Kif2c ^flox/flox mice of C57BL/6N background and Ddx4-Cre (Vasa-Cre) transgenic mice (#:006954, The Jackson Laboratory) of FVB background were used (Gallardo et al., 2007). C57BL/6N mice were purchased from SLC and all mice were housed in a 12-h light/dark cycle at 24 ± 1°C, and 60 ± 10% humidity. The mice had free access to food (MF; Oriental Yeast Co., Ltd.) and water. All methods were performed in accordance with relevant guidelines and regulation. Animal handling and experiments were conducted according to protocols approved by the Tohoku University Institutional Animal Care and Use Committee (2019noudou-003-02, 2019noukumikae-030-06).

Germ cell-specific Kif2c conditional knockout mice (Kif2c cKO) of the C57BL/6N strain were generated using i-GONAD, as previously reported (Gurumurthy et al., 2019). To generate Kif2c ^flox/flox mice, firstly, one loxP site was inserted in an intron between exons 1 and 2 (That mice call loxP site 5′- knock-in mice) using the i-GONAD method. Next, loxP site 5′- knock-in mice were mated with each other, and the other loxP site was inserted in the intron between exon 10 and exon 11 into its zygotes at day 0.7. These mice harbored a loxP site flanking exons 2 and 10 and were thus called Kif2c ^flox/flox mice. The crRNA (Integrated DNA Technologies) and ssODN (Integrated DNA Technologies) were designed with the following sequences to insert the loxP site between exon 1 and exon 2:5’- AAATTTACATTTACCTCGGA -3’ and 5’- CCAGGGCTACACAGAGAAACCCTGTCTCGAAAAAACAAACAAACAAACAAACAAAAAAAAACAAAAACAAAACAAAACAAATTTCCCTTCCATAACTTCGTATAGCATACATTATACGAAGTTATGAGGTAAATGTAAATTTGAAGTCTTTAGGATCCTAA -3’. The crRNA and ssODN were designed with the following sequences to insert the loxP site between exons 10 and 11:5’- CATACGACATATTTTCGGCC -3’ and 5’- AGGCCAGCCTAGGCTATAGTGTGAGATCCTAATTATGACAGACATTCTATGTAGTTCACATACGACATATTTTCGATAACTTCGTATAGCATACATTATACGAAGTTATGCCTGGTTTTCTTTTGCCCTTTGTAATACATCCTGAGCAGTGGGGTCAGAGGAGGGGCTGTTTGGCTTAAAGTCT -3’. For the preparation of gRNA, crRNA and tracrRNA (Integrated DNA Technologies) were annealed at 94°C for 2 min, mixed with ssODNs and Cas9 nuclease, and injected into the oviduct on day-0.7 pregnant mice. The primer sequences used for genotyping were as follows: to detect the loxP site between exon 1 and exon 2, Fw1:5’-CCACTGCCTGGCTGTTTGTA-3’ and Rv1:5’-CCACTGCCTGGCCAAACTTA-3’. To detect the loxP site between exons 10 and 11, Fw2:5’-GAGGTTTGGAGGTGCAGTCT-3,’ Rv2:5’-ACCCCACTGCTCAGGATGTA-3’ were used.

The PCR product from the mutant DNA was sequenced using Sanger sequencing to confirm successful insertion. Kif2c ^flox/flox mice were crossed with Ddx4-Cre (Vasa-Cre) transgenic mice (#006954, Jackson Laboratory) to generate Kif2c ^flox/+; Vasa-cre mice. Furthermore, Kif2c ^flox/flox mice were crossed with Kif2c ^flox/+; Vasa-cre mice to generate Kif2c ^flox/-; Vasa-cre mice. Kif2c ^flox/-; Vasa-cre mice were called germ cell-specific Kif2c conditional knockout (Kif2c cKO) mice in this study. The primer sequences used for genotyping to detect Cre were as follows: Fw: 5’- CACGTGCAGCCGTTTAAGCCGCGT-3’, Rv: 5’- TTCCCA-TTCTAAACAACACCCTGAA-3’. Kif2c ^flox/+; Vasa-cre mice, Kif2c ^flox/-; Vasa-cre mice and Kif2c ^+/−; Vasa-cre mice were crossed with C57BL/6N mice at least three times in this study.

Mice were euthanized by cervical dislocation under anesthesia and the tissues were collected. We used an anesth etic mixture of medetomidine (0.3 mg/kg, Medetomin injection Meiji; Meiji Seika Pharma), midazolam (4.0 mg/kg, Midazolam Injection Sandoz; Sandoz K. K), and butorphanol tartrate (5.0 mg/kg, Vetorphale; Meiji Seika Pharma).

To evaluate male fertility, a single 12-week-old male mouse (control or Kif2c cKO) was caged with two 8-week-old female Wild Type (WT) mice for 2 months. Mating was verified by monitoring for the presence of vaginal plugs. Plugged females were replaced with fresh mice. Litter size was scored for females with plugs. To evaluate female fertility, two 8-week-old female mice (control and Kif2c cKO) were caged with a 12-week-old male WT mouse for 5 months. Female mice were separated their pups immediately after the litters and were mated again 1 week later. This process was repeated five times, and the cumulative litter size was scored.

RNA extraction was performed using NucleoSpin® RNA (U0955S, TaKaRa Bio) following the manufacturer’s protocol. Briefly, tissues were homogenized in buffer RA1, after which the filtrate solution was incubated DNase reaction mixture and washed buffer RAW2 and RAW3 three times. The washed solution was extracted with RNase-free H₂O via centrifugation. Reverse transcription was conducted using ReverTra Ace -α-® (FSK-101F, TOYOBO) following the manufacturer’s protocol to obtain cDNA. The obtained cDNA was amplified through qPCR using TB Green® Premix Ex Taq™ II (RR820S; TaKaRa Bio). The primer sequences used were as follows: Fw: 5′- ATGGAGTCGCTTCACGCAC-3′. Rv: 5′- CCACCGAAACACAGGATTTCTC-3′.

The testes and ovaries were collected from control mice and Kif2c cKO mice, and testis weight were measured. Testes were fixed in Bouin’s solution, and ovaries were fixed in 4% formaldehyde. The fixed tissues were gradually dehydrated using a graded series of ethanol and xylene solutions at different concentrations and finally embedded in paraffin. Sections were rehydrated, testes were stained with periodic acid–Schiff (PAS), and ovaries were stained with hematoxylin and eosin and examined under a microscope (BX50; Olympus).

Immunohistochemistry (IHC) was performed as previously described (Hiradate et al., 2022). Tissue sections were placed in an antigen retrieval buffer (pH 9.0; 1 × TE buffer) for 20 min at 90°C. The following primary antibodies were used: rabbit anti-KIF2C antibody (PA5-109879, Thermo Fisher Scientific), mouse anti-SCP3 antibody (sc-74569, Santa Cruz), mouse anti-α-tubulin antibody (sc-32293, Santa Cruz), mouse anti-DDX4/MVH antibody (ab27591, Abcam). The acrosomes were stained with Alexa Fluor 568–conjugated Lectin PNA (L32458; Thermo Fisher Scientific). Sections were examined under a fluorescence microscope (BZ-X710; Keyence).

The TUNEL assay was performed using an in situ apoptosis detection kit (MK500, TaKaRa Bio) as previously described (Hiradate et al., 2022). Briefly, paraffin-embedded testes sections at 8 weeks were rehydrated and labeled with FITC-dUTP by TdT enzyme for 75 min at 37°C. The sections were washed thrice with PBS and incubated with Hoechst 33,342 for nuclear staining. The sections were examined under a fluorescence microscope (BZ-X710, Keyence). One hundred randomly chosen seminiferous tubules per mouse were examined, and TUNEL signals that merged with the nucleus were counted as TUNEL-positive.

To stain individual cells, samples were prepared using the following protocol: seminiferous tubules were fixed in 2% formaldehyde containing 0.1% Triton X-100 for 10 min at room temperature (RT). Pieces of tubules were placed in a drop of fixing solution on a glass slide and the pieces were minced in the drop. The slides were gently tapped onto coverslips using tweezers, frozen in liquid nitrogen for 30 s, and then the coverslips were removed. The slides were washed three times with PBS for 5 min. The slides were treated with blocking buffer (Blocking One, 03953-95, Nacalai Tesque) for 1 h at RT and incubated with the primary antibody overnight at 4°C. The following primary antibodies were used: mouse anti-α-tubulin (Santa Cruz), rabbit anti-α-tubulin (ab52866, Abcam), mouse anti-γ-tubulin (T6557, Sigma-Aldrich), rabbit anti-KIF2C (Thermo Fisher Scientific), rabbit anti-BubR1 (11504-2-AP, Proteintech), and mouse anti-SCP3 (Santa Cruz). The slides were washed thrice with PBS containing 0.1% Tween 20 for 5 min and incubated with Alexa Fluor-dye secondary antibodies (Thermo Fisher Scientific) for 1 h at RT. The ovaries were examined under a fluorescence microscope (BZ-X710; Keyence).

Paraffin-embedded ovaries at 8 weeks were sectioned at 5 µm and the ovarian follicles were counted from every fifth serial section, with 20 sections per ovary. The follicles were divided into primordial (POF), primary (PF), secondary (SF), and antral (AF) follicles and counted without duplication.

To collect oocyte at metaphase Ⅱ, 8-weeks-old mice were superovulated by intraperitoneal injection of 5 IU pregnant mare serum gonadotropin (SEROTROPIN^Ⓡ, Asuka Animal Health) and after 48 h, intraperitoneally injected 5 IU human chorionic gonadotropin (HCG; Mochida Pharmaceutical). Fifteen hours after hCG injection, cumulus-oocyte complexes were collected, and cumulus cells were removed in HTF medium containing 0.1% hyaluronidase (18240-36, Nacalai Tesque) for 5 min at 37.5°C with 5% CO₂. The composition of HTF medium is follow: 101.6 mM NaCl, 4.7 mM KCl, 0.37 mM K₂PO₄, 0.2 mM MgSO₄·7H ₂ O, 2 mM CaCl₂, 25 mM NaHCO₃, 2.78 mM glucose, 0.33 mM sodium pyruvate, 21.4 mM sodium lactate, 286 mg/L penicillin G, and 228 mg/L streptomycin. Oocytes were washed three times with washing buffer (0.1% polyvinyl alcohol and 1% Bovine Serum Albumin in PBS) and fixed with 2% formaldehyde containing 0.2% Triton X-100 in PBS for 40 min at RT. Oocytes were washed thrice with washing buffer and treated with blocking buffer (Blocking One) for 1 h at RT. Oocytes were incubated with the primary antibody overnight at 4°C. The following primary antibodies were used: mouse anti-α-tubulin (Santa Cruz), rabbit anti-α-tubulin (ab52866, Abcam), and rabbit anti-KIF2C (Thermo Fisher Scientific). The oocytes were washed thrice with washing buffer for 5 min and incubated with Alexa Fluor-dye secondary antibodies (Thermo Fisher Scientific) for 1 h at RT. Samples were examined under a fluorescence microscope (BZ-X710, Keyence).

Testes were lysed using RIPA buffer (08714-04, Nacalai Tesque) containing 1% phosphatase inhibitor (07575-51, Nacalai Tesque) and incubated for 30 min on ice. The lysates were centrifuged at 15,000 × g for 15 min at 4°C and eluted with the sample buffer for 5 min at 100°C. Proteins were separated by SDS-PAGE, transferred to a PVDF membrane, and treated with a blocking buffer (Blocking One, 03953-95, Nacalai Tesque) for 30 min at RT. Membranes were then incubated with the primary antibody overnight at 4°C. The following primary antibodies were used: rabbit anti-KIF2C (Thermo Fisher Scientific) and mouse anti-β-actin (sc-47778, Santa Cruz). After washing three times with TBS-T for 5 min, the membranes were incubated with anti-rabbit or anti-mouse HRP-conjugated antibodies (W4011 and W4021, Promega) for 1 h at RT. Images were captured using an ImageQuant LAS 500 (Cytiva).

Each experiment was conducted in at least three independent mice of each genotype. The quantitative analyses were used BZ-X analyzed (BZ-X710, Keyence). Statistical tests were used KyPlot 6.0 version 6.0.2 (KyensLab Inc.). Sample sizes, statistical tests, and p-values are showed in the manuscript, figures, and figure legends.

RT-qPCR analysis of various tissues showed that Kif2c mRNA was highly expressed in the adult testis (Figure 1A). Next, we performed immunohistochemistry to determine KIF2C localization in germ cells of the testis. We found that KIF2C localized as a punctate on chromosomes at the spindle tip at metaphase in spermatocytes (Figure 1B arrowheads at top). Moreover, KIF2C localized along the manchette, a microtubule-based skirt-like transient structure around the nucleus of elongating spermatids (Figure 1B arrows at bottom). SCP3 accumulates in centromeres at metaphase in meiosis (Parra et al., 2009; Parra et al., 2006). Therefore, we examined the definite localization of KIF2C by co-staining with SCP3. KIF2C and SCP3 colocalized at metaphase in meiosis (Figure 1C). Therefore, KIF2C localized to centromeres at metaphase in meiosis. In contrast, KIF2C did not localize to the spindle tip at mitotic metaphase in spermatogonia (Supplementary Figure S1A). These results suggest that KIF2C functions at centromeres in meiotic metaphase spermatocytes and at the manchette in elongated spermatids.

We created germ cell-specific Kif2c conditional knockout mice using the Cre-loxP system. We designed a pair of gRNA and ssODN to insert loxP sites flanking exons 2 and 10 using i-GONAD methods (Gurumurthy et al., 2019) (Figure 2A). We confirmed the successful incorporation of the loxP 34 bp sequence at the target site through genotyping and Sanger sequencing (Figure 2A). Subsequently, the generated Kif2c ^flox/flox mice (loxP sites were inserted homozygously into the Kif2c gene) were crossed with Ddx4-Cre (Vasa-Cre) transgenic mice to generate germ cell-specific Kif2c homozygous knockout (Kif2c ^flox/-; Vasa-cre: Kif2c cKO) mice. Kif2c ^flox/+; Vasa-cre mice and Kif2c ^+/−; Vasa-cre mice are heterozygous knockout mice in germ cells; we used Kif2c ^flox/+; Vasa-cre mice as control (Ctrl) mice for male analysis and Kif2c ^+/−; Vasa-cre mice as Ctrl mice for female analysis. Both Ctrl mice showed normal fertility, comparable to that of WT mice (data not shown). Western blot (WB) analysis confirmed that KIF2C proteins were almost completely lost in Kif2c cKO mouse testes (Figure 2B). Furthermore, IHC confirmed that the expression of KIF2C disappeared in the male germ cells of Kif2c cKO mice (Supplementary Figure S1B arrowheads and arrows). Next, we assessed the fertility of Kif2c cKO male mice and found that Kif2c cKO mice were completely infertile (Figure 2C). Furthermore, the testis size was also significantly smaller (Figure 2D), and the testis weight of the Kif2c cKO mice was significantly lower than that of the Ctrl mice (Figure 2E). These results indicate that Kif2c is essential for male fertility.

To investigate the cause of male infertility and decreased testicular weight in Kif2c cKO mice, we performed a histological analysis of seminiferous tubules using periodic acid-Schiff hematoxylin (PAS-H) staining. Only a few haploid germ cells were present in Kif2c cKO mice compared to Ctrl mice (Figure 3A, low magnification). In more detailed observation, chromosomes aligned at the equator in meiotic metaphase spermatocyte in Ctrl mice (Figure 3A, orange arrowheads at high magnification). However, chromosome misalignments at meiotic metaphase were detected in the majority of spermatocytes in Kif2c cKO mice (Figure 3A, red arrowheads at high magnification). Round and elongated spermatids were not completely absent and only a few spermatids were present (Figure 3A; black and white arrowheads at high magnification; Figure 3B). These results suggest that the number of haploid germ cells was significantly decreased due to chromosomal misalignments at meiotic metaphase in Kif2c cKO mice. To investigate the cause of germ cell decrease, we performed a TdT-mediated dUTP nick-end labeling (TUNEL) assay. TUNEL assay showed that the number of apoptotic cells was significantly increased in Kif2c cKO mice (Figure 4A). Many TUNEL-positive cells were spermatocytes at metaphase, and apoptotic signals were often detected in elongated spermatids (Figures 4B, C). These results indicate that cell death was frequently occurred at metaphase spermatocyte in Kif2c cKO mice. A few spermatocytes differentiated into spermatids in Kif2c cKO mice, however cell death was also induced in elongated spermatids in Kif2c cKO mice. Furthermore, no spermatozoa are present in cauda epididymis in Kif2c cKO mice (Supplementary Figure S2). These results suggest that some kind of abnormality inducing cell death occurred in elongated spermatids in Kif2c cKO mice.

To analyze defects in spermatocyte metaphase in Kif2c cKO mice, we conducted immunocytochemistry on spermatocytes from postnatal day (PD) 20–21 testis, and examined meiotic spindle structure and chromosome alignment. In Kif2c cKO mice, three main types of defects were more frequently observed than in Ctrl mice (Figures 5A–C). The first was partial chromosomal misalignment (Figure 5BII), the second was multiple misaligned chromosomes (Figure 5BIII), and the third was multipolar spindles (Figure 5BIV). The majority of defects observed in metaphase spermatocytes of Kif2c cKO mice were chromosomes misalignment (Figure 5C). In addition, 37% of spermatocytes were showed normal chromosome alignment (Figure 5C). Therefore, chromosome congression at the equator were impaired in Kif2c cKO mice. Next, we investigated the attachment between the kinetochores and the spindle microtubules in Kif2c cKO mice. Unattached kinetochore-spindle microtubule was increased in Kif2c cKO mice spermatocytes, but many cells with normal attached were observed in Kif2c cKO mice (Figures 6A, B). Kinetochore-spindle microtubule attachment and chromosome alignment at the equator are monitored and regulated by a ubiquitous safety device called the spindle assembly checkpoint (SAC) (Burke and Stukenberg, 2008; Caudron et al., 2005; Cleveland et al., 2003; Musacchio and Salmon, 2007; Nasmyth, 2001). When the spindle microtubules have correctly attached to kinetochores and aligned at the equator, SAC is inhibited and it proceeds to anaphase (Musacchio and Salmon, 2007). Thus, we analyzed the behavior of SAC system. The SAC component protein BubR1 was accumulated in metaphase spermatocytes of Kif2c cKO mice (Figures 6C, D). In Kif2c cKO mice, SAC is activated due to chromosome misalignment, leading to meiotic arrest at metaphase. These results suggest that KIF2C functions in the regulation of kinetochore microtubule flux to align chromosomes, rather than in spindle formation or kinetochore-spindle microtubule attachment.

Because KIF2C localized to the manchette in elongated spermatids, KIF2C may also function in the regulation of microtubule dynamics in the manchette. We analyzed the structure of the manchette through staining for α-tubulin using ICC. The manchette begins to appear in step 9 of elongated spermatids, gradually moves to the caudal end of the head, and disassembles in step 14 (Figure 7A Ctrl). In Kif2c cKO mice, the manchette is normally formed and maintained in step 9 elongated spermatids compared to that in Ctrl mice (Figure 7A). However, in step 10-12 elongated spermatids, the head structure was abnormally elongated, and the manchette showed abnormal morphologies, such as being overly elongated in Kif2c cKO mice (Figures 7A, B). Moreover, in step 13-14 elongated spermatids, which occurred when the manchette was gradually disassembled, were not disassembled in Kif2c cKO mice (Figure 7A). These results show that the manchette dynamics were compromised in Kif2c cKO mice (Figure 7C). Therefore, it is suggested that KIF2C regulates microtubule dynamics in the manchette by depolymerizing microtubules and maintains manchette structure (Figure 7C).

In oocyte, KIF2C localize to centromeres in meiotic metaphase (Illingworth et al., 2010; Vogt et al., 2010), and thus we investigated the effect of KIF2C deficit in female fertility and chromosome alignment in oocyte. In contrast to male mice, female mice had sufficient fertility, and even after repeated litters, fertility did not decline and was maintained (Figure 8A). The number of follicles in Kif2c cKO mice was comparable to that in the Ctrl mice (Figure 8B), and the morphology of the ovaries was also normal (Figure 8C). As chromosome misalignment is frequently observed in metaphase spermatocytes, we analyzed chromosome congression in metaphase oocytes using ICC. Chromosome misalignment was slightly increased in Kif2c cKO mice, but there was no significant difference compared with that in Ctrl mice (Figures 9A–C). These results showed that unlike Kif2c cKO male mice, Kif2c cKO female mice had sufficient fertility for a long period, and chromosome alignment in the meiotic metaphase was minorly affected.