Nexin-Dynein regulatory complex component DRC7 but not FBXL13 is required for sperm flagellum formation and male fertility in mice

Flagella and cilia are evolutionarily conserved cellular organelles. Abnormal formation or motility of these organelles in humans causes several syndromic diseases termed ciliopathies. The central component of flagella and cilia is the axoneme that is composed of the ‘9+2’ microtubule arrangement, dynein arms, radial spokes, and the Nexin-Dynein Regulatory Complex (N-DRC). The N-DRC is localized between doublet microtubules and has been extensively studied in the unicellular flagellate Chlamydomonas. Recently, it has been reported that TCTE1 (DRC5), a component of the N-DRC, is essential for proper sperm motility and male fertility in mice. Further, TCTE1 has been shown to interact with FBXL13 (DRC6) and DRC7; however, functional roles of FBXL13 and DRC7 in mammals have not been elucidated. Here we show that Fbxl13 and Drc7 expression are testes-enriched in mice. Although Fbxl13 knockout (KO) mice did not show any obvious phenotypes, Drc7 KO male mice were infertile due to their short immotile spermatozoa. In Drc7 KO spermatids, the axoneme is disorganized and the ‘9+2’ microtubule arrangement was difficult to detect. Further, other N-DRC components fail to incorporate into the flagellum without DRC7. These results indicate that Drc7, but not Fbxl13, is essential for the correct assembly of the N-DRC and flagella.


Introduction
Cilia and flagella are evolutionarily-conserved microtubule-based organelles extending from the surface of many cell types and are used for sensing and motility [1][2][3]. Defects in the formation or motility of these organelles are associated with human diseases such as impaired mucociliary clearance and recurrent respiratory infections, which is called primary ciliary dyskinesia (PCD) [4]. The central component of motile cilia and flagella is the axoneme, the '9+2' microtubule arrangement that consists of a central pair (CP) of two singlet microtubules surrounded by nine outer microtubule doublets ( Fig 1A) [5,6]. Each doublet microtubule is composed of an A-tubule and a B-tubule. In addition to the microtubule arrangement, the axoneme contains several accessory structures such as inner dynein arms (IDAs), outer dynein arms (ODAs), radial spokes (RSs), and the Nexin-Dynein Regulatory Complex (N-DRC). The IDAs and ODAs are motor complexes attached to the A-tubule and drive axoneme beating by sliding the neighboring doublet microtubules in an ATP dependent manner [7,8]. The RSs extend from the doublet microtubules towards the CP and mediate signal transduction between the CP and the dynein arms [9].
In addition to ciliary motility defects, PCD patients can exhibit infertility due to abnormal flagellar movement of the spermatozoa. Although detailed morphology and motility of the spermatozoa were not described, an asthenozoospermia patient with a mutation in GAS8 was reported [27]. In mice, it was reported that Lrrc48 mutant males were infertile in addition to displaying abnormal ciliary motility [26]. Further, Iqcg (Drc9) KO male mice were infertile due to severe defects in flagellar formation and immotility of spermatozoa [28]. Recently, we reported that Tcte1 (Drc5) KO male mice were infertile due to defective sperm motility [29]. We showed that TCTE1 interacts with DRC3, FBXL13 (DRC6), and DRC7 in cultured cell [29], but functional roles of FBXL13 and DRC7 in mammals have not been elucidated. In Drosophila melanogaster, it has been reported that insertional mutations of CG34110 (ortholog of mouse Drc7) affect sperm movement into the female storage receptacle, suggesting that DRC7 may regulate flagellar motility [30]. In this report, we knocked out Fbxl13 and Drc7 and analyzed their function in mouse fertility.

Fbxl13 and Drc7 are evolutionarily conserved and testis-enriched genes
Fbxl13 and Drc7 are evolutionarily conserved genes in many eukaryotes, including Chlamydomonas, Trypanosoma, mouse, and human (S1 and S2 Figs, S1 Table). Localization of each N-DRC component has been analyzed with biochemical and cryo-electron tomography analyses of Chlamydomonas wild-type and mutant axonemes ( Fig 1A) [13,14]. To investigate spatial expression of Fbxl13 and Drc7 in mice, we conducted multi-tissue RT-PCR using cDNA obtained from adult mice. Fbxl13 is expressed strongly in testis with weaker signals detected in lung, ovary, and uterus. Drc7 is also expressed abundantly in testis with weak expression in brain, thymus, lung and ovary (Fig 1B), suggesting that Fbxl13 and Drc7 may have some roles in spermatogenesis. In mice, the first wave of spermatogenesis occurs within the first 35 days  [13,14]. The numbers shown in the N-DRC correspond to each DRC component. FBXL13 (DRC6) and DRC7 are located next to TCTE1 (DRC5). (B) The expression of mouse Fbxl13 and Drc7 was examined by RT-PCR using RNA isolated from various organs. Both Fbxl13 and Drc7 are testis-enriched with weak expression detected in other tissues. Actb was used as a loading control. (C) The expression of Fbxl13 and Drc7 was examined by RT-PCR using RNA isolated from testis at various postnatal days. Actb was used as a loading control.
https://doi.org/10.1371/journal.pgen.1008585.g001 DRC7 is essential for male fertility of postnatal development. [31]. To determine which stage of spermatogenesis Fbxl13 and Drc7 start to express, we conducted RT-PCR using cDNA obtained from postnatal testis. Both genes were strongly expressed from three weeks around when spermiogenesis occurs (Fig 1C). These results suggest that Fbxl13 and Drc7 function during spermiogenesis when flagellum elongation and head morphogenesis occur. FBXL13 or DRC7 localization in the mouse testis and spermatozoa was not analyzed due to the unavailability of effective antibodies against FBXL13 and DRC7. However, in a previous study, DRC7 was detected in the tail fraction of human spermatozoa using proteomic analysis [32]. Further, in Drosophila, DRC7 is localized in the entire flagellum of the spermatozoa [30].

Generation of Fbxl13 or Drc7 knockout mice
To uncover the function of Fbxl13 and Drc7 in vivo, we generated Fbxl13 and Drc7 single knockout (KO) mice. To generate Fbxl13 KO mice we mutated the gene in ES cells using the CRISPR/Cas9 system [33] (S3A Fig). Because exon 10-12 overlapped with Lrrc17, we designed guide RNAs (gRNA) targeting exon 1 and exon 7. These gRNA sequences were inserted into pX459 plasmids that express humanized CAS9 [33]. The plasmids were co-transfected into EGR-G01 ES cells [34] to excise the region flanked by the two gRNA target sequences. We obtained 5 mutant clones with biallelic large deletions out of 48 clones examined with three clones used for injecting into 8-cell embryos. The chimeric mice obtained with biallelic deletions were mated with wild-type females, and heterozygous mice with a 30983 bp deletion were obtained, suggesting that this gene might not be involved in male fertility. To confirm the function of this gene, homozygous KO mice were obtained by subsequent mating (S3B Fig). We confirmed that mRNA was not detected in the Fbxl13 -/testis with RT-PCR using primers designed within exon 1 and 20 (S3C Fig). No overt abnormalities were observed in Fbxl13 -/mice.
For Drc7 KO mice, we established mutant ES clones (EGR-G101) [34] using the targeting vector obtained from Knockout Mouse Project (KOMP). Ten clones out of 48 clones were targeted correctly and three clones were injected into 8-cell embryos. Chimeric mice were mated with wild-type females. The mutant mice obtained were mated with CAG-Cre transgenic mice to delete the region flanked by loxP sites (tm1b allele) (

Fbxl13 is not required for male fertility
In order to analyze the fertility of Fbxl13 -/male mice, individual wild-type or homozygous males were caged with wild-type females for two months. Homozygous males sired a comparable number of pups to the wild-type males (S4A Fig). To investigate if there were any subtle abnormalities not affecting male fertility, we examined spermatogenesis with periodic acid-Schiff (PAS) staining of testis sections, however, no abnormalities were observed in spermatogenesis of Fbxl13 KO testis (S4B Fig). The morphology of KO spermatozoa obtained from the cauda epididymis was also normal (S4C Fig). Further, we examined sperm motility using a Computer Assisted Sperm Analyzer (CASA). No significant differences were found in both motility rates and progressive motility rates (S4D Fig). In addition, velocity parameters such as average path velocity (VAP), straight line velocity (VSL), and curvilinear velocity (VCL) [35] were normal in Fbxl13 KO mice (S4D Fig). Taken together, these results indicate that Fbxl13 is not essential for male fertility, sperm formation, or sperm motility.

Fbxl13 is not required for ciliogenesis in the trachea
Although no overt abnormalities were observed in Fbxl13 -/mice, Fbxl13 expression was detected in tissues other than testis. To examine the role of FBXL13 in other ciliated tissue, multicilia of tracheal epithelia were analyzed as Fbxl13 expression was detected in the lungs (Fig 1B). No morphological abnormality was observed by scanning electron microscopy (S5A Fig) and the '9+2' arrangement of microtubules was observed in the Fbxl13 -/mice (S5B Fig).
These data indicate that motile cilia can form in the Fbxl13 -/trachea.

Drc7 is required for male fertility, head morphogenesis, and sperm flagellum formation
To analyze the fertility of Drc7 -/male mice, individual heterozygous or homozygous males were caged with wild-type females for two months. Although the formation of plugs was observed, homozygous males failed to sire any pups (Fig 2A). To investigate the causes of infertility in Drc7 -/male mice, we observed testis morphology and found that Drc7 -/testes were slightly smaller than those of Drc7 +/- (Fig 2B and 2C). In testis sections with PAS staining, spermatogonia, spermatocytes, and round spermatids were observed in homozygous mice; however, elongated spermatids were hardly detected ( Fig 2D, Stage VII-VIII). Further, elongated spermatids were not released and abnormally retained in stage IX seminiferous tubules of Drc7 -/mice ( Fig 2D). In addition, abnormal head shapes were observed in stage XI (step 11) of Drc7 -/testis ( Fig 2D).
PAS staining of cauda epididymis sections indicated that the control tubules were filled with mature spermatozoa, but spermatozoa were hardly observed in the Drc7 -/epididymis ( Fig 2D, Cauda). When we observed the spermatozoa obtained from the Drc7 -/cauda epididymis, they exhibited significantly shorter tails and abnormal head shapes (Fig 2E and 2F), and were immotile (number of motile spermatozoa = 0 out of 101 spermatozoa observed, number of males = 4) (S1 and S2 Movies). Further, aberrant localization or loss of mitochondria were observed with Mitotracker staining (Fig 2E). These results indicate that Drc7 is necessary for spermiogenesis, especially for flagellum formation and head morphogenesis.

Other N-DRC components in Drc7 KO spermatozoa
We investigated the interaction of N-DRC components. We focused on DRC3, GAS8, and TCTE1 because these DRCs were shown to be localized adjacent to DRC7 in Chlamydomonas flagella ( Fig 1A) [13][14][15]. We co-expressed FLAG-tagged mouse DRC7 and PA-tagged mouse DRC3, GAS8, or TCTE1 in HEK293T cells and confirmed that mouse DRC7 interacts with mouse DRC3 and TCTE1 but not RSPH4A, a radial spoke protein, using immunoprecipitation analysis ( Fig 3A). Because PA-tagged GAS8 bound non-specifically to a mouse monoclonal anti-FLAG antibody (Fig 3A), we utilized a rabbit polyclonal antibody for FLAG for immunoprecipitation and the results showed that DRC7 can also interact with GAS8 ( Fig 3B) Next, we confirmed if the localization of other N-DRC components were disrupted in the short flagellum of Drc7 KO spermatozoa. Because we could not obtain an effective TCTE1 antibody, we crossed Drc7 mutant mice with a mouse line that has a FLAG sequence knocked into the 3' end of the Tcte1 locus [29]. With Western blot analysis using FLAG antibody, we found that the amount of TCTE1-FLAG decreased in the Drc7 KO spermatozoa (Fig 4A). In addition, the amount of DRC3 and GAS8 decreased as well; however, the decreased levels could be due to the short tails of Drc7 KO spermatozoa. To normalize the amount of loaded protein we utilize acetylated tubulin signal from Western blot analysis. Ten times the volume of protein lysate was loaded and the results showed that the amount of TCTE1, DRC3 and GAS8 decreased in Drc7 KO spermatozoa. In contrast, the signal of RSPH6A and RSPH9, proteins localized in the radial spokes, was comparable to control. We also performed immunofluorescence analysis to further detect the localization of DRC3 and GAS8. Both DRC3 and GAS8 were localized in the sperm flagellum and their fluorescence intensities were significantly Drc7 is required for male fertility, head morphogenesis, and sperm flagellum formation. (A) Drc7 +/or Drc7 -/males were mated with wild-type females and the number of pups born per plug is reported. Although 17 plugs were observed, Drc7 -/male mice did not sire offspring. Error bars represent S.D. ��� p < 0.001 (unpaired Student's t-test). (B) Testis morphology and (C) testis/body weight of Drc7 +/and Drc7 -/adult mice. Error bars represent S.D. � p < 0.05 (unpaired Student's t-test). Scale bar, 5 mm. (D) PAS staining of seminiferous tubules and cauda epididymis of adult mice. In stage VII-VIII, normal spermatocytes (red arrowhead) and round spermatids (white arrowhead) were observed in both Drc7 +/and Drc7 -/testis; however, elongated spermatids (black arrowhead) were hardly detected in Drc7 -/-. In stage IX, elongated spermatids were released in Drc7 +/but they were retained in Drc7 -/-(red arrow). Stage XI step 11 spermatids of Drc7 +/mice had normal, hook shaped heads (black arrow) whereas Drc7 KO step 11 spermatids have abnormal "club-shaped" heads (white arrow). In cauda epididymis, spermatozoa were hardly seen in Drc7 -/mice. Scale bar, 50 μm. (E) Observation of spermatozoa extracted from cauda epididymis. Sperm mitochondria (midpiece) is stained with Mitotracker (red). Spermatozoa from Drc7 -/mice have abnormal head shapes, short tails, disrupted midpieces, and are immotile. Hoechst (white), nucleus. Scale bar, 10 μm. (F) The measurement of sperm tail length. Drc7 KO spermatozoa exhibit shorter tails. ��� p < 0.001 (unpaired Student's t-test).
https://doi.org/10.1371/journal.pgen.1008585.g002  Drc7 KO spermatozoa had decreased levels of DRC3, GAS8 and TCTE1. In contrast, the signal of RSPH6A and RSPH9 were detected lower in Drc7 KO spermatozoa (Fig 4B and 4C). These results suggest that the structure of the N-DRC is disrupted because of the absence of DRC7.

Axoneme formation and manchette removal is impaired in Drc7 -/testis
To further analyze the causes of abnormal sperm morphology in Drc7 -/mice, we observed microtubule structures that play important roles in tail elongation and head morphogenesis during spermiogenesis (Fig 5). At the early steps of spermiogenesis, the axoneme composed of microtubules elongates and a transient microtubule structure called the manchette is assembled [36]. The manchette is involved in sperm head elongation, and abnormal head morphology is often related to defects in the formation, function, and movement of the manchette [37][38][39][40]. To observe the axoneme and manchette, we performed immunofluorescence using an anti-α-tubulin antibody on spermatogenic cells obtained from seminiferous tubules. From control mice, the elongating axoneme was observed in all steps. In contrast, we could not find the axoneme in Drc7 -/spermatids, suggesting that axoneme formation is severely impaired in Drc7 KO spermatids. The formation of the manchette was comparable between the control and Drc7 KO spermatids in the early step of spermiogenesis, but in later steps, the movement of the manchette was aberrant in Drc7 KO mice. In the control, the manchette tightened around the nucleus correctly as it moves caudally. On the other hand, in the Drc7 KO mice, the caudal movement of the manchette was delayed, which leads to the contraction of the at comparable levels. (B) Immunostaining of DRC3 (left) and GAS8 (right) in spermatozoa. Enlarged pictures of Drc7 KO spermatozoa are shown in right panel. Hoechst (blue), nucleus; acetylated-tubulin (green); DRC3 or GAS8 (red). Scale bar in upper, 50 μm, below, 10 μm. (C) Signal intensity ratio of DRC3/acetylated tubulin (upper) and GAS8/acetylated tubulin (lower) shown in (B). The signal of DRC3 and GAS8 were significantly weaker in Drc7 KO spermatozoa. Error bars represent S.D. ��� p < 0.001 (unpaired Student's t-test).
https://doi.org/10.1371/journal.pgen.1008585.g004 manchette at an unusual location and subsequent formation of the "club-shaped" head. These results indicate that the failure of axoneme elongation precedes the formation of abnormal sperm heads in the Drc7 KO spermatids.

Ultrastructural analysis of the testis in Drc7 -/mice
To investigate axoneme elongation in Drc7 -/mice in more detail, we observed the testis from Drc7 +/and Drc7 -/mice using transmission electron microscopy (TEM). Consistent with the PAS staining, normal round spermatids (step 8) were observed in Drc7 -/testis (Fig 6A and  6B). The manchette formation in elongating Drc7 KO spermatids was also comparable with that of Drc7 heterozygous spermatids (Fig 6C and 6D). Further, in step 10-11 elongating spermatids, the basal body, a structure for axoneme extension, was correctly located on the posterior side of the nucleus in the Drc7 KO spermatids (Fig 6C' and 6D'). However, it was difficult to find the flagella in the Drc7 -/testis (Fig 6E-6H). In the rare Drc7 KO flagellum found, the arrangement of mitochondria and outer dense fibers were abnormal, and the inner '9+2' arrangement of microtubules were missing (Fig 6F). Very rarely in these Drc7 KO flagellum, scattered microtubule-like structures were found (Fig 6G and 6H), which is consistent with the detection of acetylated tubulin in the mature Drc7 KO spermatozoa (Figs 4B, 4D and 5). These results suggest that the basal body can be located correctly, but flagellum elongation was impaired because the axoneme structure was disrupted in the Drc7 -/mice.

Drc7 is not required for ciliogenesis in the brain and trachea
Drc7 expression was detected in tissues other than testis, however no overt abnormalities were observed in Drc7 -/mice. To examine the role of DRC7 in other ciliated tissues, multicilia of ependymal and tracheal epithelia were observed because Drc7 expression was detected in these tissues. No morphological abnormalities were observed by scanning electron microscopy (S5A

Normal pups are obtained from Drc7 KO spermatozoa with ICSI
To investigate if Drc7 KO sperm nuclei can produce viable pups despite impaired spermiogenesis, we conducted intracytoplasmic sperm injection (ICSI). Ninety wild-type oocytes were injected with Drc7 KO spermatozoa obtained from cauda epididymis and 42 of them were developed to the 2-cell stage (Fig 7A). These embryos were transplanted into pseudopregnant female mice and 4 heterozygous pups were obtained (Fig 7B). These pups grew normally ( Fig  7C). These results showed that Drc7 KO sperm nuclei could activate eggs and produce viable pups, and the infertility phenotype seen in Drc7 KO males could be rescued with ICSI.

Discussion
Fbxl13 and Drc7 are evolutionarily conserved genes that are present in nearly all eukaryotes including Chlamydomonas, Trypanosoma, flies, mice, and humans. In mice, we showed that these genes are testis-enriched and start to express during the haploid phase of spermatogenesis, when several flagellar proteins are translated to form the axoneme [41]. Despite showing testisenriched expression, Fbxl13 is not necessary for male fertility, sperm formation, or sperm motility. According to the previous report in Chlamydomonas, DRC6 (FBXL13 ortholog) is located in the most distal region adjacent to the B-tubule [10,13]; therefore, FBXL13 may not play critical roles compared to proteins localized at the base of the N-DRC. It is also possible that there is either a paralogue or an analog that replaces FBXL13 in mouse flagella. In Chlamydomonas, a structural paralogue (FAP43) of a flagella protein (FAP244) has been shown to replace FAP244 [42]. BLAST searches against Chlamydomonas DRC6 in mice reveal that FBXL4 and FBXL7 are potential orthologs. These two proteins may replace the function of FBXL13.
In contrast, in the absence of DRC7, male mice were infertile due to flagellar malformation. This phenotype is different from CG34110 (ortholog of mouse Drc7) mutants in Drosophila melanogaster [30], which do not exhibit short flagella but show impaired sperm motility, suggesting that flagellar formation may be slightly different in Drosophila and mice. For example, in Drosophila, intraflagellar transport (IFT) is not required for the assembly of flagella [43,44], whereas it is reported that IFT proteins are required for sperm flagellum formation in mouse [45]. Immunostaining of the axoneme during spermiogenesis suggests that the axoneme formation is severely impaired in Drc7 KO spermatids. Consistent with this result, we could not detect any organized '9+2' microtubule arrangement in the Drc7 KO testis with electron microscopy. Mitochondrial sheaths and other accessory structures were also misshapen. Because other N-DRC components were reduced in the Drc7 KO spermatozoa, DRC7 may stabilize the N-DRC, which is essential for proper axoneme elongation. A disorganized axonemal structure could lead to impaired IFT transport on the B-tubule of the outer-doublets, which may cause flagellar shortening in the Drc7 KO spermatozoa. In contrast, in the Tcte1 KO mice, the axoneme structure was normal, but their sperm motility was impaired [29]. In Chlamydomonas, it is indicated that DRC5 (TCTE1 orthologue) is localized to a more distal region from the A-tubule than DRC7, suggesting that DRC7 is essential to localize TCTE1 in the correct place, and that TCTE1 regulates flagellar motility. This is supported by the observation that TCTE1 decreased in Drc7 KO spermatozoa, but DRC7 was detected in the Tcte1 KO spermatozoa [29].
Despite severe phenotypes found in spermatozoa, no defects were recognized in the ependymal and tracheal cilia of the Drc7 KO mice. A recent study has shown that drc7 mutant Chlamydomonas does not display an obvious motility defect, which is consistent with no obvious ciliary abnormalities found in the Drc7 KO mice in this study. As sperm flagellum has accessory structures such as outer dense fibers, fibrous sheaths, and mitochondrial sheaths, sperm flagellum formation may be different from that of cilia and Chlamydomonas flagella. This idea is supported by several KO mouse lines that show normal ciliary formation, but exhibit impaired sperm flagellum formation [46]. Alternatively, it is possible that other proteins compensate the function of DRC7 in the ependymal and tracheal cilia. Proteins used for axonemal formation are known to be different between cilia and sperm flagellum [47].
In addition to flagellar deformation, abnormal sperm head morphology was observed in Drc7 KO mice. Because manchette movement is abnormal, it is likely that tightening of the sperm heads by the manchette occurs at an incorrect place is causing the abnormal morphology. Since flagellar deformation precedes the formation of the misshapen heads in Drc7 KO testis, impaired manchette movement may be a secondary effect of abnormal flagellar formation. As it is suggested in the previous study [48], proper flagellar formation may be required to remove the manchette properly. It is noteworthy to mention that mutations of axonemal proteins often exhibit both abnormal flagellar formation and misshapen heads in mice [37,39,48].
In summary, our results indicate that Drc7 is essential for flagellar formation and male fertility. The phenotype of Drc7 KO mice is similar to human reproduction disorders called multiple morphological abnormalities of the flagellum (MMAF). The genes identified as associated with MMAF thus far account for only one-third of cases with the remaining two-thirds having unknown etiology [49]. By clarifying the interactions between the N-DRC proteins, this may give us a better understanding of the mechanism involved in sperm flagellar formation and may lead us to uncover the remaining causative genes of MMAF.

Ethics statement
All animal experiments were conducted in accordance with the guidelines of "Animal experiment rules" established by the Research Institute for Microbial Diseases, Osaka University, and were approved by the Animal Care and Use Committee of the Research Institute for Microbial Diseases, Osaka University (#Biken-AP-H30-01).

Animals
All mice used in this study were purchased from CLEA Japan (Tokyo, Japan) or Japan SLC (Shizuoka, Japan). All mice were maintained under specific-pathogen-free conditions with ad libitum feeding. The mouse line with FLAG knocked into the 3' end of the Tcte1 locus (Tcte1 tm2(Maik) ) was maintained in our laboratory [29].

Isolation of RNA and RT-PCR
RNA was isolated and purified from multiple adult tissues of C57BL/6N mice at different ages with TRIzol (15596018, Thermo Fisher Scientific). Reverse transcription was performed using purified RNA and the SuperScript III first-strand synthesis system (18080051, Thermo Fisher Scientific). PCR was carried out using KOD Fx neo (KFX-201, TOYOBO, Osaka, Japan). The primers for each gene are listed in S2 Table.

Generation of Fbxl13-deficient mice
Fbxl13 large deletion (LD) KO mice were generated as previously described [33]. We designed two gRNAs that recognize either exon 1 or exon 7 to avoid the cleavage of nearby genes (S2A Fig) and inserted the sequences into the pX459 V2.0 plasmid (#62988, Addgene, Cambridge, MA, USA). EGR-G01 embryonic stem (ES) cells [34] were co-transfected with 1.0 μg of two gRNAs inserted vectors. Nine ES clones out of 48 had a large deletion of Fbxl13. ES cell clones that possessed the mutation were injected into ICR embryos and chimeric blastocysts were transferred into the uteri of pseudopregnant females. Generated chimeric male mice were mated with B6D2F1 female mice to obtain heterozygous KO mice. Fbxl13 KO mice were maintained by sibling matings.

Generation of Drc7-deficient mice
A targeting vector for Drc7 (#PG00082_Z_3_D01) was obtained from the KnockOut Mouse Project (KOMP) and electroporated into EGR-G101 ES cells [34] after linearization. Potentially targeted ES cell clones were selected with G418. Correctly targeted ES cell clones and germ-line transmission were determined via PCR using primers listed in S2 Table. Ten ES clones that were correctly targeted were obtained out of 48 clones. Heterozygous KO mice were generated from the ES cells in the same way as Fbxl13 KO mice.

Genotyping analysis
PCR was performed using KOD FX neo (KFX-201, TOYOBO). The primers for each gene are listed in S2 Table. In vivo male fertility check Each control (wild-type or Drc7 +/-) or KO male was caged with 2 or 3 8-week-old B6D2F1 female mice for 2 months. Plugs and the number of sired pups were checked every morning.
Cauda epididymal spermatozoa were dispersed in TYH medium [50] or non-capacitating medium [51] for 10 min before the observation of morphology and motility. For mitochondria staining, spermatozoa were dispersed in TYH medium containing 10 nM MitoTracker Red FM (M22425, Thermo Fisher Scientific) and 10 μg/mL of Hoechst 33342 (H3570, Thermo Fisher Scientific). After 10 min incubation at 37˚C under 5% CO 2 in air, spermatozoa were observed using a phase contrast microscope (BX50, Olympus).

Sperm velocity analysis
Sperm velocity was analyzed as described previously [52]. Cauda epididymal spermatozoa were suspended in TYH medium [50]. Sperm velocity was measured using the CEROS sperm analysis system (software version 12.3; Hamilton Thorne Biosciences, Beverly, MA, USA) at 10 min and 2 h after incubation. More than 200 spermatozoa were analyzed for each male.

Protein extraction of testis and spermatozoa
Protein samples were extracted from testis and spermatozoa using lysis buffer (6 M urea, 2 M thiourea, and 2% sodium deoxycholate) and centrifuged at 15,300 × g for 15 min at room temperature. Obtained supernatants were subjected to SDS-PAGE for immunoblotting.

Immunoprecipitation
Harvested cells were lysed with a solution containing 1% Triton X-100, 50 mM NaCl, 20 mM Tris-HCl, pH 7.4, Protease Inhibitor Cocktail (25955, Nacalai tesque, Kyoto, Japan), and incubated for 30 min at 4˚C. The lysates were centrifuged at 15,300 × g for 15 min at 4˚C, and the resulting supernatants were incubated for 60 min at 4˚C with FLAG (M2 or PM020) conjugated Dynabeads (10009D, Thermo Fisher Scientific). The immune complexes were then washed three times with a solution containing 40 mM Tris-HCl, 150 mM NaCl, 0.1% Triton X-100, and 10% glycerol and eluted with sample buffer with 2-mercaptoethanol before SDS-PAGE and immunoblot analysis.

Immunoblotting
Proteins obtained from culture cells or tissues were separated by SDS-PAGE under reducing conditions and transferred to PVDF membrane using the Trans Blot Turbo system (Bio-Rad, Foster City, CA, USA). After blocking with 5% skim milk (232100, Becton Dickinson, Cockeysville, MD, USA), the membrane was incubated with primary antibodies overnight at 4˚C and then incubated with HRP-conjugated secondary antibodies. Chemiluminescence was detected by Chemi-Lumi One Super (02230, Nacalai tesque) or Chemi-Lumi One Ultra (11644, Nacalai tesque) using Image Quant LAS 4000 mini (GE Healthcare, Buckinghamshire, UK). Antibodies used are listed in S3 Table. Antibody for RSPH6A was generated as previously described [48].

Immunofluorescence
Immunofluorescence for mature spermatozoa was conducted as described previously [48] with slight modifications. Spermatozoa collected from the cauda epididymis were diluted in PBS, spread onto glass slides, and incubated at 37˚C until dry. The samples were fixed with 4% paraformaldehyde in PBS for 20 min and fixed with 100% methanol at -30˚C for 3 min. After three 10 min washes with 0.1% Triton X-100 in PBS, the slides were blocked with 5% BSA and 10% goat serum for 1 h at room temperature and incubated with primary antibodies overnight at 4˚C. After three 10 min washes with 0.1% Triton X-100 in PBS, the slides were incubated with secondary antibody at room temperature for 90 min and then washed with 0.1% Triton X-100 in PBS three times for 10 min each. To stain the nucleus, the slides were incubated with Hoechst 33342 (H3570, Thermo Fisher Scientific) for 15 min, washed with 0.1% Triton X-100 in PBS three times for 10 min each, and then mounted using Immu-Mount (9990402, Thermo Fisher Scientific). Slides were observed with a BX-53 microscope (Olympus).
Immunofluorescence for manchette was conducted as described previously [48] with slight modifications. Germ cells including spermatids were squeezed out from the seminiferous tubules onto slide glasses and incubated at 37˚C until dry. The samples were fixed with 4% paraformaldehyde in PBS for 15 min. After three 5 min washes with PBS, the samples were permeabilized with 0.3% Triton X-100 for 15 min, washed with PBS three times for 5 min each, and blocked with 5% BSA and 10% goat serum diluted in PBS for 1 h at room temperature. Incubations with antibodies were performed as described above. Samples were observed using a Nikon Eclipse Ti microscope connected to a Nikon C2 confocal module (Nikon, Tokyo, Japan). Fluorescence images were false-colored and cropped using ImageJ software (version 2.0.0, NIH, Bethesda, MD, USA). Antibodies used are listed in S3 Table.

Ultrastructural analysis using transmission electron microscope (TEM)
Testis samples were prepared as previously described [55]. The sections were observed using a JEM-1400 plus electron microscope (JEOL, Tokyo, Japan) at 80 kV with a CCD Veleta 2K × 2K camera (Olympus).
Brain and trachea samples were prepared as previously described [56] with slight modifications. Ultrathin sections stained with uranyl acetate solution for 5 min, briefly washed three times with distilled water, stained with Reynolds lead staining solution for 5 min, and washed three times with distilled water. The sections were observed using a JEM-1200EX electron microscope (JEOL) at 80 kV.

Ultrastructural analysis using scanning electron microscope (SEM)
Brain and trachea samples were prepared as previously described [56]. The samples were observed using a JCM-5000 NeoScope Table Top SEM (JEOL).

Ciliary motility analysis
Ependymal cilia were prepared as previously described [56]. The motility of ependymal cilia was observed under a 60x objective (LUCPlan FL N, Olympus) on a differential interference contrast microscope (BX53, Olympus) and recorded at 200 fps through a HAS-220 high-speed camera (DITECT, Tokyo, Japan).

Intracytoplasmic sperm injection (ICSI)
Mature oocytes were collected from superovulated B6D2F1 female mice of 7-8 weeks of age, 13-15 h after injection of human chorionic gonadotropin (hCG, ASKA Animal Health, Tokyo, Japan). After treatment with 1 mg/ mL of hyaluronidase (088-07243, FUJIFILM Wako Pure Chemical) for 5 min to remove the cumulus cells, oocytes were placed in KSOM medium at 37˚C under 5% CO 2 in air until subjected to ICSI. Wild-type sperm heads were separated from the tail by applying a few piezo pulses. Because mutant spermatozoa proved difficult in separating the head, the whole spermatozoa was used for subsequent ICSI. The sperm head was injected into a mature oocyte using a piezo manipulator (PrimeTech, Tsuchiura, Ibaraki, Japan), as previously described [57]. The following day, two-cell embryos were transferred to pseudopregnant females. Genotyping was conducted one week after the birth.