Novel bi-allelic variants in DNAH10 lead to multiple morphological abnormalities of sperm flagella and male infertility

Abstract

Multiple morphological abnormalities of sperm flagella (MMAF) is a severe form of asthenoteratozoospermia, characterized by morphological abnormalities and reduced motility of sperm, causing male infertility. Although approximately 60% of MMAF cases can be explained genetically, the etiology of the remaining cases is unclear. Here, we identified two novel compound heterozygous variants in the gene, dynein axonemal heavy chain 10 (DNAH10), in three patients from two unrelated Pakistani families using whole-exome sequencing (WES), including one compound heterozygous mutation (DNAH10: c.9409C>A [p.P3137T]; c.12946G>C [p.D4316H]) in family 1 and another compound heterozygous mutation (DNAH10: c.8849G>A [p.G2950D]; c.11509C>T [p.R3687W]) in family 2. All the identified variants are absent or rare in public genome databases and are predicted to have deleterious effects according to multiple bioinformatic tools. Sanger sequencing revealed that these variants follow an autosomal recessive mode of inheritance. Hematoxylin and eosin (H&E) staining revealed MMAF, including sperm head abnormalities, in the patients. In addition, immunofluorescence staining revealed loss of DNAH10 protein signals along sperm flagella. These findings broaden the spectrum of DNAH10 variants and expand understanding of the genetic basis of male infertility associated with the MMAF phenotype.

INTRODUCTION

Sperm motility plays a crucial role in male fertility, as it enables sperm to move through the reproductive tract and fertilize the egg. Asthenoteratozoospermia, a complex disorder that causes reduced sperm motility and morphologically abnormal sperm, results in male infertility. Flagella and cilia are narrow thread-like structures with a common structural component known as the axoneme. The axoneme comprises nine peripheral A and B microtubule doublets arranged circumferentially around a central microtubule doublet, resulting in an evolutionarily conserved 9 + 2 pattern along the axoneme. In peripheral microtubule doublets, microtubule A contains inner and outer dynein arms (IDAs and ODAs, respectively), establishing mechanical links in the axoneme. ODAs and IDAs are intricate proteins, with multiple components, such as light, heavy, and intermediate polypeptide chains. Dynein arms, in their entirety, hydrolyze adenosine triphosphate (ATP) and assist in ciliary and flagellar movement.1

Multiple morphological abnormalities of sperm flagella (MMAF) is an acute manifestation of asthenoteratozoospermia, characterized by morphologically abnormal sperm, such as bent, coiled, irregular caliber, absent, or short flagella.2,3 However, the etiology of male infertility is multifactorial, including urinogenital infections, immuno-hormonal abnormalities, and genetic disorders.4,5 In mammals, pathogenic variants of the axonemal dynein heavy chain (DHC) gene family, such as dynein axonemal heavy chains DNAH1,2 DNAH2,6 DNAH3,7 DNAH6,8 DNAH9,9 DNAH8,10 DNAH7,11 and DNAH17,12 have been reported to cause male infertility. The DNAH10 (Online Mendelian Inheritance in Man [OMIM] accession No. 605884) gene, which encodes an IDA heavy chain component of sperm flagella and respiratory cilia, is required for the assembly of both sperm flagella and respiratory cilia.13,14 Studies have shown that mutations of DNAH10 cause MMAF in humans and mice, resulting in male infertility.1,15,16 A recent study identified a homozygous missense DNAH10 mutation in a patient with primary ciliary dyskinesia (PCD) and suggested that DNAH10 mutations may be a causative factor in this condition.17

The role of genetics in MMAF has been widely studied,18 and mutations in 36 genes have been linked to MMAF in humans.19 Even so, the genetic cause remains unidentified in 40% of patients with MMAF. Therefore, further research is needed to explore the genetic etiology of MMAF and enhance understanding of male infertility associated with MMAF. In the present study, utilizing whole-exome sequencing (WES), we identified novel compound heterozygous DNAH10 mutations in three patients with male infertility and MMAF from two unrelated Pakistani families. Our study expands understanding of MMAF caused by novel variants in DNAH10.

PATIENTS AND METHODS

Sample collection

In this study, we recruited three patients, P1 (IV:4), P2 (IV:1), and P3 (IV:3), from two unrelated Pakistani families (family 1 and family 2 as presented in Figure 1) with primary male infertility. Each patient completed a comprehensive questionnaire covering his physical condition and infertility history. None of the individuals in either family had a history of smoking, drinking, toxic chemical exposure, PCD symptoms, or other diseases. All three affected individuals had normal karyotypes and were physically normal in terms of height, weight, and external genitalia. In all three cases, erectile and ejaculatory functions were normal, and testicular diameters were normal. However, they were infertile.

Figure 1.

Identification of novel DNAH10 variants in two unrelated Pakistani families. The pedigree of (a) family 1, with one patient, P1 (IV:4), and (b) family 2, with two patients, P2 (IV:1) and P3 (IV:3). The red arrows indicate the individuals selected for whole-exome sequencing (WES). The two double lines indicate a consanguineous marriage. Verification of novel compound heterozygous DNAH10 missense variants by Sanger sequencing of genomic DNA from all available family members of both (c) family 1 and (d) family 2, which revealed a recessive mode of inheritance. The arrowheads represent variant sites. DNAH10: dynein axonemal heavy chain 10; P: patient; WT seq: wild-type DNA sequence; MT seq: mutant DNA sequence; Het: heterozygous; WT: wild-type; F: female; M: male.

P1 (IV:4), P2 (IV:1), and P3 (IV:3) sperm smear slides were prepared for analysis of sperm morphology and immunostaining. Blood samples were also obtained from the patients and available family members.

All the participants provided informed consent. This study was conducted in accordance with the guidelines of the University of Science and Technology of China (USTC; Hefei, China) and was approved by the institutional ethics committee of the USTC (Approval No. 2019-KY-168).

WES, gene filtration, and Sanger sequencing

WES was conducted as described in a previous study.20 Briefly, a QIAamp DNA Blood Mini Kit (51206; QIAGEN, Hilden, Germany) was used to extract total genomic DNA from peripheral blood obtained from the three patients in family 1 and 2 and available family members, following the guidelines of the manufacturer. WES was performed for P1 (IV:4) from family 1. In family 2, WES was performed for P2 (IV:1), P3 (IV:3), their fertile brother (IV:6), and their mother (III:2). DNA integrity was determined by 1% agarose gel electrophoresis and then fragmented by Covaris-focused ultrasonication. An AIExome Enrichment Kit V1 (iGeneTech, Beijing, China) was used to capture known exons and exon‒intron boundary sequences. DNA sequencing libraries were constructed according to the manufacturer’s instructions. Sequencing was performed using the HiSeq2000 platform (Illumina, San Diego, CA, USA). The Burrows‒Wheeler Aligner, using the default parameters, was employed to align sequencing reads to Genome Reference Consortium Human Build 37 (GRCh37). The Sequence Alignment/Map (SAM) files from each sample were converted to Binary Alignment/Map (BAM) files, sorted, and merged using SAMtools. PCR duplicates were then removed using Picard software (http://broadinstitute.github.io/picard/, last accessed on 2024 June 07). The files were processed using the Genome Analysis Toolkit HaplotypeCaller (http://www.broadinstitute.org/gatk/, last accessed on 2024 June 07). A local indel realigner was used to realign the BAM data. SNVs and indels were found in the gathered coding exonic intervals using GATK’s UnifiedGenotyper. Filtration of detected variants was performed as described in the flow chart in Supplementary Figure 1 ^{(166.3KB, tif)} . Subsequently, Sanger sequencing of available family members and patients from family 1 (III:1, III:2, IV:2, and P1 [IV:4]) and family 2 (III:2, IV:6, P2 [IV:1], and P3 [IV:3]) was used to confirm inheritance patterns. Supplementary Table 1 lists the primers used for Sanger sequencing.

Supplementary Table 1.

Polymerase chain reaction and sanger sequencing primers for novel dynein axonemal heavy chain 10 variants

Gene	cDNA change	AA change	Product size (bp)	Forward primer	Reverse primer
DNAH10	c.G8849A	p.G2950D	383	GAGTCCCTTCCAGACTTTTC	AACTCCCTTAAGAGCTGAGC
DNAH10	c.C9049A	p.P3137T	563	CATGGTGGCCCATCCTGTAG	ATAAGGACATGCCAGGGTCC
DNAH10	c.C11059T	p.R3687W	459	CCTTGCTGCTCTAGAGTGAC	CAGAGAAATGGACGCTCCAA
DNAH10	c.G12946C	p.D4316H	362	ACTCTACTCAGGAGGACTTC	TTCAGCCAGAGACTTCGTCA

cDNA: complementary DNA; DNAH10: dynein axonemal heavy chain 10

Semen and sperm morphology analysis

Semen analysis was performed following a minimum of 5 days of sexual abstinence. Sperm concentration and motility were assessed according to the 6^th edition of the laboratory manual of the World Health Organization (WHO).21 The sperm morphologies of the affected individuals and that of a normal control were analyzed using hematoxylin and eosin (H&E) staining of sperm smear slides, as described elsewhere.1,15 Two separate morphological analyses of around 200 spermatozoa from the normal control and patients were used to show the percentage of sperm with different flagellar defects. For morphologically abnormal spermatozoa, we used the reference ranges for fertile men: short (<1%), bent (<13%), coiled (<17%), irregular caliber (<23%), and absent (<5%), as indicated in previous studies.1,22 We used the Student’s t-test to evaluate the significance of differences in sperm flagellar morphologies between the fertile control and the patients. The data are presented as the mean ± standard deviation (s.d.). Differences were considered significant at ^*P < 0.05, ^**P < 0.01, and ^***P < 0.001.

Conservation and in silico analysis of novel missense DNAH10 variants

DNAH10 protein amino acid sequences of different organisms were obtained from the GenBank database (https://www.ncbi.nlm.nih.gov/datasets/gene/, last accessed on 2024 September 17). The evolutionary conservation of affected residues across species was analyzed using Clustal Omega.23 A schematic diagram of the domain composition of the DNAH10 protein was drawn using the Domain Graph, version 1.0 tool integrated in GPS 5.0 software.24 The global frequencies of novel missense DNAH10 variants were checked via the 1000 Genomes Project and Genome Aggregation Database (gnomAD). To analyze the potentially damaging effects of novel missense variants on the DNAH10 protein, we used PolyPhen-2,25 SIFT,26 and MutationTaster.27 We evaluated the potential effect of novel missense variants on the structural stability of the DNAH10 protein using MUpro (https://mupro.proteomics.ics.uci.edu/, last accessed on 2024 October 20) and I-Mutant2.0 bioinformatic tools (https://folding.biofold.org/cgi-bin/i-mutant2.0.cgi, last accessed on 2024 October 20).28,29

Immunostaining

Immunofluorescence staining of DNAH10 was conducted using sperm smear slides from P1 (IV:4), P2 (IV:1), and P3 (IV:3), as previously described.12 Briefly, the sperm smear slides were washed with phosphate-buffered saline (PBS) twice and then fixed with 4% paraformaldehyde solution. Subsequently, the slides were permeabilized using 0.1% Triton X-100 solution in PBS, followed by blocking with 3% nonfat milk solution. The primary antibodies used were rabbit polyclonal anti-DNAH10 (1:100; bs-11022R; Bioss, Beijing, China) and anti-α-tubulin (1:200; F2168, Sigma-Aldrich, St. Louis, MO, USA), which were incubated at 37°C overnight. The next day, the slides were incubated with Alexa Fluor 488-conjugated goat anti-mouse IgG (1:100; A21121, Invitrogen, Carlsbad, CA, USA) and Alexa Fluor 555-conjugated donkey anti-rabbit IgG (1:200; A31572; Invitrogen) secondary antibodies at 37°C for 1 h. Next, the slides were mounted with VECTASHIELD mounting medium (H-1000; Vector Laboratories, Burlingame, CA, USA), supplemented with Hoechst 33342 (H21492; Invitrogen). A Nikon ECLIPSE 80i microscope (Nikon, Tokyo, Japan), equipped with a Hamamatsu charge-coupled device, was used to obtain images of spermatozoa.

Data availability

The authors confirm that the data presented in the article and its supplementary Information support the conclusions drawn in this study. Additional data can be provided by the corresponding author upon request.

RESULTS

Clinical characteristics of the affected individuals

Two unrelated Pakistani families were enrolled in this study. Family 1 had two siblings, IV:1 (43 years old with five children) and P1 (IV:4, 37 years old). The wife (IV:3) of P1 (IV:4) had normal height and weight, regular menstrual cycles, and no other reproductive-related medical history, but she had not conceived after 17 years of marriage. Family 2 had one fertile brother, IV:6 (44 years old with three children), and two patients, P2 (IV:1) and P3 (IV:3). In family 2, the patients’ wives also had normal height and weight, regular menstrual cycles, and no other reproductive-related diseases, but they had not become pregnant after 20 years and 23 years of marriage, respectively (Figure 1a and 1b). For P1 (IV:4), P2 (IV:1), and P3 (IV:3), semen analysis was conducted twice, according to WHO guidelines.21 All three patients had normal semen volumes (>1.4 ml) and sperm concentrations (>16 × 10⁶ ml⁻¹). However, more than 80% of their sperm were immotile, with the percentage of sperm with progressive motility and total motility from P1 (mean ± s.d.: 14.2% ± 1.1% and 17.1% ± 2.0%, respectively), P2 (mean ± s.d.: 11.5% ± 2.1% and 14.0% ± 1.4%, respectively), and P3 (mean ± s.d.: 15.0% ± 1.4% and 20.0% ± 1.4%, respectively) far below the lower limit of the normal reference value (Table 1), indicating that the affected individuals have asthenozoospermia.

Table 1.

Clinical characteristics of patients harboring dynein axonemal heavy chain 10 compound mutations

Characteristic	Reference value	P1 (IV:4; family 1)	P2 (IV:1; family 2)	P3 (IV:3; family 2)
Fertility status	-	Infertile	Infertile	Infertile
Age at diagnosis (year)a	-	37	39	49
Years of marriageb	-	17	20	23
Height/weight (cm/kg)	-	155/70	172/78	175/82
Karyotype	-	46,XY	46,XY	46,XY
Semen parameters, mean±s.d.
Semen volume (ml)	≥1.4	2.5±0.4	1.9±0.1	2.1±0.1
Sperm concentration (×106 ml−1)	≥16	20.0±4.2	21.0±1.4	22.0±1.4
Progressive motility (%)	≥30	14.2±1.1	11.5±2.1	15.0±1.4
Total motility (%)	≥42	17.1±2.0	14.0±1.4	20.0±1.4
Immotile sperm (%)	-	82.9±2.0	86.0±1.4	80.0±1.4

^aAge at the manuscript submission. ^bYears of marriage at the manuscript submission. Reference values are according to 6^th edition of WHO manual in 2021.21 -: no reference value; s.d.: standard deviation; WHO: World Health Organization

Identification of novel compound heterozygous missense DNAH10 variants in the patients

To identify the genetic cause of the patients, we performed WES on available family members and patients from families 1 and 2. Through filtering of WES data, we identified a novel compound heterozygous DNAH10 missense variant (NM_001372106.1: c.9409C>A [p.P3137T]; c.12946G>C [p.D4316H]) in P1 (IV:4) from family 1 (Supplementary Figure 1a ^{(166.3KB, tif)} ) and another novel compound heterozygous DNAH10 missense variant (NM_001372106.1: c.8849G>A [p.G2950D]; c.11059C>T [p.R3687W]) in P2 (IV:1) and P3 (IV:3) from family 2 (Supplementary Figure 1b ^{(166.3KB, tif)} ). Further validation and segregation analysis of the variants using Sanger sequencing of P1 (IV:4) and his available family members indicated that the novel compound heterozygous DNAH10 missense variant in P1 (IV:4) was recessively inherited from his parents, as we found that his biological father (III:1) was heterozygous for one variant (c.9409C>A). However, his biological mother (III:2) was heterozygous for another variant (c.12946G>C). Moreover, his sister (fertile control) was found to have a wild-type allele (Figure 1c). In family 2, Sanger sequencing of P2 (IV:1), P3 (IV:3), one control brother (IV:6), and their biological mother (III:2) was carried out. We found that P2 (IV:1), P3 (IV:3), and their fertile brother (IV:6) were heterozygous for one variant (c.8849G>A), but their biological mother had a wild-type allele. This finding suggests that their deceased biological father should be heterozygous for this variant. For another variant (c.11059C>T), we found that P2 (IV:1), P3 (IV:3), and their biological mother were heterozygous. The fertile brother had a wild-type allele, indicating that the novel compound heterozygous DNAH10 missense variants in P2 (IV:1) and P3 (IV:3) were recessively inherited from their parents (Figure 1d).

Next, we performed in silico analysis to evaluate the pathogenicity of these mutations and found that the novel compound heterozygous missense variants identified in DNAH10 were located on ATPases associated with diverse cellular activities (AAA-ATPases) and DHC C-terminal domains and resulted in the substitution of conserved amino acids (Figure 2a). Moreover, these novel variants were rare in the 1000 Genomes Project database and the gnomAD database and were predicted to be deleterious by Polyphen-2, SIFT, and MutationTaster tools (Table 2). In addition, the MUpro and I-Mutant2.0 bioinformatic tools showed that the structural stability of the DNAH10 protein was decreased due to the novel missense DNAH10 variants (Supplementary Table 2). These results suggest that the identified compound heterozygous missense DNAH10 variants should be the disease-causing mutations in patients.

Figure 2.

Graphical illustration and conservation analysis of identified novel DNAH10 variants. (a) Multiple sequence alignment of the DNAH10 protein showing evolutionary conservation across different species. The red arrows show the positions of conserved mutated amino acids in family 1 and family 2. (b) The mutation sites of four novel compound heterozygous DNAH10 missense variants in family 1 and family 2 and previously reported mutations. The schematic diagrams show the positions of the variants at the cDNA (transcript ID: NM_001372106.1) and protein (NP_001359035.1) levels. The variants marked in red indicate those we identified in this study. DNAH10: dynein axonemal heavy chain 10; cDNA: complementary DNA; UTR: untranslated region; DHC-N1: dynein heavy chain, N-terminal region 1; DHC-N2: DHC, N-terminal region 2; AAA: ATPases associated with a variety of cellular activities; MT: microtubule-binding stalk of dynein motor.

Table 2.

In silico analysis of the novel compound heterozygous missense variants in dynein axonemal heavy chain 10

cDNA change in patient	Amino acid change	PolyPhen-2	SIFT	MutationTaster	1000 Genomes project	GnomAD
P1 (IV:4) in family 1
c.9409C>A	p.P3137T	Damaging	Deleterious	Disease causing	0.000799	0.0002321
c.12946G>C	p.D4316H	Damaging	Deleterious	Disease causing	NA	0.000004013
P2 (IV:1) and P3 (IV:3) in family 2				Disease causing
c.8849G>A	p.G2950D	Damaging	Deleterious	Disease causing	0.002995	0.001206
c.11059C>T	p.R3687W	Damaging	Deleterious	Disease causing	NA	0.00001291

NA: not available; cDNA: complementary DNA

Supplementary Table 2.

Analysis of the effect of novel dynein axonemal heavy chain 10 protein variants on the protein structural stability

Gene	Patient	Protein alteration	MUpro		I-Mutant2.0
			DDG (kcal/mol)*	Protein stability	DDG# (kcal/mol)	RI	pH	Protein stability
DNAH10	P1 (IV:4) (family 1)	p.P3137T	−1.7860969	Decrease	−1.68	8	6.6	Decrease
		p.D4316H	−1.5218294		−1.20	5
	P2 (IV:1), P3 (IV:3) (family 2)	p.G2950D	−1.0815977	Decrease	−0.75	8	6.6	Decrease
		p.R3687W	−0.77081054		−0.43	6

*Free energy change value; ^#DG (new protein) − DG (wild type) in kcal/mol. DDG <0: decreased stability; DDG >0: increased stability. RI: Reliability index; DNAH10: dynein axonemal heavy chain 10

Overall, including the four novel variants we found in the present study and 12 previously reported variants,1,15,16,17 16 variants in DNAH10 in total have been identified so far, which are distributed in different domains of the DNAH10 protein (Figure 2b). Notably, spermatozoa motility in our patients was higher than that of patients carrying other DNAH10 mutations in previous studies (Supplementary Table 3), which may be due to differences in mutations.

Supplementary Table 3.

Phenotypes and dynein axonemal heavy chain 10 variants previously reported and in this study

Disease (phenotype)	Variants	Mutation type	Motility (%)	Progressive motility (%)	References
Asthenoteratozoospermia	NM_001372106.1: c.13192G>A (p.G4398R)	Homozygous missense	0	0	Tu et al. 2021
Asthenoteratozoospermia	NM_001372106.1: c. 7955C>T (p.T2652M)	Homozygous missense	0	0	Tu et al. 2021
Asthenoteratozoospermia	NM_001372106.1: c.6017G>A (p.R2006Q); c.12241 (p.R4081C)	Compound heterozygous missense	9.2	1.7	Tu et al. 2021
Asthenoteratozoospermia	NM_001372106.1: c.7614dupC (p.E2539Rfs*26); c.12589delA (p.S4197Afs*5)	Compound heterozygous frameshift	II-1: 10.3 II-2: 11.8	II-1: 1.9 II-2: 2.4	Tu et al. 2021
Asthenoteratozoospermia	NM_001372106.1: c.1483A>G (p.K495E); c.11915G>A (p.G3972E)	Compound heterozygous missense	-	-	Li et al. 2022
Asthenoteratozoospermia	NM_001372106.1: c.2868delG (p.L957*)	Homozygous frameshift	6.3	3.8	Li et al. 2022
Asthenoteratozoospermia	NM_001372106.1: c.11174T>C (p.M3725T); c.13046C>T (p.T4349I)	Compound heterozygous missense	13.3	2.4	Li et al. 2022
Asthenoteratozoospermia + PCD	NM_001372106.1: c.772C>T (p.R258W)	Homozygous missense	0	0	Wang et al. 2023
Asthenoteratozoospermia	NM_001372106.1: c.9409C>A (p.P3137T); c.12946G>C (p.D4316H)	Compound heterozygous missense	P1: 17.1	P1: 14.25	This study
Asthenoteratozoospermia	NM_001372106.1: c.8849G>A (p.G2950D); c.11059C>T (p.R3687W)	Compound heterozygous missense	P2: 14.0 P3: 20.0	P2: 11.5 P3: 15.0	This study

PCD: primary ciliary dyskinesia

Patients harboring DNAH10 mutations displayed typical MMAF phenotypes

Considering that reported mutations in DNAH10 have been reported to cause morphological abnormalities in human spermatozoa,1,15 we wondered whether the compound heterozygous mutations in DNAH10 we identified resulted in a similar phenotype. Thus, we performed H&E staining of spermatozoa from the patients and fertile control group to examine sperm morphology (Figure 3a). In the fertile control group, more than 82.3% of spermatozoa had a normal morphology of flagella. However, the proportion of spermatozoa with normal flagella obtained from patients was significantly reduced, with 4.5% in P1 (IV:4), 2.8% in P2 (IV:1), and 2.8% in P3 (IV:3). Moreover, the three patients showed significantly higher percentages of coiled (>30%), irregular caliber (P1: 18.8%, P2: 24.5%, and P3: 23.6%), and bent (P1: 31.0%, P2: 28.0%, and P3: 31.2%) sperm flagellar defects. P2 (IV:1) and P3 (IV:3) also exhibited a significant increase in the number of short sperm flagella, compared to the fertile control group (Figure 3b). We also analyzed the sperm head, midpiece/neck, and residual cytoplasm. In more than 50% of patients’ spermatozoa, the sperm heads were abnormal, mainly pyriform and tapered. However, the midpiece/neck and residual cytoplasm were comparable to the controls (Supplementary Table 4). Collectively, these results pointed to the presence of asthenoteratozoospermia in the patients with the MMAF phenotype.

Figure 3.

Sperm flagella from patients harboring DNAH10 mutations exhibit multiple morphological abnormalities. (a) Representative images of H&E-stained spermatozoa from the fertile control and the patients (P1 [IV:4] in family 1, and P2 [IV:1] and P3 [IV:3] in family 2). The green arrows represent sperm head defects, such as pyriform, tapered, amorphous, and round. Scale bars = 10 µm. (b) Statistical analyses of sperm flagellar morphologies from the fertile control and the patients. The differences were calculated using the Student’s t-test. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001. DNAH10: dynein axonemal heavy chain 10; H&E: hematoxylin and eosin; NS: non-significant.

Supplementary Table 4.

Semen analysis and morphological details of patients and control used in this study

Parameters	Fertile control		P1 (family 1)		P2 (family 2)		P3 (family 2)
Semen parametersa	A1	A2	A1	A2	A1	A2	A1	A2
Semen volume (ml)	-	-	2.5	2.0	1.8	2	2	2.2
Sperm concentration (×106 ml−1)	-	-	17	23	20	22	23	21
Progressive motility (%)	-	-	15	13.5	10	13	16	14
Nonprogressive motility (%)	-	-	3.5	2.2	3.0	2.0	3.0	7.0
Total motility (%)	-	-	18.5	15.7	13	15	19	21
Immotile sperm (%)	-	-	81.5	84.3	87	85	81	79
Flagellar defectsb	A1	A2	A1	A2	A1	A2	A1	A2
Normal (%)	80.0 (160)	84.5 (169)	4.0 (8)	5.0 (10)	3.5 (7)	2.0 (4)	3.0 (6)	2.5 (5)
Absent (%)	3.0 (6)	1.0 (2)	7.0 (14)	4.0 (8)	6.0 (12)	8.0 (16)	4.0 (8)	3.0 (6)
Short (%)	1.0 (2)	1.0 (2)	5.0 (10)	3.0 (6)	7.0 (14)	5.0 (10)	4.5 (9)	5.0 (10)
Bent (%)	8.0 (16)	6.0 (12)	27.0 (54)	35.0 (70)	26.0 (52)	30.0 (60)	28.5 (57)	34.0 (68)
Coiled (%)	7.0 (14)	6.0 (12)	35.5 (71)	37.0 (74)	33.0 (66)	30.5 (61)	34.5 (69)	33.5 (67)
Irregular (%)	1.0 (2)	1.5 (3)	21.5 (43)	16.0 (32)	24.5 (49)	24.5 (49)	25.5 (51)	22.0 (44)
Sperm head defectsc	A1	A2	A1	A2	A1	A2	A1	A2
Pyriform (%)	9.5 (19)	11.0 (22)	27.2 (55)	23.0 (46)	19.7 (40)	24.5 (49)	22.0 (44)	25.0 (50)
Round (%)	0	0	0	0.5 (1)	3.4 (7)	2.5 (5)	3.0 (6)	2.0 (4)
Tapered (%)	10.5 (21)	8.5 (17)	23.2 (47)	26.5 (53)	18.7 (38)	20.5 (41)	31 (62)	25.5 (51)
Double (%)	0.5 (1)	0	0.4 (1)	0.5 (1)	0	0	0	0
Amorphous (%)	0	0	4.4 (9)	2.5 (5)	6.4 (13)	8.0 (16)	5.0 (10)	7.5 (15)
Small acrosomal area (%)	0	0	0	0	0.4 (1)	0	0	0
Abnormal residual cytoplasm (%)	2.0 (4)	4.0 (8)	3.4 (7)	2.0 (4)	3.9 (8)	3.0 (6)	2.5 (5)	1.0 (2)
Normal (%)	77.5 (155)	76.5 (153)	41.0 (83)	45.0 (90)	47.3 (96)	41.5 (83)	36.5 (73)	39.0 (78)

^aTwo different semen analyses of the patients were performed with a minimum of 5 days of sexual abstinence; ^b,cFor each patient and control, two different sperm morphological analyses were performed. The values in brackets show corresponding sperm numbers. -: not available; P: patient; A: analyzer

Loss of DNAH10 protein signals in patients’ spermatozoa

To better understand the pathogenic effects of DNAH10 missense mutations, we explored the effect of the identified novel DNAH10 variants on the localization of the DNAH10 protein using immunofluorescence staining. Consistent with previously reported results,15 we found that in the fertile controls, the DNAH10 protein was specifically localized along the sperm flagella. However, such signals were hardly seen in the sperm flagella of the patients harboring the novel DNAH10 compound heterozygous missense variants (Figure 4), indicating that the identified mutations severely impaired DNAH10 protein localization, thereby causing defective spermatozoa.

Figure 4.

Loss of DNAH10 protein signal in spermatozoa from patients from two unrelated Pakistani families. Representative images of spermatozoa from the patients and the fertile control stained with anti-DNAH10 antibody (red), anti-α-tubulin antibody (green), and Hoechst (blue). Scale bars = 10 µm. P1: patient 1, IV:4 in family 1; P2: patient 2, IV:1 in family 2; P3: patient 3, IV:3 in family 2; DNAH10: dynein axonemal heavy chain 10.

DISCUSSION

This study identified novel DNAH10 variants in three patients with the MMAF phenotype from two unrelated Pakistani families using WES. These patients’ spermatozoa exhibited a high percentage of flagellar defects, and more than half had abnormal head morphology (>50%). DNAH10 is expressed in cilia and flagella.30 Eukaryotic cilia and flagella require several dynein microtubule motor complexes for their assembly and movement.31,32 These complexes have one or more DHCs with ATPase and microtubule motor activity. DHCs are large, evolutionarily conserved proteins with approximately 4500 residues.33 All DHC proteins contain a ring of six AAA-ATPase and linker, stalk, buttress, and C-terminal domain extensions. Moreover, axonemal DHCs function as motors associated with axonemal microtubules.34,35 DNAH10 also encodes an axonemal DHC, a component of the IDA on the microtubule doublet of sperm flagella.36 Axonemal dyneins are responsible for ciliary motion in organisms, and anomalies in axonemal dyneins cause human diseases.37

In a previous study, researchers identified compound heterozygous missense, homozygous missense, and frameshift DNAH10 variants in five infertile men with the MMAF phenotype.15 All amino acid mutations were found in conserved domains of the DNAH10 protein, including AAA-ATPases and DHC C-terminal domains. These pathogenic variants led to disorganization of the axonemal ultrastructure, causing male infertility associated with MMAF without PCD symptoms in human and mouse models.15 Similarly, another study identified homozygous frameshift and compound heterozygous missense DNAH10 variants through WES in men with asthenoteratozoospermia who exhibited the MMAF phenotype without PCD symptoms.1 In both these studies,1,15 researchers reported reduced progressive and total sperm motility, with loss of DNAH10 protein expression along the sperm flagella of men harboring pathogenic DNAH10 variants. Consistent with the previous studies, our findings showed reduced sperm motility in the patients.

It is worth noting that progressive sperm motility was as high as 11.5% in our patients, which is higher than that in a previous report.15 This may be due to the differences in mutations. In the study by Tu et al.,15 in two patients with homozygous DNAH10 mutations, the sperm was completely immotile, whereas about 10% of sperm from the other patients were motile. In contrast, in our patients, the DNAH10 mutations were all compound heterozygous missenses and were closer to the C-terminus. In our study, perhaps a small amount of DNAH10 mutant protein was expressed in our patients but was not detected by immunofluorescence staining due to limitations in specimen quality and antibody performance. Other possible reasons for the phenotypic difference are population and geographical differences. Our patients were from Pakistan, whereas all the previous patients were from China.

In summary, we found novel compound heterozygous missense DNAH10 variants that severely impaired DNAH10 protein localization, resulting in morphologically abnormal sperm flagella and reduced sperm motility, causing asthenoteratozoospermia with MMAF without PCD symptoms in patients from two unrelated Pakistani families. Although they have difficulty in conceiving naturally, assisted reproductive techniques, especially intracytoplasmic sperm injection (ICSI), is a promising treatment strategy for patients with MMAF. As these patients carry bi-allelic compound heterozygous mutations in DNAH10, one of the mutations is inevitably transmitted to their offspring. The other allele is inherited from the offspring’s mother, so genetic screening of the patient’s wife is necessary and helpful. If the wife is wild-type for DNAH10, the offspring should be heterozygous, and their health, especially their respiratory and reproductive systems, should not be affected because these mutations are consistent with Mendelian recessive inheritance. If the patient’s wife also carries the mutation in DNAH10, a preimplantation genetic diagnosis of the embryo is necessary. Our research broadens the spectrum of known DNAH10 variants and offers useful information for individuals with MMAF seeking genetic and reproductive counseling.

AUTHOR CONTRIBUTIONS

MS performed most experiments and prepared the draft. MZ, WS, MU, AH, GM, and FR collected patient samples and performed semen analysis. HZ performed the WES analysis. IA, TA, and YR participated in the in silico analysis and graph presentation. SXF and QHS reviewed and revised the manuscript. QHS supervised the findings of this study. All authors read and approved the final manuscript.

COMPETING INTERESTS

All authors declared no competing interests.

Supplementary Figure 1

Variant filtration pipeline of two unrelated Pakistani families. The flowcharts show the variant filtration process utilizing WES data from families 1 (a) and 2 (b). MAF: minor allele frequency; SNV: single nucleotide variant.