A novel homozygous mutation in the DNAAF3 gene leads to severe asthenozoospermia and teratospermia

Abstract Primary ciliary dyskinesia (PCD) is an autosomal recessive genetic disorder characterized by ultrastructural defects in the cilia or flagella of cells, causing respiratory abnormalities, sinusitis, visceral transposition, and male infertility. DNAAF3 plays an important role in the assembly and transportation of axonemal dynein complexes in cilia or flagella and has been shown to be associated with PCD. To date, only two cases of PCD with infertility associated with DNAAF3 mutations have been reported, and no mouse models for this gene have been successfully constructed. This study was conducted on an infertile Chinese male patient with a history of bronchitis. Examination of the patient's semen revealed severe asthenozoospermia and teratospermia. Whole exome sequencing revealed a new homozygous loss‐of‐function DNAAF3 mutation. CRISPR‐Cas9 gene‐editing technology was used to construct the same mutation in C57/B6 mice, revealing that homozygous C57/B6 mice were characterized by severe hydrocephalus and early death. The results of this study expand the mutation spectrum of DNAAF3 and confirm its correlation with PCD pathogenesis. This study provides new insights on the mechanisms underlying male infertility related to DNAAF3 mutation and PCD.

arms diverge from nine pairs of peripheral doublet microtubules and are responsible for converting the chemical energy of adenosine triphosphate (ATP) into mechanical force, which generates power for the sperm flagella. 57][8] In addition, factors affecting sperm maturation or ejaculation processes may also impact sperm motility, such as reproductive tract infections, 9 varicocele, 10 lifestyle 11 and toxic substances, 12 among others.
Primary ciliary dyskinesia (PCD) is an autosomal recessive genetic disease that affects approximately 1:10000 people worldwide. 13,14It is characterized by ciliary dyskinesia and results in a variety of clinical manifestations, including chronic bronchitis, bronchiectasis, interstitial pneumonia, sinusitis, secretory otitis media, situs inversus and so on.Sperm flagella are a special type of cilia and an abnormal sperm flagellar structure can cause the sperm to lose motility.Therefore, PCD is often associated with male infertility. 15though the ciliary pathogenesis of the PCD respiratory phenotype has been widely studied, there are relatively few reports on the impact of PCD on male infertility. 16According to a systematic review of the validated monogenic causes of human male infertility by the International Male Infertility Genomics Consortium (IMIGC) in 2020, eight PCD-related genes are at least moderately linked to male infertility, namely, CCDC39, CCDC40, DNAAF2, DNAAF4, DNAAF6, LRRC6, RSPH3 and SPEF2. 17e DNAAF3 gene is located on human chromosome 19q13 and contains 12 exons.It encodes axonemal assembly factor 3, a protein required for the assembly of axonemal dyneins and the assembly and trafficking of the axonemal dynein complex. 18,19otein expression abnormalities caused by DNAAF3 gene mutations have been shown to lead to assembly defects in the axonemal dynein complex, further causing abnormal ciliary structure and function, which can lead to respiratory system abnormalities, sinusitis, visceral transposition, etc. 18,19 The dynein arm is a molecular motor for sperm flagellar movement.When DNAAF3 gene mutations cause structural abnormalities, the outer and inner dynein arms may be absent at the ultrastructural level, resulting in immotile sperm or severe asthenozoospermia, thus affecting male reproductive potential.However, there are very few reports on PCD in infertile males with DNAAF3 mutations and on their semen characteristics, with only two cases reported thus far. 19,20rthermore, no disease model has been established for this gene in mice.
In this study, we identified a novel DNAAF3 mutation in a Chinese male suffering with PCD and severe asthenozoospermia.Using CRISPR-Cas9 gene-editing technology, we developed a mouse model with homozygous mutations that mimicked the identified variation.This study expands the spectrum of DNAAF3 mutations and supplements the pool of male infertility cases attributed to DNAAF3 mutations.Moreover, using mouse models, this study confirmed its correlation with PCD pathology.These findings offer valuable insights into the mechanisms underlying male infertility associated with DNAAF3 and PCD.

| Patients
A 24-year-old male patient with primary infertility who was treated at the Reproductive Medicine Center of the First Affiliated Hospital of Sun Yat-sen University was enrolled in this study (Figure 1).The patient's parents were first cousins (consanguineous marriage).The patient also had a history of bronchitis.There was no clear acquired aetiology, such as orchitis, epididymitis, cryptorchidism, severe vari-

| Semen examination
The patient, following standard clinical semen collection requirements, abstained from ejaculation for 2-7 days and provided a semen sample through masturbation.According to the laboratory manual on Human Semen Analysis and Processing by the World Health Organization (WHO, 5th edition), the obtained semen sample was analysed using computer-assisted sperm analysis (CASA) to determine sperm quantity and vitality parameters.Subsequently, semen smears and staining were performed to evaluate sperm morphology.Additionally, sperm vitality was measured using the hypotonic swelling (HOS) test (normal range: ≥50% 3,21 ).Tests for genital tract inflammation were conducted, including semen leukocyte peroxidase (POX) staining (normal range: ≤1 × 10 6 /mL, 3,21 ) and seminal plasma elastase detection (normal range: <290 ng/ mL 22 ).

| Whole exome sequencing
Peripheral blood samples were used for the extraction, fragmentation, and library construction of genomic DNA (gDNA) using the QIAamp DNA Blood Mini Kit (QIAGEN), Covaris S220 focused ultrasonicator (Covaris) and NEBNext Ultra II DNA Library Prep Kit for Illumina (NEB), respectively.A TruSeq Exome Enrichment Kit (Illumina) was used to enrich the coding regions and intron/ exon boundaries.Sequencing was performed on the NextSeq 550DX (Illumina) platform using the Truseq SBS Kit V4-HS reagent kit (Illumina).The sequence reads were aligned to the reference genome (GRCh37/hg19) using the Burrows-Wheeler Aligner.

| Sanger Sequencing
After designing primers corresponding to the mutation site of DNAAF3 (NM_001256714), the target sequence was amplified by PCR using genomic DNA from the patient as a template.PCR products were Sanger sequenced using conventional methods. 23

| Mouse model construction
The mouse model construction process is illustrated in Figure 2. The mice were kept in a facility meeting specific pathogen free standards, with a 12-hour light/dark cycle.In vitro-transcribed sgRNA (Table 1) and donor vectors were constructed and microinjected into fertilized C57BL/6JGpt mouse zygotes with Cas9, sgRNA and a donor for homologous recombination.Seven-base insertions (c.871_872ins CAGGCTC, p.Q291fs) in DNAAF3 were generated to produce the same mutation as that in the patient.After injection, the surviving zygotes were transferred to pseudopregnant female mice, and the mice were allowed to give birth.The F0 progeny born from the recipient mice were tail-and toe-clipped at 5-7 days old, genotyped by PCR and sequenced using the primers shown in Table 2 to confirm their genotype.After reaching sexual maturity, F0 heterozygous mice were crossed with wild-type mice.F1 progeny were genotyped by PCR and sequenced to confirm their genotype.Heterozygous F1 mice were crossed to obtain homozygous C57 mice in the F2 generation.
After crossing heterozygous C57 mice with homozygous 129 wild-type mice, the F1 progeny were used to extract genomic DNA for PCR and sequenced to confirm their genotype.Heterozygous F1 mice carrying both C57 and 129 genetic backgrounds were crossed to obtain homozygous mice in the F2 generation that possessed both C57 and 129 genetic backgrounds simultaneously.Genotyping was performed at 5-7 days after birth, and no deaths were observed at that time.The lung and brain tissues of mice were stained with haematoxylin and eosin (H&E).Additionally, the testicular tissues were stained with both H&E and periodic acid Schiff (PAS) stain.

| Statistical analysis
Statistical analyses were performed using SPSS software (version 22.0; IBM Corp., Chicago, IL, USA).Student's t-test was used to compare body weight and testicular volume between wild-type and mutant mice.The Pearson chi-squared test was used to determine any statistical differences between actual and expected genotype frequencies in the F2 generation.p < 0.05 was considered statistically significant.

| Semen examination
Table 3 provides a detailed display of the patients' semen examination results.Semen volume, pH, total sperm count and sperm concentration were normal.However, severe asthenozoospermia was observed, with a sperm progressive motility (PR) of 0% and F I G U R E 2 Workflow for the construction of DNAAF3 homozygous mutant mice.CRISPR/Cas9 technology was used to modify C57 zygotes, introducing a 7-nucleotide insertion (c.871_872ins CAGGCTC, p.Q291fs) into the DNAAF3 gene to create the same mutation as that observed in the patient.After zygote manipulation, the embryos were transferred, resulting in heterozygous F1 offspring.Heterozygous F1 mice were then bred to obtain homozygous F2 offspring.In the F2 generation, the genotypic distribution of wild-type (WT) mice, heterozygotes, and homozygotes deviated from the Mendelian inheritance ratios with percentages of 15, 57% and 27%, respectively.To obtain hybrid mice with both C57 and 129 genetic backgrounds, heterozygous F1 mice were crossed with WT 129 strains.The F1 generation mice obtained from this cross possessed both C57 and 129 genetic backgrounds and were heterozygous for the targeted mutation.Further mating among these F1 mice resulted in homozygous F2 offspring.In the F2 generation, the genotypic distribution of WT mice, heterozygotes, and homozygotes deviated from the Mendelian inheritance ratios with percentages of 4.41%, 39.32% and 56.27%, respectively.This value is higher than the mean MAI of fertile men (1.58 ± 0.2). 3,21However, both the fifth and sixth editions of the WHO laboratory manual do not specify a range of reference values for MAI, and the sixth edition of the WHO laboratory manual states that no sharp limit exists between fully fertile and subfertile men.Based on this, we calculated the patient's multiple anomalies index (MAI) as 1.92.This value was higher than the mean MAI of fertile men (1.58 ± 0.2). 3,21However, both the fifth and sixth editions of the WHO laboratory manual do not specify a range of reference values for MAI, and the sixth edition of the WHO laboratory manual states that no sharp limit exists between fully fertile and subfertile men. 3,21The sperm HOS test revealed a vitality rate of 45%, suggesting a decrease in sperm viability in the patient.The patient's semen POX staining was 0.15 × 10 6 /mL and seminal plasma elastase was 42.9 ng/mL, which were both within normal ranges, indicating no reproductive tract infection in the patient.

| WES and Sanger sequencing
Whole-exome sequencing (WES) revealed a novel mutation site in exon 7 of DNAAF3, characterized by a homozygous insertion of seven bases (c.871_872ins CAGGCTC, p.Q291fs).This mutation was identified as a frameshift mutation.This variant was not found in the normal control population of the Exome Aggregation Consortium (ExAC) (http:// exac.broad insti tute.org/ ) database or the 1000 Genomes Project (http:// www.1000g enomes.org/ data) database.Sanger sequencing was performed to confirm this mutation in the patient; however, no familial verification was performed.

| Pathogenicity analysis of DNAAF3 mutation
DNAAF3 (c.871_872ins CAGGCTC, p.Q291fs) was identified as a loss-of-function (LOF) variant.This mutation resulted in the incorrect translation of 20 amino acids, followed by premature termination (Figure 1).According to American College of Medical Genetics and Genomics (ACMG) guidelines, 24 this variant has been classified as pathogenic.Computational analysis using the online variant pathogenicity prediction tool MutationTaster (http:// www.mutat ionta ster.org/ ) indicated that this mutation is 'disease causing', with a probability of 100%.
After embryo transfer, heterozygous F1 offspring were obtained and crossed to obtain homozygous F2 offspring.The genotype frequencies of wild-type mice, heterozygotes, and homozygotes in the F2 generation were 15.43%, 57.41% and 27.16%, respectively, which exhibited no significant difference with expected frequencies (25%, 50% and 25%) (p = 0.09).Figure 3 displays the phenotypic characteristics of the mutant mice.When compared to wild-type mice, homozygous mice exhibited enlarged heads and weak bodies (Figure 3C,D).The body weight of homozygous mice was significantly lower than that of wild-type mice (6.49 ± 0.93 vs. 15.50 ± 0.37 g, p = 0.0015) (Figure 3E) (The original data can be found in Table S1).Micro-computed tomography revealed an expanded cranial contour (Figure 3F).The anatomical results revealed that the brains of homozygous mice were larger Compared to wild-type mice, homozygous mice exhibit growth defects, with enlarged heads and smaller bodies.The body weight of homozygous mice was significantly lower than that of wild-type mice (6.49 ± 0.93 vs. 15.50 ± 0.37 g, p = 0.0015) (The original data can be found in Table S1).(F) Micro computed topography (MicroCT) images of the skulls of homozygous and wild-type mice.The skull outline is enlarged in homozygous mice.(G-H) Results of anatomical examination of the skulls of homozygous and wild-type mice.Compared to wild-type mice, homozygous mice exhibit increased brain volume, enlarged ventricles, and thinning of the cerebral cortex.(I, J) Results of anatomical examination of the testes of homozygous and wild-type mice.The testes size is significantly reduced in homozygous mice compared to that in wild-type mice (8.90 ± 0.74 vs. 27.28 ± 4.03 mm 3 , p < 0.0001) (The original data can be found in Table S2).(K-N) Normal morphology of brain tissue in homozygous and wild-type mice stained with haematoxylin & eosin (scale: 500 and 50 μm).(O-R) Haematoxylin and eosin (H&E) staining of lung tissue sections (scale: 100 and 50 μm).Compared to wild-type mice, homozygous mice exhibit increased proliferation and a dense arrangement of alveolar epithelial cells, congested and dilated blood vessels (black arrow in R), and thickening of alveolar septa (red arrow in R). (S-V) H&E staining of testicular tissue sections (scale: 100 and 50 μm).Both homozygous and wild-type mice exhibit prepubertal characteristics without mature sperm.In homozygous mice, the diameter of the seminiferous tubules was reduced, the walls were thin, and the lumens were narrow or occluded with a decrease in the hierarchy and quantity of spermatogenic cells.(W-Z) Periodic acid Schiff (PAS) staining of testicular tissue sections (scale: 100 and 50 μm).The germ cell types in wild-type mice were more diverse than those in homozygous mice.Elongated spermatids were observed in the lumen line (red arrow in X), indicating the movement of the nucleus from a central position in the cell so that the acrosome approaches the surface of the cytoplasm and points toward the basal membrane of the tubule, signifying Stage 8 spermatids.Whereas, in homozygous mice, most of the germ cells were pachytene spermatocytes (red arrow in Z) and acrosomic granules were not observed, which indicated that the stage of germ cells was before the round spermatid stage.
(Figure 3G,H), ventricles were more dilated (Figure 3G), cerebral cortex was thinner (Figure 3G), and testicular volume was significantly lower than that of wild-type mice (8.90 ± 0.74 vs. 27.28 ± 4.03 mm 3 , p < 0.0001) (Figure 3I,J) (The original data can be found in Table S2).H&E staining of brain tissue from homozygous mice revealed normal morphology (Figure 3K-N).The H&E -stained lung tissue sections of homozygous mice exhibited increased proliferation and dense arrangement of alveolar epithelial cells, congested and dilated blood vessels, and thickened alveolar septa compared to those in wild-type mice (Figure 3O-R).
The testicular tissue sections stained with H&E (Figure 3S-V By crossing the C57 heterozygous and wild-type 129 strains, heterozygous mice with both C57 and 129 genetic backgrounds were obtained in the F1 generation and were subsequently mated to obtain homozygous F2 offspring.The genotype frequencies of wild-type mice, heterozygotes, and homozygotes in the F2 offspring of the C57/C57 cross were 4.41%, 39.32% and 56.27%, respectively, which were significantly different from the expected frequencies (p < 0.001).Homozygous mice with both C57 and 129 genetic backgrounds had a high mortality rate, and all died before PND21.

| DISCUSS ION
The axonemal dynein complex, comprising the ODA and IDA, is responsible for generating and regulating the beating of cilia and flagella. 25The axonemal dynein arm is first assembled in the cytoplasm and then delivered to the axoneme during ciliogenesis.7][28] Three DNAAF family members (DNAAF2, DNAAF4 and DNAAF6) play significant roles in male fertility. 17DNAAF3 has been identified as one of the PCD-related pathogenic genes, and based on the Global Variome shared LOVD database, as of 7th July 2023, a total of 212 common variants have been reported (https:// datab ases.lovd.nl/ shared/ genes/ DNAAF3).
In this study, we identified a novel mutation in the DNAAF3 gene through WES of samples from an infertile Chinese male; bioinformatics analysis showed that this variant was highly pathogenic, and may have led to PCD-related phenotypes, such as bronchitis and severe asthenozoospermia, in this male.This study also presented the semen examination results of the patient and established a homozygous mutant mouse model.The results of this study expand the spectrum of DNAAF3 mutations and provide important clues for investigating the mechanisms underlying DNAAF3-related male infertility associated with PCD.
The first reported case of DNAAF3 mutation dates back to 2012. 19tchison et al. reported the clinical features of 10 patients carrying DNAAF3 mutations, with each patient displaying typical symptoms of PCD, including chronic cough, recurrent chest infections, sinusitis, severe lung disease, bronchiectasis, and hearing loss. 19Four of the 10 patients also presented with situs inversus.Additionally, the study reported the first case of infertility associated with a DNAAF3 mutation, in which the patient was diagnosed with immotile sperm; however, more detailed semen analysis results were not provided.More recently, Wan et al. reported a second case of DNAAF3 mutationrelated male infertility, with the patient presenting with sinusitis and recurrent lower respiratory tract infections, but no situs inversus. 20e semen examination indicated severe asthenozoospermia (1.7% progressive motility and 0.49% non-progressive motility) but normal morphology.Transmission electron microscopy and immunostaining analysis of the patient's sperm revealed the complete absence of the ODA and partial absence of the IDA.The patient produced offspring through intracytoplasmic sperm injection (ICSI).
The newly identified case of homozygous DNAAF3 mutationrelated infertility in our study also presented with a history of bronchitis and infertility but without situs inversus.Semen examination of the patient indicated a normal sperm count, but approximately 99.2% of the sperm were immotile and approximately 99.1% exhibited morphological abnormalities, suggesting severe asthenozoospermia and teratospermia.This finding is similar to the semen examination results reported for mutations in the DNAAF1/2/3/4/6/7 family proteins, where the sperm count is usually within the normal range, but with impaired motility and varying degrees of morphological abnormalities. 20,25,29,30Although Wan et al. reported normal sperm morphology in their DNAAF3 mutation case, 20 we observed that reduced flagellar length was an important characteristic of sperm morphology in the patient with DNAAF3 mutation in our study.Previously reported human patients with DNAAF2, 4, 6 and 7 mutations also exhibited significantly decreased flagellar length, and transmission electron microscopic examination of the sperm flagella revealed defects in or the absence of the ODA and IDA. 25,29,30Chlamydomonas, an algal genus widely used to study flagellar motility, with DNAAF3 mutations also displayed shortened flagella. 192][33] Although our study did not analyse the ultrastructure of the patient's sperm, Wan et al.'s research validated the presence of ODA and IDA abnormalities in the sperm flagella of DNAAF3 mutation patients using electron microscopy. 20These findings suggest that the presence of the ODA and IDA is crucial for the stability of sperm flagella.
Animal models can provide direct evidence supporting an as- Moreover, the study found that male Drosophila flies with these mutations were unable to reproduce, as their chordotonal neuron cilia and sperm flagella lacked dynamic protein arms and exhibited axoneme damage. 18These studies have provided strong evidence confirming the correlation between DNAAF3 gene mutations and PCD.
However, no researchers have successfully developed a DNAAF3 mutant mammalian model, such as a mouse model.
In the present study, we used CRISPR-Cas9 technology to generate C57BL/6 mice carrying the same homozygous DNAAF3 mutation.The proportion of homozygotes in the F2 offspring was significantly lower than the Mendelian inheritance ratio, which may be related to the intrauterine loss of homozygotes during pregnancy, which is similar to the results obtained after constructing DNAAF5 mutant mice. 34This indicates that DNAAF3 is essential for embryonic growth and development and that its functional impairment is embryonically lethal.Furthermore, the homozygous mutant mice exhibited severe hydrocephalus and early lethality.5][36] Motile cilia in the ependymal layer play a crucial role in facilitating the flow of cerebrospinal fluid, and their dysfunction can lead to the accumulation of cerebrospinal fluid and congenital hydrocephalus. 37,38In humans, PCD-associated hydrocephalus is less common than the typical symptoms, such as recurrent respiratory infections, male infertility, and situs inversus.The DNAAF3 mutant male patient included in this study only exhibited bronchitis and infertility phenotypes without evidence of hydrocephalus.However, hydrocephalus is frequently observed in PCD mouse models, possibly due to the significant anatomical and physiological differences in brain development between humans and mice. 39fferent strains of mice exhibit varying susceptibilities to and severities of PCD-associated hydrocephalus.C57BL/6 and FVB/N mice are highly sensitive to PCD-related hydrocephalus and are prone to early death, while 129S6/SvEv and C3H/He mice appear to have lower sensitivity and may reach normal lifespans. 39Therefore, in this study, heterozygous C57BL/6 mice were crossed with 129 background mice to obtain DNAAF3 mutant mice exhibiting both genetic backgrounds.However, none of these mice survived until sexual maturity in the present study.The preliminary results of the mouse model constructed in this study provide evidence for a correlation between DNAAF3 and the phenotype of PCD-associated hydrocephalus.
Noteworthily, PAS staining of the testicular tissue revealed stage 8 germ cells in wild-type mice, whereas in homozygous mice, the seminiferous tubules exhibited a reduced number and stratification of spermatocytes and their development was arrested before the round spermatid stage.This finding suggests that the gene mutation may cause a blockade in germ cell development in mice; however, further evidence is needed to confirm this.We cannot exclude the possibility that the spermatogenic blockade observed in homozygous mice might be a secondary effect resulting from multi-system pathologies such as hydrocephalus and pulmonary oedema or a primary alteration induced by the gene mutation itself.Moreover, we observed that the lumens of seminiferous tubules in homozygous mice were narrowed or obstructed.However, owing to a lack of well-preserved efferent duct tissue samples, we were unable to perform haematoxylin and eosin staining combined with immunofluorescence to investigate potential efferent duct multicilial defects, as described by Hoque et al. 40 Future research is warranted to further explore whether the structural abnormalities are caused by increased pressure or other factors associated with efferent duct multicilial defects.
This study has several limitations.First, owing to the lack of patient cooperation, we were unable to obtain blood samples from the parents for pedigree verification.Thus, we could not determine whether the mutation was inherited from the parents.Additionally, owing to the patient's noncompliance, we could not obtain an adequate number of semen samples; therefore, we did not perform scanning and transmission electron microscopy on the patient's sperm.The effect of this mutation on the ultrastructure of human sperm, thus, requires further investigation.Second, we were unable to conduct functional gene validation studies using testicular tissue and sperm samples from the patient.These functional validation experiments could help determine the localisation, quantification, and extent of the impact of the mutation on DNAAF3 and expression of other related proteins, providing more direct evidence to support the pathogenicity of the mutation and its association with infertility.Third, the infertile male subject in this study did not undergo assisted reproductive treatments; therefore, we could not observe whether ICSI treatment could help the patient produce healthy offspring.Fourth, regarding the construction of the DNAAF3 mutant mouse model, we did not conduct ultrastructural examinations of mouse neurons and respiratory cilia.In addition, we were unable to cocele, history of exposure to reproductive toxic drugs, including chemotherapy or radiation or history of exposure to chemical toxins or radiation that could cause reproductive damage.This study was conducted in accordance with the principles of the Declaration of Helsinki and was approved by the Independent Ethics Committee for Clinical Research and Animal Trials of the First Affiliated Hospital of Sun Yat-sen University (approval number: [2021]326; date of approval: May 12th, 2021).Written informed consent was obtained from the patient.

F I G U R E 1
An illustration of the mutation site in the DNAAF3 gene observed in the patient in this study (A), as well as a pedigree chart (B).(A) Whole exome sequencing (WES) revealed that the patient (IV1) had a homozygous insertion of seven nucleotides (c.871_872ins CAGGCTC, p.Q291fs) in exon 7 of the DNAAF3 gene on chromosome 19.This represents a novel homozygous mutation, which results in the premature termination of translation after the insertion site, leading to a missense of 20 amino acids.Sanger sequencing confirmed this homozygous mutation.(B) The parents of the patient (III1 and III2) are first cousins, indicating a possible heterozygous DNAAF3 mutation; however, Sanger sequencing was not performed to validate this assumption.
All procedures were approved by the Independent Ethics Committee for Clinical Research and Animal Trials of the First Affiliated Hospital of Sun Yat-sen University and performed in accordance with the People's Republic of China regulations concerning the use of animals in research.
)showed prepubertal characteristics in both wild-type and homozygous mice, with no mature sperm.To determine the exact stage of germ cells with the development of the acrosome cap or the morphology of the younger generation of spermatids, the testicular tissues were subjected to PAS staining.As shown in Figure3W-Z, the germ cell types in wild-type mice were more diverse than those in homozygous mice.Elongated spermatids were observed in the lumen line (red arrow in Figure3X), indicating the movement of the nucleus from a central position in the cell so that the acrosome approaches the surface of the cytoplasm and points toward the basal membrane of the tubule, signifying stage 8 spermatids.Whereas in homozygous mice, most of the germ cells were pachytene spermatocytes (red arrow in Figure3Z), and acrosomic granules were not observed, which indicated that the stage of germ cells was before the round spermatid stage.The diameter of the seminiferous tubules in homozygous mouse testes decreased, the walls became thinner, and the lumens narrowed or became blocked, with a decrease in the number and hierarchy of germ cells (Figure3V,Z).All of the homozygous mice died before postnatal day 21 (PND21).