAXIP1 by SMART analysis (Figure S3A), suggesting that 1700010I14RIK may have similar roles to that of TSNAXIP1. The TSNAXIP_N domains between TSNAXIP1 and 1700010I14RIK are 36% identical at the amino acid sequence level.We confirmed that 1700010I14Rik was predominantly expressed in the testis by performing RT-PCR using RNA obtained from multiple mouse tissues (Figure 4A). RT-PCR using RNA obtained from postnatal mouse testes revealed that 1700010I14Rik started to express gradually postnatally from 2 weeks (Figure 4B) consistent with Tsnaxip1 (Figure 1B). We then generated 1700010I14Rik KO mice using the CRISPR/Cas9 system in the same way as Tsnaxip1 KO mice to examine the function of 1700010I14RIK in vivo. Two gRNAs were designed to largely delete the ORF of 1700010I14Rik (Figure 4C). Of the 145 fertilized oocytes that were electroporated, 120 two-cell embryos were transplanted into the oviducts of six pseudo-pregnant female mice and 28 pups were born. Fifteen of the 28 pups had a large deletion and subsequent crosses were performed to obtain homozygous KO mice. 1700010I14Rik KO mice did not show any obvious abnormalities in development or behavior. We performed genomic PCR for genotyping (Figure 4D) using the listed primers (Table S1), and subsequent Sanger sequencing revealed that the 1700010I14Rik KO mice have a 19,812 bp deletion (Figure 4E). To examine if 1700010I14Rik plays a role in male fertility, we caged individual 1700010I14Rik KO males with three females for 3 months. Mating tests revealed that 1700010I14Rik KO mice exhibited impaired male fertility compared to the WT mice (Figure 4F), suggesting that 1700010I14Rik is important for male fertility.

FIGURE 4.

1700010I14RIK is important for male fertility and sperm motility. (A) Expression patterns of 1700010I14Rik in mouse multiple tissues. Actb was used as a loading control. Br: brain, Th: thymus, Lu: lung, He: heart, Li: liver, Sp: spleen, Ki: kidney, Te: testis, Ut: uterus, Ov: ovary, and NC: negative control (water). 1700010I14Rik is shown as I14Rik. (B) Expression of 1700010I14Rik in testes at various postnatal days. Actb was used as a loading control. NC: negative control (water). (C) CRISPR/Cas9 targeting scheme of mouse 1700010I14Rik. White boxes indicate untranslated regions, while black shaded boxes indicate protein coding regions. Primers for genotyping are indicated in red. (D) Genotyping of 1700010I14Rik mutant mice. Fw #1-Rv #1 (for KO) and Fw #2-Rv #2 (for WT) primers in (C) were used. NC: negative control (water). (E) Wave pattern sequence of 1700010I14Rik KO mice. 1700010I14Rik KO allele exhibits a 19,812 bp deletion. (F) The number of pups born per plug detected. Three WT females were caged with each WT or 1700010I14Rik KO male for 3 months. p = 1.59×10⁻¹¹ (Welch’s t-test). (G) Trajectories of moving spermatozoa analyzed by CASA after 120-min incubation in TYH medium. Light blue tracks highlight progressively motile spermatozoa; green for motile; magenta for late (slow moving). Red points indicate immotile spermatozoa. (H) Sperm motility was analyzed 10 and 120 min after incubation in TYH medium. Percentages of progressively motile spermatozoa are shown. p = 0.349 for 10 min and p = 0.022 for 120 min (Welch’s t-test). (I) Sperm flagellar beating patterns were analyzed after 10- and 120-min incubation. Moving spermatozoa were videotaped at 200 frames per second and one cycle of beating was indicated by superimposing videotaped images.

To identify the cause of impaired fertility found in 1700010I14Rik KO males, we examined the testes; however, no abnormalities were found in testis gross morphology and weights between the control and 1700010I14Rik KO mice (Figure S3B,C). PAS staining of both testis and epididymis showed no obvious differences between the two genotypes (Figure S3D). Furthermore, when we observed the morphology of spermatozoa obtained from the cauda epididymis by phase contrast microscopy, no obvious defects were found in 1700010I14Rik KO spermatozoa (Figure S3E). We then performed CASA to evaluate sperm motility (Figure 4G). Although the percentage of motile spermatozoa in 1700010I14Rik KO mice was comparable to that of the control mice (Figure S4A), the percentage of progressively motile spermatozoa was significantly lower in 1700010I14Rik KO mice than that of the control mice at 120-min incubation (Figure 4H). Further analyses showed that velocity parameters such as VSL, VCL, and VAP at 120-min incubation were significantly impaired in 1700010I14Rik KO mice compared to the control mice (Figure S4B). In addition, 1700010I14Rik KO spermatozoa showed an asymmetric waveform (Figure 4I Movies S3 and S4) and low maximum bending angles (Figure S4C), which is similar to Tsnaxip1 KO spermatozoa (Figure 3E and Figure S2C). These results suggest that the function of 1700010I14RIK may be similar to that of TSNAXIP1.

3.4 ∣. Tsnaxip1 and 1700010I14Rik double-KO mice exhibit impaired male fertility but are not infertile

Because both TSNAXIP1 and 1700010I14RIK contain the same TSNAXIP1_N domain and each KO mice showed similar phenotypes, TSNAXIP1 and 1700010I14RIK may interact. Therefore, we co-expressed FLAG-tagged TSNAXIP1 and PA-tagged 1700010I14RIK in human embryonic kidney 293T (HEK293T) cells and performed an immunoprecipitation analysis using an anti-PA antibody. After immunoblot analysis, we found that TSNAXIP1 can interact with 1700010I14RIK (Figure S5A). We then examined the protein amounts of TSNAXIP1 in 1700010I14Rik KO males; however, there were no significant differences in the protein amounts of TSNAXIP1 between WT and 1700010I14Rik KO testes and spermatozoa (Figure S5B), indicating that 1700010I14RIK depletion does not affect the amount of TSNAXIP1.

Because TSNAXIP1 exists in 1700010I14Rik KO testis and spermatozoa, TSNAXIP1 may compensate the function of 1700010I14RIK. Therefore, we analyzed the phenotypes of Tsnaxip1 and 1700010I14Rik double-KO (dKO) males that were obtained by breeding of each KO mouse line. We caged individual dKO males with three females for 3 months and found that dKO males exhibited impaired fertility (Figure S6A), but were not infertile. This result indicates that male mice can still have some pups without both Tsaxip1 and 1700010I14Rik. Consistent with Tsnaxip1 and 1700010I14Rik single-KO males, dKO spermatozoa exhibited normal morphology (Figure S6B), but impaired motility with circular travel paths (Figure S7A-D). Further, consistent with Tsnaxip1 and 1700010I14Rik single-KO spermatozoa, dKO spermatozoa showed an asymmetric waveform (Figure S7E Movies S5 and S6) and low maximum bending angles (Figure S7F), but additional phenotypes not seen in either single KO were not found.

3.5 ∣. TSNAXIP1 is localized in the sperm tail and fractionated with axonemal proteins

Because TSNAXIP1 was suggested to associate with TSN, we examined protein amounts of TSN in the testes and spermatozoa of Tsnaxip1 KO, 1700010I14Rik KO, and dKO mice. Immunoblot analysis revealed that KO of Tsnaxip1, 1700010I14Rik, or both genes did not affect protein amounts of TSN in the testes and spermatozoa (Figure 5A,B), while it was reported that the absence of TSN caused the loss of TSNAX in mouse testes.⁸ These results support the idea that the function of TSNAXIP1 may be different from that of TSN.

FIGURE 5.

TSNAXIP1 was detected in the sperm tail. (A) Immunoblot analysis of TSN in testes of WT, Tsnaxip1 KO, 1700010I14Rik KO, and dKO (Tsnaxip1 and 1700010I14Rik) mice. Acetylated tubulin was used as a loading control. 1700010I14Rik is shown as I14Rik. (B) Immunoblot analysis of TSN in epididymal cauda spermatozoa of WT, Tsnaxip1 KO, 1700010I14Rik KO, and dKO mice. Acetylated tubulin was used as a loading control. (C) Head and tail separation of mouse spermatozoa. TSNAXIP1 was detected in the tail fraction. IZUMO1 and acetylated tubulin were detected as a marker for heads and tails, respectively. Asterisk indicates nonspecific bands. (D) Fractionation of sperm proteins using different lysis buffers. TSNAXIP1 was found in the SDS-soluble fraction that contains axonemal proteins. SLC2A3, acetylated tubulin, and ODF2 were detected as a marker for Triton-soluble, SDS-soluble, and SDS-resistant fractions, respectively. Asterisk indicates nonspecific bands. (E) Ultrastructure of sperm tails in Tsnaxip1 and 1700010I14Rik dKO mice. Cross-sections of midpiece and principal piece were observed using transmission electron microscopy.

We then analyzed the localization of TSNAXIP1 in the mature spermatozoa with immunofluorescence, but anti-TSNAXIP1 antibody did not work for immunofluorescence. Therefore, we sonicated mature spermatozoa to separate sperm heads and tails for immunoblotting and detected the TSNAXIP1 band in the tail fraction (Figure 5C). Further, we performed a sperm fractionation assay as described previously^19,20 and detected TSNAXIP1 in the Triton X-100 resistant/SDS-soluble fraction that contains axonemal proteins such as acetylated tubulin (Figure 5D). These results suggest that TSNAXIP1 may be localized in the axoneme and play roles in regulating the axoneme bending rather than RNA and DNA binding with TSN. Of note, no abnormalities were found in the ultrastructure of the axoneme with transmission electron microscopy even in Tsnaxip1 and 1700010I14Rik dKO spermatozoa (Figure 5E).

4 ∣. DISCUSSION

In this study, we showed that Tsnaxip1 and 1700010I14Rik have the same Pfam domain called TSNAXIP1_N (Figure S3A), are predominantly expressed in the mouse testes (Figures 1A and 4A), and begin expression postnatally from 2 weeks (Figures 1B and 4B) when primary spermatocytes are first seen.²⁴ Further, we generated KO mice of each gene using the CRISPR/Cas9 system and found that each single-KO mouse line exhibited reduced male fertility due to impaired sperm motility. Both KO spermatozoa exhibit an asymmetric waveform (Figures 3E and 4I), suggesting that both TSNAXIP1 and 1700010I14RIK may be involved in regulating flagellar bending patterns.

TSNAXIP1 was found to bind to TSNAX, an interacting protein of TSN that plays roles in RNA metabolism.²⁸ It was reported that aged Tsn KO mice (8-month old) as well as young Tsn KO mice (2-month old) showed abnormal vacuoles in testicular sections⁸; however, no abnormalities were found in the testicular sections of even aged Tsnaxip1 KO mice (8-month old) (Figure S1B), suggesting that the interaction of TSNAXIP1 with the TSN-TSNAX complex is not critical for the function of TSN-TSNAX in spermatogenesis. Rather, TSNAXIP1 may play roles in regulating flagellar bending patterns as mentioned above. This idea is supported by the result that showed TSNAXIP1 localization in the sperm tail (Figure 5C). Further, TSNAXIP1 was found in the Triton X-100 resistant/SDS soluble fraction (Figure 5D), suggesting that TSNAXIP1 may be associated with the axoneme.

We also found that 1700010I14Rik KO mice exhibited impaired sperm motility with abnormal flagellar waveforms, which is similar to Tsnaxip1 KO mice. Using the heterologous expression system, we reveal that TSNAXIP1 can interact with 1700010I14RIK, although their relationship in vivo remains unknown. At least, our result showed that TSNAXIP1 was not depleted in the testes and spermatozoa of 1700010I14Rik KO mice. Further, we found that Tsnaxip1 and 1700010I14Rik dKO male mice were subfertile, indicating that TSNAXIP1 and 1700010I14RIK are not essential for male fertility even if these genes work together. It remains to be determined if the common TSNAXIP_N domain is involved in the regulation of sperm motility.

In conclusion, we found two testis-specific genes, Tsnaxip1 and 1700010I14Rik, that are important for sperm motility and male fertility in mice. Amino acid sequences of both TSNAXIP1 and 1700010I14RIK are conserved well between mice and humans (Figure S8A,B). According to Genome Aggregation Database (gnomAD), a TSNAXIP1 variant that causes a frameshift (p.Glu222ArgfsTer4) was found at high frequency (allele frequency: 0.001400) in humans, and two examples were found to possess the homozygous variant. Further analyses of Tsnaxip1 and 1700010I14Rik may lead to a better understanding of the etiology of human infertility.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.