KCTD19 associates with ZFP541 and HDAC1 and is required for meiotic exit in male mice

Meiosis is a cell division process with complex chromosome events where various molecules must work in tandem. To find meiosis-related genes, we screened evolutionarily conserved and reproductive tract-enriched genes using the CRISPR/Cas9 system and identified potassium channel tetramerization domain containing 19 (Kctd19) as an essential factor for meiosis. In prophase I, Kctd19 deficiency did not affect synapsis or the DNA damage response, and chiasma structures were also observed in metaphase I spermatocytes of Kctd19 KO mice. However, spermatocytes underwent apoptotic elimination during the metaphase-anaphase transition. We were able to rescue the Kctd19 KO phenotype with an epitope-tagged Kctd19 transgene. Immunoprecipitation-mass spectrometry identified zinc finger protein 541 (ZFP541) and histone deacetylase 1 (HDAC1) as binding partners of KCTD19, indicating that KCTD19 is involved in chromatin modification. Phenotyping of Zfp541 KO spermatocytes demonstrated XY chromosome asynapsis and recurrent DNA damage in the late pachytene stage, leading to apoptosis. In summary, our study reveals that KCTD19 associates with ZFP541 and HDAC1, and that both KCTD19 and ZFP541 were essential for meiotic exit in male mice. Author summary Meiosis is a fundamental process that consisting of one round of genomic DNA replication and two rounds of chromosome segregation producing four haploid cells. To properly distribute their genetic material, cells need to undergo complex chromosome events such as a physical linkage of homologous chromosomes (termed synapsis) and meiotic recombination. The molecules involved in these events have not been fully characterized yet, especially in mammals. Using a CRISPR/Cas9-screening system, we identified the potassium channel tetramerization domain containing 19 (Kctd19) as an essential factor for meiosis in male mice. Further, we identified zinc finger protein 541 (ZFP541) and histone deacetylase 1 (HDAC1) as binding partners of KCTD19. By observing meiosis of Zfp541 knockout germ cells, we found that Zfp541 was also essential for meiotic completion. These results show that the KCTD19/ZFP541 complex plays a critical role and is indispensable for male meiosis and fertility.

damage response, and chiasma structures were also observed in metaphase I 23 spermatocytes of Kctd19 KO mice. However, spermatocytes underwent apoptotic elimination 24 during the metaphase-anaphase transition. We were able to rescue the Kctd19 KO 25 phenotype with an epitope-tagged Kctd19 transgene. Immunoprecipitation-mass 26 spectrometry identified zinc finger protein 541 (ZFP541) and histone deacetylase 1 (HDAC1) 27 as binding partners of KCTD19, indicating that KCTD19 is involved in chromatin 28 modification. Phenotyping of Zfp541 KO spermatocytes demonstrated XY chromosome 29 asynapsis and recurrent DNA damage in the late pachytene stage, leading to apoptosis. In 30 summary, our study reveals that KCTD19 associates with ZFP541 and HDAC1, and that 31 both KCTD19 and ZFP541 were essential for meiotic exit in male mice. To determine KCTD19 localization, we performed immunostaining of testicular 155 sections with a specific antibody against KCTD19 (Rat mAb #2; Fig. 2F). KCTD19 signals 156 started to appear in the nuclei of spermatocytes in seminiferous stage III -IV (Fig. 2G), 157 corresponding to early pachytene stage. The signal continuously localized in the nuclei of 9 spermatocytes ( Fig. 2G; stage VII -VIII and X -XI). During the metaphase-anaphase 159 transition in meiosis, KCTD19 signal spread throughout the cell ( Fig. 2G;  Due to an apparent defect in meiosis in Kctd19 del/del male mice, we examined DNA 166 double-strand breaks (DSBs) and synapsis by immunostaining γH2AX and synaptonemal 167 complex protein 3 (SYCP3), respectively. γH2AX signals appeared in the leptotene/zygotene 168 stage and disappeared in the pachytene/diplotene stage, except for the XY body ( Fig. 3A  169 and 3B), suggesting that Kctd19 del/del spermatocytes underwent DSB initiation and resolution 170 as controls. Also, homologous chromosomes in Kctd19 del/del spermatocytes synapsed in 171 pachytene stage and desynapsed in diplotene stage remaining physically connected at 172 chiasmata without obvious defects ( Fig. 3A and 3B). However, the diplotene population 173 declined in juvenile Kctd19 del/del males (P20), but not in adult males (Fig. 3C) 174 To uncover the cause of apoptosis in metaphase spermatocytes, we stained spread 175 chromosomes with Giemsa's staining. We observed a normal number of bivalent 176 chromosomes with chiasmata ( Fig. 3D)  To elucidate KCTD19 function, we identified interacting proteins by 207 immunoprecipitation (IP) and mass spectrometry (MS). We lysed Kctd19 del/del and juvenile 208 (PND21) WT testis with non-ionic detergent (NP40) and incubated the lysate with antibodies 209 (rabbit pAb and rat mAb #1) and protein G-conjugate beads. The specific co-IPed proteins 210 were visualized by SDS-PAGE and silver staining ( Fig. 5A and B). When eluted samples 211 were subjected to MS analysis, HDAC1 (histone deacetylase 1) and ZNF541 (Zinc finger 212 protein 54; ZFP541) were reproducibly detected with both antibodies (Fig. 5C), consistent 213 with a prior study (18). KCTD19 and HDAC1 association was confirmed by reciprocal IP with 214 an anti-HDAC1 antibody (Fig. 5D). 215 HDAC1 is a modulator of chromatin structure and disruption of HDAC1 results in embryonic 216 lethality before E10.5 (27) In previous reports, KCTDs were implicated in HDAC degradation 217 (15, 17). We examine the behavior of HDAC1 in Kctd19 del/del testis by immunoblotting 218 analysis and immunostaining with the anti-HDAC1 antibody. HDAC1 protein levels and 219 localization were comparable between Kctd19 del/del and WT testis ( Fig. 5E and 5F). The 220 HDAC1 staining intensity was the strongest in spermatocytes in stage X -XI and lost in 221 elongating spermatids (Fig. 5F), reminiscent of the KCTD19 staining pattern (Fig. 2G). 222 These results indicated that KCTD19 works together with HDAC1 in regulating meiotic exit. The second factor identified by co-IP MS analysis, Zfp541, is evolutionally 226 conserved (Fig. S3) and specifically expressed in testis (Fig. 6A). Further, the expression 227 begins around PND10 -12 and was then continuously detected with increasing signal 228 intensity at PND 28 (Fig. 6B), reminiscent of Kctd19 rtPCR (Fig. 1B). The mouse ZFP541 229 protein comprises 1363 amino-acid residues and has five C2H2 type zinc finger motifs, one 230 12 ELM2 domain, and one SANT domain based on SMART software (20), indicating 231 KCTD19/ZFP541 binds DNA. To reveal the function of ZFP541 and its relationship with 232 KCTD19, we analyzed Zfp541 KO phenotype with chimeric mice (chimeric analysis) (4) (5). 233 To disrupt gene function completely and minimize an effect on a juxtapose gene, 234 Napa, we designed two sgRNAs targeting the sequence upstream of the start codon and 235 intron 8 (Fig 6C), and transfected embryonic stem (ES) cells expressing EGFP (28) with two 236 pairs of sgRNA/Cas9 expressing plasmids (pair 1: gRNA 1 and 3; pair 2: gRNA 2 and 4; Fig  237   6C). We Screened 32 clones for each pair, and obtained 13 and 11 mutant clones with 238 biallelic deletion for pair 1 and 2. Accounting for ES cell quality and off-target cleavages, we 239 produced chimeric mice with one ES cell clone from pair 1 (1 -3 #2) and pair 2 (2 -4 #3) 240 ( Fig 6D and E). 241 First, we examined spermatogenesis with HePAS staining of testicular sections. 242 Almost no round spermatids with GFP were observed in chimeric mice (Fig 6F), as seen in 243 Kctd19 del/del testis sections. Zfp541 deficient spermatocytes were eliminated by apoptosis in 244 stage X -XII seminiferous tubules without reaching metaphase ( Fig 6G). Next, we 245 performed immunostaining with the antibodies against KCTD19. The KCTD19 intensity 246 became weaker, although not lost, in the nuclei of Zfp541 deficient spermatocytes than that 247 of adjacent WT spermatocytes ( Fig 6H). On the other hand, the immunofluorescence 248 intensity of HDAC1 was comparable between Zfp541 deficient and WT spermatocytes (Fig  249   6I). Finally, we examined the DNA damage response and synapsis in a XX/XY (Host/ES) 250 chimeric male mouse (29), in which all spermatocytes are derived from the mutant ES cells 251 ( Fig. S4A and S4B). Zfp541 deficient spermatocytes initiated DSBs in the leptotene/zygotene 252 stage and resolved the breaks in the early pachytene stage (Fig. 6J) We also detected CUL9 and DNTTIP1 in the IP-MS analysis with rabbit-generated anti-295 KCTD19 antibody, albeit not with the rat antibody. These factors can be excellent targets in 296 future research because knockdown experiments from other groups showed that CUL9 297 protects mouse eggs from aneuploidy (40)  In summary, our results showed that KCTD19 associates with ZFP541 and HDAC1 306 and are essential for meiotic exit. Further comparable studies will unveil the exact functions 307 16 of KCTD19 and ZFP541, which will give some insight into the molecular mechanism in male 308 meiosis. Kctd19-ΔPOZ, respectively. The genotyping primers (GeneDesign, Osaka, Japan) and 348 amplification conditions are available in Table S1.  Table S1.

Generation of Zfp541 KO ES cells and chimeric mice. 362
Zfp541 KO embryonic stem (ES) cells were generated using methods previously 363 described (5). Briefly, EGR-G01 ES cells were transfected with two pX459 plasmids 364 (Addgene plasmid #62988) with the target sequences (Table S1) Sepharose 4B (GE Healthcare). The purified KCTD19 protein with a complete adjuvant was 394 injected into female rats. After 17 days of injection, lymphocytes were collected from iliac 395 lymph nodes and hybridomas were generated (44,45

Fertility analysis of KO mice 418
To examine fertility, sexually mature male mice were housed with wild-type females 419 (B6DF1) for at least three months. Both plug and pup numbers were recorded at approximately 420 10 AM to determine the number of copulations and litter size.

Morphological and histological analysis of testis 435
To observe testis gross morphology and measure testicular weight, 11-12 week-old 436 male mice were euthanized after measuring their body weight. The whole testis was observed 437 using BX50 and SZX7 (Olympus, Tokyo, Japan) microscopes. For histological analysis, testes

Apoptosis detection in testicular section 448
TdT-mediated dUTP nick end labeling (TUNEL) staining was carried out with In Situ 449

22
Apoptosis Detection Kit (MK500, Takara Bio Inc., Shiga, Japan), according to the 450 manufacturer's instruction. Briefly, testes were fixed with Bouin's fixative, embedded in paraffin, 451 and sectioned (5 µm). After paraffin removal, the slides were boiled in citrate buffer (pH 6.0; 452 1:100; ab93678, abcam, Cambridge, UK) for 10 min and incubated in 3% H2O2 at room 453 temperature for 5 min for endogenous peroxidase inactivation, followed by a labeling reaction 454 with TdT enzyme and FITC-conjugated dUTP at 37°C for 1 h. 455 The antibodies used in this study are listed in Table S2. For preparing metaphase chromosome spreads, seminiferous tubules were 501 unraveled using forceps in ice-cold PBS and transferred to a 1.5-mL tube with 1 mL of 502 accutase (12679-54, Nacalai Tesque), followed by clipping the tubules, and a 5 min incubation 503 at room temperature. After filtration with a 59 µm mesh and centrifugation, the cells were 504 resuspended in 8 mL of hypotonic solution [1% sodium citrate] and incubated for 5 min at room 505 temperature. The suspension was centrifuged and 7 mL of supernatant was aspirated. The 506 cells were then resuspended in the remaining 1 mL of supernatant and 7 mL of Carnoy's 507 Fixative (75 % Methanol, 25% Acetic Acid) were added gradually while shaking. After 2 508 washes with Carnoy's Fixative, the cells were resuspended ~ 0.5 mL of Carnoy's Fixative and 509 dropped onto a wet glass slide. The slide was stained with Giemsa Stain Solution (079-04391, 510 wako) and observed using a BX53 (Olympus) microscope. 511 512

Immunostaining of metaphase I cells 513
For cytological analysis of metaphase I cells, seminiferous tubule squashes were 514 performed as previously described (47). In brief, seminiferous tubules were incubated in 515 fix/lysis solution [0.1 % TritonX-100, 0.8 % PFA in PBS] at room temperature for 5 min. Tubule 516 bunches were then put on glass slides with 100 µL of fix/lysis solution, minced into 1.0 ~ 3.0 517 mm segments with forceps, and arranged so that no tubule segment overlaped. After removing 518 the excess amount of fix/lysis solution, a coverslip and pressure was applied to disperse cells, 519 followed by flash freezing in liquid nitrogen for 15 sec, and removing the coverslip with forceps 520 and a needle. For longer storage, the slide glasses were kept at -80 ºC with the coverslip. 521

25
The slides were blocked and permeabilized in 10% goat serum and 0.1% Triton X-522 100 for 20 min in PBS, and incubated with primary antibody overnight at 4°C. After incubation 523 with AlexaFlour 488/546-conjugated secondary antibody (1:200) at room temperature for 1 h, 524 samples are counterstained with Hoechst 33342 (1:2000) and mounted with Immu-Mount. Z-525 stack images were taken using a BZ-X700 (kyence, Osaka, Japan) microscope and stacked 526 using ImageJ software. The antibodies used in this study are listed in Table S2. 527 528

Immunoprecipitation and mass spectrometry analysis 529
Proteins from testis were extracted using NP40 lysis buffer [50 mM Tris-HCl (pH7.5), 530 150 mM NaCl, 0.5% NP-40, 10% Glycerol]. Protein lysates were mixed with Dynabeads 531 Protein G (Thermo)-conjugated with 2.0 μg of antibody. The immune complexes were 532 incubated for 1 h at 4°C and washed 3 times with NP40 lysis buffer. Co-immunoprecipitated 533 products were then eluted with 18 µL of 100 mM Gly-HCl (pH2.5) and neutralized with 2µL of 534 1 M Tris. The antibodies used in this study are listed in Table S2. Half of the eluted amount 535 was subjected to SDS-PAGE and silver staining (06865-81, Nacalai Tesque). The remaining 536 half amount was subjected to mass spectrometry (MS) analysis. 537 The proteins were reduced with 10 mM dithiothreitol (DTT), followed by alkylation with 538 55 mM iodoacetamide, and digested by treatment with trypsin and purified with a C18 tip (GL-539 Science, Tokyo, Japan). The resultant peptides were subjected to nanocapillary reversed-540 phase LC-MS/MS analysis using a C18 column (25 cm × 75 um, 1.6 µm; IonOpticks, Victoria, 541 Australia) on a nanoLC system (Bruker Daltoniks, Bremen, Germany) connected to a tims 542 TOF Pro mass spectrometer (Bruker Daltoniks) and a modified nano-electrospray ion source 543 (CaptiveSpray; Bruker Daltoniks). The mobile phase consisted of water containing 0.1% formic 544 acid (solvent A) and acetonitrile containing 0.1% formic acid (solvent B). Linear gradient elution 545 26 was carried out from 2% to 35% solvent B for 18 min at a flow rate of 400 nL/min. The ion 546 spray voltage was set at 1.6 kV in the positive ion mode. Ions were collected in the trapped 547 ion mobility spectrometry (TIMS) device over 100 ms and MS and MS/MS data were acquired 548 over an m/z range of 100-1,700. During the collection of MS/MS data, the TIMS cycle was 549 adjusted to 1.1 s and included 1 MS plus 10 parallel accumulation serial fragmentation 550 (PASEF)-MS/MS scans, each containing on average 12 MS/MS spectra (>100 Hz), and 551 nitrogen gas was used as the collision gas. 552 The resulting data were processed using DataAnalysis version 5.1 (Bruker Daltoniks), 553 and proteins were identified using MASCOT version 2.6.2 (Matrix Science, London, UK) 554 against the SwissProt database. Quantitative value and fold exchange were calculated by 555 Scaffold4 (Proteome Software, Portland, OR, USA) for MS/MS-based proteomic studies. 556 557

Chimeric analysis 558
For distinguishing ESC-derived germ cells, GFP was stained by immunofluorescence 559 or immunohistochemistry. The antibodies used in this study are listed in Table S2.

Declaration of interests 567
The authors declare no competing interests. 568 569

Data availability statement 570
The authors declare that the data that support the findings of this study are 571 available from the corresponding author upon request.