2.1. Patient

The patient was a 32-year-old man who presented to the outpatient clinic of our hospital for semen examination due to fertility issues. Physical examination of the testis, epididymis, and vas deferens revealed normal findings; moreover, his blood hormone levels were within the normal range: follicle-stimulating hormone, 4.39 IU/L; luteinizing hormone, 1.5 IU/L; testosterone, 1.57 ng/mL; estradiol, 13.78 pg/mL; and prolactin, 17.7 ng/mL. Neither the patient nor his family had any PCD-related symptoms. The patient was diagnosed with asthenozoospermia and teratozoospermia based on three semen analyses. The control sample was obtained from a fertile man who was planning to have a second child. This study was approved by the Ethics Committee of Zhejiang Provincial People’s Hospital, China (No. 2022091), and written informed consent was obtained from the patient (see Supplemental Material (available here)).

2.2. Semen Parameters and Sperm Morphological Analyses

The patient and healthy control were asked to abstain from sex for 2–7 days before collecting their masturbation semen samples in the hospital semen retrieval room. The semen samples were completely liquefied after being placed in a 37°C incubator for 30 min. Semen parameters, including semen volume, sperm progressive motility, and sperm concentration, were evaluated using computer-assisted sperm analysis (SAS medical, SAS-II, Beijing, China). Simultaneously, sperm morphology was assessed using Papanicolaou stain, and the samples were observed under bright field at 1,000x magnification. All semen parameters were evaluated according to the 5th edition of the WHO laboratory manual for the examination and processing of human semen [17].

2.3. Scanning Electron Microscopy (SEM)

Prepared spermatozoa samples were immobilized overnight with 2.5% phosphate-buffered glutaraldehyde at 4°C. Immobilized spermatozoa samples were dehydrated using a series of 50%, 70%, and 90% graded ethanol, wherein the concentration was changed after every 15 min. Finally, the samples were incubated in 100% ethanol twice for 20 min each. After drying, the sperm samples were observed via SEM at 10 kV (Nova nano450, Thermo Fisher FEI).

2.4. Transmission Electron Microscopy (TEM)

The prepared sperm samples were fixed overnight with 2.5% phosphate-buffered glutaraldehyde. These samples were washed thrice with 0.1 M phosphate buffer (PB; pH 7.2) and fixed with 1% osmium tetroxide in 0.1 M PB for 1 h at 4°C. Dehydration was performed sequentially using 50%, 70%, and 90% ethanol as well as 100% acetone. The samples were then allowed to polymerize for 48 h at 60°C. Subsequently, ultrathin sections were stained with 3% uranyl acetate and lead citrate. The ultrastructure of these samples was observed and photographed via TEM at 100 kV (Tecnai, Thermo Fisher FEI).

2.5. Immunostaining

The sperm samples were fixed on slides with 4% paraformaldehyde for 30 min. These samples were then permeabilized with 0.2% Triton-X for 15 min and blocked with 1% BSA for 60 min, followed by overnight incubation at 4°C with FSIP2 (rabbit monoclonal, Invitrogen, 1 : 100) and acetylated α-tubulin (mouse monoclonal, Santa Cruz, 1 : 200) antibodies. Next, the primary antibody was washed with PBS, and anti-rabbit (Abcam6802, 1 : 500, green) and anti-mouse (Abcam6789, 1 : 500, red) secondary antibodies were added to the slides and incubated with sperm for 1 h. Finally, the sperm samples were stained with DAPI for 3–5 min and observed under a scanning confocal immunofluorescence microscope (Leica sp8, Germany). Among the FSIP2 key domains, the protein domains were predicted using CDD (https://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi). Moreover, we visualized the 3D structural model of FSIP2 key domains using I-TASSER (http://zhang.bioinformatics.ku.edu/I-TASSER/) (see Supplemental Material) and PyMOL2.5.2 (https://pymol.org) to identify the impact of FSIP2 mutation sites on protein structure and function.

2.6. WES and Sanger Sequencing

Peripheral blood samples (2 mL) were collected from the patient and his parents, and DNA was extracted from whole blood using a blood genomic DNA extraction kit (QIAamp DNA Blood Mini Kit, Qiagen). WES was performed using Novaseq 6000 system based on Illumina sequencing technology. Patient gene sequences were aligned with normal human genome sequences, and loci with pathogenic potential were screened via Mutation Taster, SIFT, Polyphen2, and GERP++ software analysis and then matched according to genotype-phenotype. The frequency of genomic variants in the general population was assessed using 1,000 Genomes Project, ESP6500 and ExAC, and gnomADMAX (see Supplemental Material). FSIP2 mutations in the proband and patient’s parents were verified using PCR and Sanger methods. The PCR cycle consists of three steps: (1) high-temperature denaturation, in which the DNA is converted to single strands at 95°; (2) low-temperature annealing (approximately 55°C), wherein the primer and single chain bind according to base complementary pairing; and (3) mid-temperature extension, in which the temperature is adjusted to the optimal reaction temperature of the DNA polymerase (approximately 72°C) to facilitate the synthesis of the complementary strand in the direction of the phosphate to five-carbon sugar (5′–3′). The following two primers were used to amplify exon 3 and exon 17 of FSIP2 (5′→3′). Exon 3 (c.272_275delinsAGGTTTTTATA), forward: GGTGAGGAGAGAGGAAAGGA and reverse: ACCCACTTTATTATTGCTGGTGA. Exon 17 (16788_16791del), forward: TGAAAAGCGGCATGATTAACCT and reverse: TGGACTTCAAGGAGGATAACTGA. PCR products were sequenced using Ex Taq DNA Polymerase Kit (Bio-Rad, Hercules, CA, USA). Finally, sequence analysis was performed using 3730XL system (GSL Biotech).

3.1. Routine Semen Analysis

Routine semen analysis revealed a normal sperm concentration and reduced sperm motility according to WHO guidelines (Table 1). Further, light microscopy and SEM analyses confirmed that approximately 90% of the patient’s sperm had an abnormal morphology, including several short tails and a few curled flagella (Figures 1(a) and 1(b)). In both healthy control and patient, TEM analysis revealed that the cross-sectional ultrastructure of the sperm mid-piece exhibited an intact “9+2” axonal architecture. However, a cross-section of the principal piece and end piece of the patient’s sperm revealed the loss of axonal structure and central pair (CP), with a disordered arrangement of the microtubule doublets (Figure 1(c)).

3.2. Gene Mutation Sites

Two novel FSIP2 mutations, exon3: c.272_275delinsAGGTTTTTATA (p.L92Vfster74) and exon 17: c.16788_16791del (p.E5596fs), were identified in the patient using WES (Figure 2). Detailed information on the variants is provided in Table 2. Sanger sequencing revealed that the patient’s father had c.16788_16791del heterozygous mutation and his mother had c.272_275delinsAGGTTTTTATA heterozygous mutation (Figure 2(a)). Functional predictions based on SIFT and Polyphen2 databases suggested that these mutations have never been reported, whereas Mutation Taster classified them as disease-causing (Table 2). Both mutations (p.L92Vfster74 and p.E5596fs) are new mutations whose gene frequencies have not been described in ExACMAX, 1000G_ALL, or GnomAD (Table 2). The two mutation sites are pathogenic according to American College of Medical Genetics and Genomics (ACMG) guidelines (Table 2).

3.3. Analysis of the Sequence and Protein Structure of the FSIP2 Mutation Sites

FSIP2 expression in the patient and control samples was determined via immunofluorescence staining. The patient’s sperm showed weak FSIP2 expression, whereas the control sperm showed positive FSIP2 expression (Figure 3).

To identify the mechanism underlying the loss of FSIP2 expression, we analyzed the sequence and protein structural features of the FSIP2 mutation site. We also compared the sequences of wild-type FSIP2 and the two patient mutation sites, c.272_275delinsAGGTTTTTATA and c.16788_16791del (Figure 4(a)). We found that c.272_275delinsAGGTTTTTATA mutation site ends early, leaving only 165 amino acids at the N-terminus, resulting in protein truncation. Notably, this serious deletion results in a significant loss of protein function. Furthermore, c.16788_16791del mutation site shared its first 5596 amino acids with wild-type FSIP2. However, subsequently, only five amino acids remained at the mutant C-terminus, whereas the wild-type C-terminus had a sequence of approximately 1400 amino acids.

Notably, the loss of informational sequence in c.272_275delinsAGGTTTTTATA resulted in the loss of protein structure and function. In the case of c.16788_16791del mutation, the loss of protein function can lead to the deletion of a key functional domain. To the best of our knowledge, the crystallographic structure of FSIP2 has not yet been reported in the literature. However, using a protein domain prediction tool, we could identify a potential functional domain in FSIP2 (5901-6774). In addition, we compared the sequences of FSIP1 and FSIP2 and found a sequence similarity of 12.5, indicating some structural similarity (see Supplemental Material). A structural model of FSIP2 and FSIP1 domains was constructed using homology modeling, which revealed the possible structure of the FSIP2 domain (5901-6774) (Figure 4(b)). The c.16788_16791del mutation led to the deletion of this critical domain, resulting in the loss of protein function and pathogenicity, as evidenced by the MMAF phenotype observed in this infertile patient.