Compound heterozygous variants of the DMRTC2 gene are associated with non-obstructive azoospermia in a patient

Dear Editor,

Male infertility comprises 50% of infertility cases globally, and one of the most severe forms is non-obstructive azoospermia (NOA), which affects approximately 10%–15% of infertile men.1 Genetic factors play an important role in the pathogenesis of NOA. Recent research progress has increasingly elucidated the pathogenesis of NOA. Multiple studies, particularly those using whole-exome sequencing, have identified an increasing number of pathogenic genes, including meiosis specific with OB-fold (MEIOB), telomere repeat binding bouquet formation protein 1 (TERB1), and ubiquitin-specific peptidase 26 (USP26)2,3,4 associated with NOA. These findings have greatly advanced our understanding of the molecular mechanisms underlying the failure of spermatogenesis. However, the genetic etiology of NOA for a considerable number of patients remains to be identified.

DMRT-like family C2 (DMRTC2; also known as DMRT7 in mice) is located on human chromosome 19, which is specifically expressed in testis tissue in humans and mice.5 Previous research has shown that Dmrtc2^−/− male mice are infertile, and they exhibit phenotypes of meiotic arrest.6 However, there have been no clinical reports of this gene variant causing NOA in humans. Our study reported the novel pathogenic variants of DMRTC2 found in a NOA patient. The methodologies are available in Supplementary Participants and Methods.

This study has been approved by the Ethics Committee of the National Institute of Family Planning (Beijing, China; Approval No. 2021010), and all participants in this study have signed written informed consent forms, agreeing that their anonymized data would be used for the analysis and publication of the results of this study. In this study, a 35-year-old man from a Chinese family had a 2-year history of infertility without contraception (Figure 1a). Bilateral, small-sized testes (12 ml; reference range: >14 ml) were detected, with no nodules in the bilateral epididymis and no varicose veins in the bilateral spermatic cord. Ultrasound examinations showed no abnormalities in the testes and epididymis. Semen examination showed complete azoospermia. A hormonal analysis showed normal blood concentrations of testosterone (12.43 nmol l⁻¹; reference range: 10.4–34.7 nmol l⁻¹) and follicle-stimulating hormone (7.24 IU l⁻¹; reference range: 2.0–7.6 IU l⁻¹) with an elevation of luteinizing hormone (8.18 IU l⁻¹; reference range: 1.8–5.2 IU l⁻¹). The histology of the testis showed spermatogenic arrest. The seminiferous tubules of the patient were found to contain spermatocytes and spermatids with the absence of spermatozoa (Johnsen score = 5), which manifested maturation arrest (Figure 1b). The patient showed a normal karyotype, and no microdeletion was identified on the Y chromosome.

Figure 1.

Identification of the novel DMRTC2 variants in a patient with NOA from a Chinese family. (a) Pedigree for the family with the variants in DMRTC2 gene. The black solid symbols represent individuals with NOA, and the arrow points to the proband (II-1). (b) Testicular histological analysis of the patient showed dysplasia of spermatogenic cells. The image shows seminiferous tubule structures where no spermatozoa is observed, but spermatocytes and spermatids are present. Scale bar = 100 µm (left) and 25 µm (right). (c) Sanger sequencing validated WES results. The red boxes indicate the locations of the DMRTC2 variants. As shown, the proband carries two mutations, while each of his unaffected parents carries only one of them. (d) Conservation analysis of the p.R46H variant. The black box indicates the mutated amino acid. Conservation analysis showed that the variant occurred at highly conserved residues through evolution across several species. DMRTC2: DMRT-like family C2; NOA: non-obstructive azoospermia; WES: whole-exome sequencing.

We identified the pathogenic variants carried through the patient by whole-exome sequencing. After bioinformatic analysis and screening, we identified that the patient carried two heterozygous variants located in the DMRTC2 gene on chromosome 19: c.G137A(p.R46H) and c.862_863insGGCAGCTGCAGCAAGA(p.E296Afs*4). Among the genes in the shortlist (Supplementary Table 1), the DMRTC2 gene is specifically expressed in male testis tissue cells, and Dmrtc2^−/− male mice display a phenotype characterized by spermiogenesis arrest.5

Supplementary Table 1.

Candidate pathogenic variations and in silico bioinformatics prediction in the affected individual

Gene	Chromosome	Type	Variation	GnomAD_ALL	GnomAD_EAS	SIFT	PP2	MT	CADD	Zygosity
OBSCN	1	Nonsynonymous SNV	NM_001098623:exon11:c.G3265A(p.G1089R)	0.000006818	0.0001337	Deleterious	Probably damaging	Polymorphism	23	Heterozygous
		Nonsynonymous SNV	NM_001098623:exon55:c.G15035A(p.R5012H)	0.00003906	0.0003563	Deleterious	Probably damaging	Disease causing	26	Heterozygous
PDE6B	4	Nonsynonymous SNV	NM_000283:exon1:c.T401C(p.L134P)	0.000003761	0.0001364	Deleterious	Probably damaging	Disease causing	26.2	Heterozygous
		Nonsynonymous SNV	NM_001350155:exon16:c.G1138C(p.A380P)	0.00007812	0.002697	Deleterious	Possible damaging	Disease causing	23.8	Heterozygous
NAA11	4	Nonsynonymous SNV	NM_032693:exon1:c.C227T(p.A76V)	0.0001976	0.005393	Deleterious	Possible damaging	Disease causing	26.3	Homozygous
BTNL2	6	Stopgain	NM_001304561:exon6:c.C1358A(p.S453X)	6.206e-7	0.000				58	Heterozygous
		Nonsynonymous SNV	NM_001304561:exon3:c.G619A(p.V207M)	0.000006821	0.000	Tolerated	Possible damaging	Polymorphism	15.66	Heterozygous
USP17L2	8	Nonsynonymous SNV	NM_201402:exon1:c.A793C(p.K265Q)	0.00004022	0.001477	Deleterious	Probably damaging	Polymorphism	22.4	Homozygous
SETX	9	Nonsynonymous SNV	NM_001351527:exon26:c.C7741T(p.H2581Y)	0.00001611	0.0003121	Tolerated	Benign	Polymorphism	2.028	Heterozygous
		Nonsynonymous SNV	NM_001351527:exon12:c.C5536T:p.R1846C	0.00005950	0.001809	Deleterious	Probably damaging	Disease causing	32	Heterozygous
DMRTC2	19	Nonsynonymous SNV	NM_001040283:exon2:c.G137A(p.R46H)	0.000006815	0.00002228	Tolerated	Probably damaging	Disease causing	24.4	Heterozygous
		Frameshift insertion	NM_001040283:exon8:c.862_863insGGCAGCTGCAGCAAGA(p.E296Afs*4)							Heterozygous
CDC45	22	Nonsynonymous SNV	NM_001178011:exon4:c.G305A(p.S102N)	0.00001425	0.000	Tolerated	Benign	Disease causing	16.59	Heterozygous
		Nonsynonymous SNV	NM_001178011:exon16:c.A1480G(p.N494D)	0.00004771	0.001582	Tolerated	Benign	Disease causing	22	Heterozygous
PRR14L	22	Nonsynonymous SNV	NM_173566:exon4:c.T4463G(p.L1488R)	0.00001740	0.0005623	Deleterious	Probably damaging	Polymorphism	22.8	Heterozygous
		Nonsynonymous SNV	NM_173566:exon4:c.T3178C(p.C1060R)	0.00003414	0.001124	Deleterious	Benign	Polymorphism	9.916	Heterozygous
TUBGCP6	22	Nonsynonymous SNV	NM_020461:exon24:c.C5351A(p.S1784Y)	0.000004353	0.0001119	Deleterious	Probably damaging	Disease causing	29.6	Heterozygous
		Nonsynonymous SNV	NM_020461:exon21:c.C4664T(p.P1555L)	0.00001611	0.0002453	Deleterious	Probably damaging	Disease causing	25.3	Heterozygous
GEMIN8	X	Nonsynonymous SNV	NM_001042480:exon4:c.G652A(p.A218T)	0.00001902	0.0003553	Tolerated	Probably damaging	Disease causing	21.9	Hemizygous

The gnomAD, http://gnomad.broadinstitute.org; SIFT, http://sift-dna.org; PP2, http://genetics.bwh.harvard.edu/pph2/, MT, http://mutationtaster.org; CADD, https://cadd.gs.washington.edu, higher scores are more deleterious. SIFT: sorting Intolerant from Tolerant; gnomAD: Genome Aggregation Database; PP2: polymorphism Phenotyping, version 2; MT: mutation taster; CADD: combined annotation dependent depletion

The result of Sanger sequencing showed that the two variants carried by the patient were inherited separately from his parents (Figure 1c). In the gnomAD database, the overall frequency and the frequency in the East Asian population of the missense variant c.G137A are both sufficiently low. Conservation assessment shows that the amino acid residue is highly conserved across different species (Figure 1d). Based on the analysis results of polymorphism phenotyping, version 2 (PolyPhen-2, possibly damaging), MutationTaster (disease causing), and combined annotation-dependent depletion (CADD, score 24.4), this variant is predicted to have a deleterious impact on the structure and function of DMRTC2 (Supplementary Table 2). The other variant c.862_863insGGCAGCTGCAGCAAGA was not available in the gnomAD database, indicating its rarity. The variant results in an insertion of 16 bases, causing a frameshift and premature termination codon, potentially yielding a truncated protein. Based on the above results, we infer that the compound heterozygous variants in DMRTC2 are the pathogenic variants for NOA in this patient.

Supplementary Table 2.

Bioinformatics prediction of DMRT-like family C2 variants identified in the non-obstructive azoospermia patient

Position (GRCh38)	Variant (NM_001040283)	Frequency&	In silico bioinformatics prediction
			SIFT	Polyphen-2	MT	CADD
Chr19: 41847565	c.G137A(pR46H)	0.000006815/0.00002228	Tolerated	Probably damaging	Disease-causing	24.4
Chr19: 41850571	c.862_863insGGCAGCTGCAGCAAGA(p.E296Afs*4)	−/−	-	-	-	-

^&gnomAD frequency in overall population and East Asian population. SIFT: sorting intolerant from tolerant; Polyphen 2: polymorphism Phenotyping, version 2; MT: MutationTaster, CADD: combined annotation dependent depletion

DMRT genes encode a highly conserved family of transcription factors. In a variety of vertebrates, DMRT genes play crucial roles in spermiogenesis.7 Prior investigations have demonstrated the pathogenicity of Dmrtc2 in mice; however, DMRTC2 variants have not been identified in humans.6 In the Dmrtc2^−/− mice, most Dmrtc2 mutant germ cells undergo apoptosis after being arrested during the pachytene stage of meiosis. Although a few of the mutant cells can survive to diplonema, they show defects in sex chromosome epigenetics during the subsequent transition from the pachytene stage to the diplotene stage, which leads to spermatogenic arrest in the Dmrtc2^−/− mice.8

Similar to some genes mentioned in a recent study on NOA-related genes, DMRTC2 is speculated to cause meiotic arrest in humans.9 The patient of the current study exhibited spermatogenic arrest at meiosis and the absence of spermatozoa. Although testicular sperm extraction was not performed on the patient, the analysis of similar cases suggests that the testicular extraction result would also be negative.9 The testicular pathological phenotype of this patient aligns with the phenotypes observed in the Dmrtc2^−/− mice. In the seminiferous tubules of Dmrtc2^−/− males, spermatocytes and spermatogonia were visible with a few round spermatids and no elongating spermatids.10 The result of bioinformatics prediction showed that both variants carried by the patient may be deleterious. This suggests that the reason for the block in meiotic progression caused by the lack of DMRTC2 may be the same between humans and mice.

It is important to note that while knockout mouse models can provide valuable functional insights, not all human variants result in phenotypes in model organisms. Therefore, this study has certain limitations, and the current results need further evaluation in functional studies in appropriate biological models. In the present study, testicular biopsy was only performed in a specific area of the right testis and may not fully reflect the pathological status of the entire testis. Because of the multifocal nature of testicular histopathological findings, this sampling method may limit a comprehensive assessment of testicular pathological changes.

In conclusion, we demonstrated that compound heterozygous variants in DMRTC2 represent a novel genetic cause of NOA in men, characterized by spermatogenic failure. Our findings expand the current evidence regarding genetic causes of NOA and provide valuable insights that may advance the development of improved clinical diagnostic strategies.

AUTHOR CONTRIBUTIONS

XYG, YS, and BBW contributed to the study design. XBL, MY, and XYG collected and interpreted the patient data. YM, TYL, and LXR analyzed the data of whole-exome sequencing and Sanger sequencing. YM drafted the manuscript. All authors read and approved the final manuscript.

COMPETING INTERESTS

All authors declare no competing interests.

SUPPLEMENTARY PARTICIPANTS AND METHODS

Study participants

A Chinese male infertility patient and his family were involved in this study. This study has been approved by the Ethics Committee of the National Institute of Family Planning, and all participants in this study have signed written informed consent forms. A detailed investigation of the patient's family and birth history was conducted, and his reproductive organs were examined by ultrasound to valuate other etiologies of infertility. The patient underwent semen analysis and microscopic examinations, along with a biopsy of the right testis, to establish a diagnosis of NOA. Additionally, the patient's karyotype and Y-chromosome microdeletions were examined.

Whole exome sequencing

The peripheral blood was collected from the patient and his parents. Extract Genomic DNA from peripheral blood samples using the QIAamp DNA Blood Mini Kit (Qiagen, Germany), and the operating procedures were carried out according to the instructions provided by the manufacturer. Exome capture was performed for DNA samples using the SureSelect Human All Exon V6 Enrichment kit (Agilent, California, USA). Sequencing was performed on the NovaSeq platform (Illumina, California, USA). The human genome assembly sequence (UCSC hg38) aligned to all reads using the Burrows–Wheeler Aligner software v0.7.9. During the screening of candidate variants, variants that met the following conditions were retained: missense, nonsense, frameshift, non-frameshift, or splice site variants; variants presenting low allele frequencies of less than 1% in the gnomAD v4.1.0 database (http://gnomad.broadinstitute.org). For genes on the X-chromosome, require the allele frequency of <0.1%.

Validation by Sanger sequencing

The candidate pathogenic variants in the DMRTC2 gene identified in the patient were verified by Sanger sequencing.

Oligonucleotide primers near the candidate variant were designed to amplify the region of interest for the pedigree participants by polymerase chain reaction (PCR): F1: ACCTGGCTATGCTTACCTTC;R1: CCTCCTGTACCCAAACTCC;F2: ACCACCTGAGTAACTGAGCG;R2: TTCCCAGCCATCCCTATC.

Ethics approval and consent to participate

We adhered to the Code of Ethics of the World Medical Association (Declaration of Helsinki, revised in 2013), and the ethical committee of the National Research Institute for Family Planning approved this study. All family members participating in this study signed informed consent.