CRISPR/Cas9-mediated genome editing reveals 12 testis-enriched genes dispensable for male fertility in mice.

Gene expression analyses suggest that more than 1000-2000 genes are expressed predominantly in mouse and human testes. Although functional analyses of hundreds of these genes have been performed, there are still many testis-enriched genes whose functions remain unexplored. Analyzing gene function using knockout (KO) mice is a powerful tool to discern if the gene of interest is essential for sperm formation, function, and male fertility in vivo. In this study, we generated KO mice for 12 testis-enriched genes, 1700057G04Rik, 4921539E11Rik, 4930558C23Rik, Cby2, Ldhal6b, Rasef, Slc25a2, Slc25a41, Smim8, Smim9, Tmem210, and Tomm20l, using the clustered regularly interspaced short palindromic repeats /CRISPR-associated protein 9 (CRISPR/Cas9) system. We designed two gRNAs for each gene to excise almost all the protein-coding regions to ensure that the deletions in these genes result in a null mutation. Mating tests of KO mice reveal that these 12 genes are not essential for male fertility, at least when individually ablated, and not together with other potentially compensatory paralogous genes. Our results could prevent other laboratories from expending duplicative effort generating KO mice, for which no apparent phenotype exists.

INTRODUCTION

Fertilization is the union of two gametes, spermatozoa and eggs, which conveys their genetic information to the next generation. Abnormal formation or function of these gametes leads to infertility with male factors contributing to nearly 50% of overall cases.12 Spermatozoa, male haploid gametes, are produced through spermatogenesis in the seminiferous tubules of the testis. During spermatogenesis, diploid spermatocytes undergo meiosis to produce round spermatids and then go through spermiogenesis that is the acquisition of specialized morphology and function. During spermiogenesis, round spermatids pack their haploid genome to form the sperm heads that contain an exocytotic vesicle called the acrosome, and they generate thread-like flagella that function as the motility apparatus for the resulting spermatozoa.345

Knockout (KO) mice are widely used to understand male fertility in vivo because there is no culture system that produces fully functional spermatozoa in vitro. KO mice provide a clear answer to whether or not a gene of interest is essential for male fertility, and because of the high sequence similarity between mouse and human genomes, also provide profound insight into the potential physiological function and functional requirement of the orthologous genes in human. Previously, it has been time consuming and costly to generate KO mice with the conventional method that utilizes homologous recombination in embryonic stem cells and subsequent generation of chimeric mice, but the emergence of genome editing technology such as the clustered regularly interspaced short palindromic repeats/CRISPR-associated protein 9 (CRISPR/Cas9) system makes it possible to generate KO mice in a short period of time with relatively low costs, which can be utilized in the field of andrology as well.6

In mice and humans, it has been reported that more than 1000–2000 genes are expressed predominantly in the testis.789 These genes are speculated to play essential roles in sperm formation and/or function because of their restricted expression in the testes. While a considerable number of these genes have been discovered over the years to be essential for male fertility,610 an even greater number of these genes have been found to be not essential for male fertility per se, which might be due to functional redundancy.81112131415 Of note, we recently reported that four genes (fertilization-influencing membrane protein [Fimp], sperm–oocyte fusion required 1 [Sof1], transmembrane protein 95 [Tmem95], and sperm acrosome associated 6 [Spaca6]) are essential for sperm–egg fusion,161718 which is astonishing considering that no other male factors essential for fusion have been found for 15 years since izumo sperm–egg fusion 1 (IZUMO1), the first male fusion factor, was discovered.19 These results indicate that screening testis-enriched genes in an organismal model with the CRISPR/Cas9 system is imperative to identify not only genes dispensable for male fertility but also genes that play critical roles in sperm formation and/or function.

Here, we deleted 12 mouse genes that are predicted to be expressed predominantly in the testis with the CRISPR/Cas9 system and revealed that all 12 genes were dispensable for male fertility. These data could prevent unnecessary expenditures and efforts by others.

MATERIALS AND METHODS

Animals

All animal experiments performed in this study were approved by the Institutional Animal Care and Use Committees of Osaka University (Osaka, Japan), National Cerebral and Cardiovascular Center (Osaka, Japan), and Kansai Medical University (Osaka, Japan) in compliance with the guidelines and regulations for animal experiments. In this study, all B6D2F1 and ICR mice were purchased from CLEA Japan, Inc. (Tokyo, Japan), Japan SLC, Inc. (Shizuoka, Japan), or Shimizu Laboratory Supplies (Kyoto, Japan).

Digital PCR

Digital PCRs (heatmaps) depicting average transcript per million values across various mouse and human reproductive and nonreproductive tissues were conducted as previously performed using our previously reported data.15 Briefly, sequences for different tissues were downloaded from Sequence Read Archives, trimmed using TrimGalore, and aligned to the human genome (GRCh38) or mouse genome (GRCm38) by HISAT2.20 Feature Counts was used to quantify gene expression in each tissue, and RUVr21 was used to batch correct tissues by removing unwanted variation. EdgeR was used to determine differential gene expression for each nonreproductive tissue against each reproductive tissue.22

Reverse transcription PCR (RT-PCR)

Mouse cDNA was prepared from multiple adult tissues of C57BL/6N using TRIzol (Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer's protocol. The obtained RNA was immediately reverse transcribed to cDNA with SuperScript IV first-strand synthesis system (Thermo Fisher Scientific) using oligo (dT) as primers. PCR primers and amplification conditions for each gene are shown in .

Phylogenetic analysis

The phylogenetic trees of candidate genes were created using the neighbor-joining method with the GENETYX software (GENETYX, Tokyo, Japan). For the amino acid sequences, we chose cattle, dog, horse, human, mouse, and pig for mammals and chicken for birds.

Generation of KO mice with the CRISPR/Cas9 system

All KO mouse lines in this study were produced with the CRISPR/Cas9 system. CRISPRdirect software was used to find off-target sequences when we chose gRNAs ().23 To generate KO mice, we performed electroporation using zygotes as described previously.2425 Cas9 protein (Thermo Fisher Scientific or Integrated DNA Technologies, Coralville, IA, USA) was mixed with crRNAs, tracrRNA (Integrated DNA Technologies or Merck, Darmstadt, Germany), and Opti-MEM (Thermo Fisher Scientific). This solution was incubated at 37°C to generate the gRNA/Cas9 ribonucleoprotein (RNP) complex, and the obtained complex was electroporated into fertilized oocytes using NEPA21 Super Electroporator (Nepagene, Chiba, Japan). The treated embryos that developed to the 2-cell stage were transplanted into the oviducts of pseudopregnant ICR females at 0.5 days after mating with vasectomized males. Pups were obtained by natural or cesarean section, and subsequent sibling crosses were performed to obtain homozygous KO mice. Genotyping was performed through PCR using primers listed in . All gene-modified mice generated in this study will be made available to other investigators through either the RIKEN BioResource Research Center, Tsukuba, Ibaraki, Japan, or the Center for Animal Resources and Development (CARD), Kumamoto University, Kumamoto, Japan.

Morphological and histological analysis of testes

After euthanasia, testes were dissected. After measuring the testicular weight, the testes were fixed in Bouin's solution (Polysciences, Warrington, PA, USA), embedded in paraffin, sectioned at a thickness of 5 μm on a Microm HM325 microtome (Microm, Walldorf, Germany), 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 then stained with Mayer's hematoxylin solution (FUJIFILM Wako) before imaging and observed using an Olympus BX53 microscope equipped with an Olympus DP74 color camera (Olympus, Tokyo, Japan).

Analysis of morphology and motility of spermatozoa

Spermatozoa from cauda epididymis were suspended in TYH medium,26 incubated at 37°C for either 10 min or 120 min, diluted, and then placed on MAS-coated glass slides (Matsunami Glass, Osaka, Japan) for morphology assessment using an Olympus BX53 microscope or placed in glass chambers for sperm motility analysis using the CEROS II sperm analysis system (software version 1.5; Hamilton Thorne Biosciences, Beverly, MA, USA).

Fertility analysis of KO lines

Sexually mature wild type (WT) or KO male mice were caged individually with one to three six-week-old female mice for at least eight weeks (except for small integral membrane protein 8 [Smim8]). Male mice were removed after the mating period, and females were kept for another three weeks to count the final litters. The numbers of pups and copulation plugs were counted every morning.

Statistical analyses

Statistical difference was determined using the Welch's t-test by Microsoft Office Excel (Microsoft Corporation, Redmond, WA, USA). Differences were considered statically significant if the P < 0.05. Data represent the mean ± standard deviation (s.d.).

RESULTS

Expression patterns of candidate genes in mouse and human

To examine the mouse and human reproductive and nonreproductive tissue expression patterns of the 12 genes in this study, we performed digital PCR as described.15 Except for lactate dehydrogenase A-like 6B (Ldhal6b) and small integral membrane protein 9 (Smim9), all genes exhibited enriched expression in mouse testis (). Because no data were found for Ldhal6b and whole tissue expression signal was weak for Smim9 in the digital PCR, we performed RT-PCR for Ldhal6b and Smim9 using mouse tissues and found that the expression of both genes was enriched in the testis (). Further, we confirmed the expression of these genes in human testis with digital PCR (), although there were no expression data for cortexin domain containing 2 (CTXND2), the human homolog of 4930558C23Rik. All the genes were conserved in humans ().

Expression patterns of specific genes in various tissues. (a) Digital PCR shows the average transcripts per million (TPM) value per tissue per gene from published mouse RNA-seq datasets.15 White = 0 TPM, Black ≥30 TPM. Hprt expression is shown as housekeeping control. (b) Expression patterns of Ldhal6b and Smim9 in mouse tissues using RT-PCR. Actb expression is shown as housekeeping control. MW: molecular weight marker. (c) Digital PCR shows the average TPM value per tissue from published human RNA-seq datasets.15 PLSCR1 and PLSCR2 are orthologs of mouse 1700057G04Rik. C1orf141 is an ortholog of mouse 4921539E11Rik. White = 0 TPM, Black ≥30 TPM. GAPDH expression is shown as housekeeping control. RT-PCR: reverse transcription PCR; Hprt: hypoxanthine guanine phosphoribosyl transferase; Actb: actin beta; C1orf141: chromosome 1 open reading frame 141; GAPDH: glyceraldehyde-3-phosphate dehydrogenase; PLSCR1: phospholipid scramblase 1; PLSCR2: phospholipid scramblase 2. The expansions of the other genes have been described in the notes of

Generation of KO mice

To investigate the function of candidate genes in male reproduction, the 12 genes were ablated individually using the CRISPR/Cas9 system. For each of the genes, either the entire protein coding region or almost all of the protein coding regions were deleted through nonhomologous end joining to ensure that each gene was completely ablated and that no partially functional transcript may still be expressed (). Genome-editing efficiency (the number of mutant pups divided by the total number of genotyped pups) ranged from 13.6% to 77.8%. The efficiency of genome-editing and mutation details are summarized in .

Phenotypic analysis of 4921539E11Rik, 4930558C23Rik, and transmembrane protein 210 (Tmem210) KO mouse lines

Herein, as an example of in vivo functional analysis, we performed a detailed phenotypic analysis on three KO lines: 4921539E11Rik, 4930558C23Rik, and Tmem210 (Figures –).

Phenotypic analysis of 4921539E11Rik KO male mice. (a) KO strategy of 4921539E11Rik. Two gRNAs (#2 and #8) were designed to target intron 1 and exon 9. Fw1/2: forward primers for genotyping; Rv1/2: reverse primers for genotyping. (b) Genotyping of 4921539E11Rik KO mice through PCR using primer sets Fw1-Rv1 and Fw2-Rv2, and subsequent Sanger sequencing. MW: molecular weight marker. (c) Comparison of testis size between WT and 4921539E11Rik KO mice. Scale bar = 3 mm. (d) Average testicular weight. Number of males = 3 each. (e) Histological analysis of testes in WT and 4921539E11Rik KO mice. Scale bar = 100 μm. (f) Spermatozoa collected from the cauda epididymis of WT and 4921539E11Rik KO mice. Scale bar = 50 μm. (g) Percentages of motile spermatozoa in WT and 4921539E11Rik KO mice. Sperm motility was measured at 10 min and 120 min after incubation in TYH medium. Number of males = 3 each. KO: knockout; WT: wild type; NS: not significant.

Phenotypic analysis of 4930558C23Rik KO male mice. (a) KO strategy of 4930558C23Rik. Two gRNAs (#3 and #12) were designed to target before and after the protein-coding region. Fw1: forward primer for genotyping; Rv1: reverse primer for genotyping. (b) Genotyping of 4930558C23Rik KO mice through PCR using primer sets Fw1-Rv1 and subsequent Sanger sequencing. MW: molecular weight marker. (c) Comparison of testis size between WT and 4930558C23Rik KO mice. Scale bar = 3 mm. (d) Average testicular weight. Number of males = 3 each. (e) Histological analysis of testes in WT and 4930558C23Rik KO mice. Scale bar = 100 μm. (f) Spermatozoa collected from the cauda epididymis of WT and 4930558C23Rik KO mice. Scale bar = 50 μm. (g) Percentages of motile spermatozoa in WT and 4930558C23Rik KO mice. Sperm motility was measured at 10 min and 120 min after incubation in TYH medium. Number of males = 3 each. KO: knockout; WT: wild type; NS: not significant.

Phenotypic analysis of Tmem210 KO male mice. (a) KO strategy of Tmem210. Two gRNAs (#1 and #2) were designed to target within the 5’-UTR and within exon 4. Fw1: forward primer for genotyping; Rv1/Rv2: reverse primers for genotyping. (b) Genotyping of Tmem210 KO mice through PCR using primer sets Fw1-Rv1 and Fw1-Rv2 and subsequent Sanger sequencing. MW: molecular weight marker. (c) Comparison of testis size between WT and Tmem210 KO mice. Scale bar = 3 mm. (d) Average testicular weight. Number of males = 3 each. (e) Histological analysis of testes in WT and Tmem210 KO mice. Scale bar = 100 μm. (f) Spermatozoa collected from the cauda epididymis of WT and Tmem210 KO mice. Scale bar = 50 μm. (g) Percentages of motile spermatozoa in WT and Tmem210 KO mice. Sperm motility was measured at 10 min and 120 min after incubation in TYH medium. Number of males = 3 each. Tmem210: transmembrane protein 210; KO: knockout; WT: wild type; NS: not significant.

4921539E11Rik KO mice were generated using two gRNAs designed upstream of the start codon in intron 1 and near the stop codon in exon 9 (). Their genotypes were confirmed by genomic PCR () using primers shown in . Sanger sequencing showed that the KO mouse had a 53 580 bp deletion (). Homozygous KO mice were viable and no overt abnormalities were found. There were no significant differences in the testis appearance (), testis weight (), and testis histology (). Further, we found no overt abnormalities in sperm morphology (). We also checked the percentage of motile spermatozoa using the computer-assisted sperm analysis (CASA) system and found no differences between the control and KO mice ().

To disrupt 4930558C23Rik, we designed two gRNAs, one before the start codon and the second after the stop codon of exon 2 (). To identify the mutant allele, we performed PCR using primers presented in (). The resulting mutant allele carrying a 597-bp deletion was identified by PCR and Sanger sequencing (). 4930558C23Rik KO mice exhibited normal testicular size (Figure and ) and histology (). Further, sperm morphology and percentage of motile spermatozoa analyzed with the CASA system were comparable between the control and KO mice (Figure and ).

Tmem210 was mutated using two gRNAs, one upstream of exon 1 and the other near the stop codon of exon 4. Primers for detecting the mutant alleles are presented in (). Mutant mice carrying 1016 bp deletion were confirmed by PCR and Sanger sequencing (). We found that the testis size, weight, and histology of KO males were comparable to those of the control mice (Figure –). In addition, there were no significant differences between the control and KO males in sperm morphology and the percentage of motile spermatozoa (Figure and ).

Fertility results of 12 testis-enriched KO mouse models

To examine the fertility of the KO males, the mice were caged with one-to-three WT females for more than 8 weeks (except for Smim8). We observed more than 13 deliveries, and all the KO males were able to sire normal numbers of pups during the mating period. There were no significant differences in average liter size between the control and KO males ().

DISCUSSION

Approximately 15% of couples face the problem of infertility. While it is estimated that half of the cases are attributed to the male,2 much of the pathogenic mechanisms underlying the male factor have not been clarified.27 Screening genes that play essential roles in male reproduction using mouse models have led to a profound increase in our understanding of the etiology of male infertility.1028 In this study, we mutated 12 genes in mice and found that these 12 genes are not essential, at least individually, for male fertility, suggesting that these genes may also not play critical roles in human male reproduction. Using this rapid CRISPR/Cas9 approach, we have reported over the past 5 years that 125 testis-restricted or epididymis-enriched genes are not essential for male fertility.8111213141525 Thus, this new study with an additional 12 genes adds to the growing list of proteins that are not feasible targets for male contraceptives.

Among the genes that we analyzed in this study, the functions of some of their paralogs have already been previously reported. For example, Chibby family member 1 (Cby1), a paralog of Cby2, has been reported to be involved in intraflagellar transport (IFT) in cilia.29 IFT is a bidirectional transport of protein complexes along axonemal microtubules, which is essential for the formation of cilia and sperm flagella.30 Tracheal epithelial cells lacking Cby1 show abnormal intracellular localization of IFT88 and IFT20.29 Because depletion of IFT88 or IFT20 causes abnormal formation of sperm axoneme and male infertility,3132 Cby2 that is expressed predominantly in the testis was hypothesized to be essential for IFT in sperm axoneme and male fertility. However, we reveal that Cby2 KO male mice are fertile. Because the expression database suggests that Cby1 and Cby3 are also expressed in the testis, these paralogs may function in a compensatory manner.

SLC25 family proteins are localized to the inner mitochondrial membrane and play roles in transporting solutes across the inner membrane.33 It has been reported that solute carrier family 25 member 2 (SLC25A2) transports arginine, lysine, ornithine, and asymmetric dimethyl L-arginine,34 while solute carrier family 25 member 41 (SLC25A41) transports ATP-Mg/Pi.35 Translocase of outer mitochondrial membrane 20-like (TOMM20L) is also predicted to be a mitochondrial protein that is localized in the outer membrane because of its sequence similarity to TOMM20, a mitochondrial outer membrane marker.36 SLC25A2, SLC25A41, and TOMM20L are all listed in the MitoCarta3.0 that is an inventory of genes encoding proteins with strong support of mitochondrial localization.37 Mammalian spermatozoa possess the mitochondrial sheath that is localized in the midpiece of the flagella to provide structural support.38 Furthermore, sperm mitochondria are suggested to be involved in various sperm functions such as sperm motility, hyperactivation, sperm capacitation, and acrosome reaction,39 and disruption of mitochondrial sheath leads to male infertility.40 Since Slc25a2, Slc25a41, and Tomm20l are highly expressed in the testis, we hypothesized that these genes might be involved in sperm mitochondrial function. However, our functional analyses reveal that the individual ablation of these genes does not impair male fertility.

We also mutated proteins that may have enzymatic activities. RAS and EF hand domain containing (Rasef) was identified as a Rab GTPase with specific expression in mouse testis.41 Rab GTPase hydrolyzes GTP to GDP and controls intracellular membrane trafficking. LDHAL6B is a homolog of LDHA that catalyzes the conversion of L-lactate to pyruvate, which is involved in glycolysis.42 LDHAL6B is also listed in the mitochondrial protein inventory, MitoCarta3.0.37 1700057G04Rik is a homolog of PLSCR1 that catalyzes redistribution of plasma membrane phospholipids between the inner and outer leaflets.43 In addition to these genes, we found five more genes that were not essential for male fertility: 4921539E11Rik without a known functional domain, 4930558C23Rik containing a cortexin domain,44 Smim8 and Smim9 each containing one transmembrane domain, and Tmem210 containing two transmembrane domains.

In summary, we identified 12 genes that are not essential for male fertility in mice. We focused purely on pups sired as an assessment of male fertility in 9/12 lines and did not investigate testis histology or sperm function. Therefore, there might be subtle abnormalities in sperm structure or function in these lines that do not affect fertility. All lines will be available as bioresources, which can be analyzed further by other researchers. Recent developments in sequencing technology such as whole exome sequencing make it possible to identify causative genes of human male infertility within a short period of time.454647 In the event that whole exome sequencing of infertile men reveals mutations in these 12 genes, these genes may be deprioritized and emphasis places on other potentially causative gene mutations or agents.

CONCLUSION

In this study, we generated KO mice for 12 testis-enriched genes using the CRISPR/Cas9 system. The KO mice displayed normal fertility, suggesting that these 12 genes are dispensable for male fertility, at least when individually ablated in mice. This report is anticipated to prevent unnecessary duplicative effort in generating KO mice with no apparent phenotypes.

AUTHOR CONTRIBUTIONS

All authors designed the study and analyzed the data. YO, HM, KS, YF, KT, and TXG performed experiments. YO, HM, and MI wrote the manuscript draft. All authors revised, read, and approved the final manuscript.

COMPETING INTERESTS

All authors declared no competing interests.

Phylogenetic analysis of the 12 testis-enriched genes in mammals and birds.

KO strategy showing the location of gRNAs and genotyping primers. KO: knockout.

Supplementary Table 1

Primer sequences and conditions used for RT-PCR of Ldhal6b and Smim9

Gene symbol	Primer set for WT	Annealing condition	Elongation condition	Band size (bp)
Ldhal6b	Fw: GAAGAGCCCGTTCTCCACAA	65˚C	72˚C	504
	Rv: AGAGTGGATCCCAAGCCTCT	30 s	30 s
Smim9	Fw: ATGAAGCCTCTGAAGCTGTTCTGC	60˚C	72˚C	414
	Rv: TCACCCACCAAAAAGCTTG	30 s	30 s
Actb	Fw: CATCCGTAAAGACCTCTATGCCAAC	60˚C	72˚C	171
	Rv: ATGGAGCCACCGATCCACA	30 s	30 s

Table 1

The sequences of guide RNA and deleted regions

Gene symbol	Guide RNAs	CRISPR/Cas9 efficiency, n/total (%)	CRISPR/Cas9 delivered gene deletion (intron, exon)a	Mutation (bp)
1700057G04Rik	Up: TTAGGATATCATCTAAACT	10/18 (55.6)	Up: gcttcttaggatatcatctaaactt	−3950
	Down: GAAACATCTGGAAACAGCT		Down: tagtctccagaaacatctggaaaca
4921539E11Rik	Up: TACCTGGAGCAGTAATAGAG	1/6 (16.7)	Up: gagaggaactctccccttatgccta	−53 580
	Down: ACCCCATTGGATAACCTGCT		Down: AACAACATGACCCCATTGGATAACC
4930558C23Rik	Up: CAGACACAGCTACCTCCACG	6/22 (27.3)	Up: GAGGTAGCTGTGTCTGGAAGACAGT	−597
	Down: ACCCCAGGGACATAAGATAT		Down: ggctatgagtttcaattgtcccata
Cby2	Up: CTGACTCTGGCTAGCCACGT	5/12 (41.7)	Up: gggcacagccagcagcaaccaatcc	−10 210
	Down: GCCCAGAAGCACGTGCCCGT		Down: ACTCTGCCCAGAAGCACGTGCCCGT
Ldhal6b	Up: CCTAGCAACGGCCGCCGTTG	3/22 (13.6)	Up: GTTGTGGCCTCAACTGTCTTCCTCA	−1281
	Down: GAAGCGGGGGTTCCGAAAGC		Down: TGTGGGGTGAAGCGGGGGTTCCGAA
Rasef	Up: AGCATGACTTCCTTGTGGCG	8/19 (42.1)	Up: acgccacaaggaagtcatgctagtt	−63 153
	Down: GCTTCAGGTTGCTGAGAGCG		Down: GTTCTTCCCCGCTCTCAGCAACCTG
Slc25a2	Up: CAGTGAGGTCTATGGCGGCT	5/15 (33.3)	Up: CGCCATAGACCTCACTGCGGGCGCA	−795+1
	Down: AGGGATGGCTCGAATCAGAG		Down: ATTCTGGATTGAAAGCCACTCTGAT
Slc25a41	Up: TTATCCAGCCTTGCGGTTTA	6/16 (37.5)	Up: ACCGCAAGGCTGGATAACCAGCCAC	−8811
	Down: ACTCGGATTAGAAAAGCGCT		Down: CACCCAGGACTCGGATTAGAAAAGC
Smim8	Up: TGGTTGTGTGCGCCCCTCGG	8/11 (72.7)	Up: CGCACACAACCACCTTATTTCGAGC	−2300
	Down: GTGAAGGGCATCGGTACATG		Down: GTGAAGGGCATCGGTACATGAGGAG
Smim9	Up: CCTCTGAAGCTGTTCTGCAT	29/39 (74.4)	Up: CATAGGACTGCTCTTGTGCCCTCTG	−12 405
	Down: GAAAATTCTGGAATCCTACT		Down: TGAATCTAATCAAGgtctgcctagt
Tmem210	Up: TTCCAATGAGTTCCATCTAG	7/9 (77.8)	Up: tAGAGGTGTGTGGTGGGGTTGGGGG	−1016
	Down: GTAGTGGAGAGCAACTAGGG		Down: TTCCACAGTAGTGGAGAGCAACTAG
Tomm20l	Up: TGCAGTGTCCTTGGGTCGCC	3/9 (33.3)	Up: gaaatccaggcgacccaaggacact	−483
	Down: GAGGTGCAGCAGACCCGCCA		Down: gtgaggtgcagcagacccgccaagg

aUppercases indicate exon sequences and lowercases indicate intron sequences. CRISPR/Cas9: clustered regularly interspaced short palindromic repeats/CRISPR-associated protein 9; Cby2: chibby family member 2; Ldhal6b: lactate dehydrogenase A-like 6B; Rasef: RAS and EF hand domain containing; Slc25a2: solute carrier family 25 member 2; Slc25a41: solute carrier family 25 member 41; Smim8: small integral membrane protein 8; Smim9: small integral membrane protein 9; Tmem210: transmembrane protein 210; Tomm20l: translocase of outer mitochondrial membrane 20-like. CRISPR/Cas9 efficiency was calculated by the number of mutant pups divided by the number of genotyped pups. CRISPR/Cas9 delivered gene deletion indicates 25 bp sequences upstream and downstream of the deleted regions

Supplementary Table 2

Primers and PCR conditions used for genotyping

Gene symbol	Primer set for WT	Annealing condition	Elongation condition	Band size (bp)	Primer set for KO
1700057G04Rik	Fw2:GGCAGGATTTCCAAGCACTG	58˚C	72˚C	362	Fw1:CTACCCCATGCAGCTGTCC
	Rv1:GACAACCTCAGAGCTAGGGC	30 s	30 s		Rv1:(as WT)
4921539E11Rik	Fw2:CAGCGGAGTCTTCAGTCCCG	60˚C	72˚C	667	Fw1:TCCCAATGGAGGAGCTAGAG
	Rv2:GTGACCTAAGCTTGATACCC	30 s	30 s		Rv1:GAGATGGCTCATCCTGGAAA
4930558C23Rik	Fw1:CTGTGCAGCTCCAGCTGACC	65˚C	72˚C	1104	(as WT)
	Rv1:TGTAGCATAGAGCTCCACCC	30 s	60 s
Cby2	Fw2:AGCAGCCACTTTCCTGTCTC	60˚C	72˚C	593	Fw1:ACAGGGCTGAACCCCTAACT
	Rv2:AGAGCTGGGGATGGAAATCT	30 s	30 s		Rv1:TGCACCATCATACCCTGCTA
Ldhal6b	Fw1:AAGAAGACCCAGTCTGTGCG	65˚C	72˚C	311	(as WT)
	Rv1:TATGCATGTGAGCCACCCAG	30 s	45 s
Rasef	Fw2:CAATGAGTGTGACTCCGAGG	60˚C	72˚C	429	Fw1:AAACCTTAGCCTAGTAATGTTGG
	Rv2:TCTTCATCCCTTATATCTGG	30 s	30 s		Rv1:ACAGGATAATTACAAACTACAGG
Smim8	Fw1:CCCCAGGGCTCTGTGAGTTCAAGG	65˚C	72˚C	616	Fw1:(as WT)
	Rv1:CAAGGGCTAGGATTATAGGTGCGTGCC	30 s	30 s		Rv2:GCCTTCAGCTGTAATGAGACTAAACTCACCAC
Smim9	Fw1:CACGACGACGGTGAGGGGTTATGC	65˚C	72˚C	573	Fw1:(as WT)
	Rv1:GCTACCTTTGTGAAACCTTCCCTTAATCCAC	30 s	30 s		Rv2:CCAAATTCCTGCCTCCTTGTGTGC
Slc25a2	Fw1:GGCTTTGGTTTGAGTAGCCT	65˚C	72˚C	452	(as WT)
	Rv1:CTGTAGATCCCACCACCAGC	30 s	45 s
Slc25a41	Fw2:TGCTCTCCTCCTTCCTCTCC	65˚C	72˚C	591	Fw1:CGTGGAGCTCATCTGCTAGG
	Rv2:GGGCCACGAAGAAGGAAGAA	30 s	30 s		Rv1:ATACACCCCATCTCCCAGCT
Tmem210	Fw1:TCCTCTGCTTGCCTCAATCT	60˚C	72˚C	809	Fw1:(as WT)
	Rv2:GGCCCGAAAGACACCAAT	30 s	30 s		Rv1:GTCCTGTCTTGGGGAATCTG
Tomm20l	Fw1: ATAAAGGGAAAAGCACCGGC	58˚C	72˚C	799	(as WT)
	Rv1: AATGTGTCCTTACCAGACCCA	30 s	30 s

Table 2

Male fertility of the 12 mutant mouse lines

Gene symbol	Genotype	Average litter size, mean±s.d.	Number of males	Number of delivery	Number of pups	Number of plugs	Mating period (week)
	Wild type	8.2±2.0	3	25	214	25	8
1700057G04Rik	−3950/−3950	7.7±1.1	4	19	145	ND	16–21
4921539E11Rik	−53580/−53580	8.8±2.4	3	24	210	30	8
4930558C23Rik	−597/−597	7.4±2.1	3	16	119	24	8
Cby2	−10210/−10210	8.7±2.4	3	24	209	29	8
Ldhal6b	−1281/−1281	9.1±1.9	3	25	227	28	8
Rasef	−63153/−63153	8.8±2.4	3	19	168	23	8
Slc25a2	−795+1/−795+1	9.3±1.8	3	20	185	24	8
Slc25a41	−8811/−8811	9.7±1.8	3	24	232	28	8
Smim8	−2300/−2300	9.4±1.4	4	16	151	16	6
Smim9	−12405/−12405	8.4±2.4	3	13	109	13	8
Tmem210	−1016/−1016	8.5±1.8	3	22	187	28	8
Tomm20l	−483/−483	8.0±0.5	3	15	120	ND	17–22

Statistical analysis of average litter size between WT and each KO mouse strain was performed by Welch’s t-test for unpaired observations and there was no significant difference. Cby2: chibby family member 2; Ldhal6b: lactate dehydrogenase A-like 6B; Rasef: RAS and EF hand domain containing; Slc25a2: solute carrier family 25 member 2; Slc25a41: solute carrier family 25 member 41; Smim8: small integral membrane protein 8; Smim9: small integral membrane protein 9; Tmem210: transmembrane protein 210; Tomm20l: translocase of outer mitochondrial membrane 20-like; s.d.: standard deviation; ND: not determined; WT: wild type; KO: knockout