Two Novel Heterozygous Variants in RecA2 Domain of DHX37 Cause 46,XY Gonadal Dysgenesis and Testicular Regression Syndrome

DHX37 (DEAH-Box Helicase 37), a member of the DEAH/RHA RNA helicase family, plays an essential role in ribosome biogenesis [Bleichert and Baserga, 2007]. Heterozygous variants in DHX37 have been associated with a range of 46,XY disorders of gonadal development types, particularly 46,XY gonadal dysgenesis and 46,XY testicular regression syndrome (TRS), with the phenotypes ranging from phenotypic females to males with atypical external genitalia or micropenis with cryptorchidism [da Silva et al., 2019; McElreavey et al., 2020; Buonocore et al., 2019; Zidoune et al., 2021; Wan et al., 2023]. The exact role of DHX37 in gonadal development is unknown, but recurrent variants in or adjacent to the highly conserved motifs within two RecA-like domains (RecA1 and RecA2) clearly establish a genotype-phenotype correlation between 46,XY gonadal dysgenesis and DHX37 variants [da Silva et al., 2019; McElreavey et al., 2020; Buonocore et al., 2019; Zidoune et al., 2021; Wan et al., 2023]. Although DHX37 variants account for approximately 10–15% non-syndromic 46,XY complete gonadal dysgenesis (CGD), and 20% 46,XY TRS, so far, only 12 different variants affecting 10 amino acids in or adjacent to the highly conserved RecA-like domains of DHX37 have been reported, all of them are missense variants [da Silva et al., 2019; McElreavey et al., 2020; Buonocore et al., 2019; Zidoune et al., 2021; Wan et al., 2023]. In this study, we discovered two novel heterozygous missense DHX37 variants affecting two additional amino acids which fall within the highly conserved RecA2 domain and expanded the variants spectrum of DHX37 in 46,XY disorders/differences of sex development (DSD).

Cases 1 and 2 were unrelated patients identified by genital abnormalities and absence of secondary sexual features, respectively. Both patients were born to non-consanguineous Chinese parents and had no family history of DSD. The clinical details of the two cases are summarized in Table 1.

Case 1, a 4-year 5-month-old boy, presented with significant micropenis (phallic structure measuring approximately 2.2 cm in length) and cryptorchidism, without palpable gonads bilaterally. There was an apparent urethral orifice with normal position and shape. Hormonal testing showed elevated gonadotropins (FSH 25.5 IU/L [normal range <0.7], LH 1.06 IU/L [<0.2]) and low testosterone levels at baseline and following 3 days human chorionic gonadotropin (hCG) stimulation (both <0.69 nmol/L). Anti-Müllerian hormone and inhibin B levels were also below the normal reference range (0.43 ng/mL [23.70–250.35] and 10.7 pg/mL [16.53–220.27], respectively). Karyotyping revealed 46,XY. Two structures in the inguinal regions (0.4 cm*0.2 cm*0.2 cm, bilaterally) had morphological appearances of testes by ultrasound (US). He underwent cystoscopy, laparoscopy, and inguinal exploration for evaluation of internal structures. Cystoscopy showed a normal urethra without a vagina. Laparoscopy demonstrated that each vas deferens ran toward its respective side of the internal inguinal ring. Neither uterus nor fallopian tubes were observed. Bilateral gonadal biopsy was performed. The histologic examination revealed hypoplastic spermatic cord, focal dystrophic calcification, with sparse immature seminiferous tubules and Leydig cell groups in the stoma, confirming the diagnosis of TRS.

Case 2, a 13-year 6-month-old female patient with phenotypically female external genitalia, was referred to us because of her short stature (145.3 cm, <-2SD for age-matched Chinese girls) and absence of secondary sexual features (B1PH1). The parents’ heights were 154.7 cm and 150.0 cm, respectively. A mild posterior labial fusion was discovered upon further examination of the external genitalia. Endocrinological tests indicated hypergonadotropic hypogonadism with remarkably elevated gonadotropins (LH 41.37 IU/L [1.24–8.64] and FSH 102.55 IU/L [<22.04]), and low E2 (<37 pmol/L [59.1–874.6]), anti-Müllerian hormone (0.01 ng/mL [0.97–11.76]) and inhibin B (<10 pg/mL [12.13–99.2]. Moreover, the patient’s serum testosterone level was 0.16 nmol/L (normal range for females: <2.04). The administration of hCG (1,500 IU/day, intramuscular injection for 3 days) slightly increased testosterone levels (0.16–0.73 nmol/L), suggesting that the gonads had poor testicular function. Karyotyping revealed 46,XY. Bone age was delayed at 11 years (Greulich-Pyle method), and the epiphyses were not fused. No obvious gonadal structure was identified on abdominal-pelvic and inguinal US. Pelvic US demonstrated no evidence of uterine structures. Surgical exploration has not yet been performed for family reasons, but a GD diagnosis can be made based on phenotype and hormonal levels.

Next-generation sequencing, including whole-exome sequencing (WES) and target panel sequencing, was performed on the BGISEQ DNBSEQ-T7 platform (BGI lnc., Shenzhen, China) with a paired-end sequencing length of 150 bp. The sequencing libraries were generated using the KAPA HyperExome Probes (Roche, Basel, Switzerland) for WES. Exome capture and sequencing of genes known to be associated with human sex development, a GenCap panel with 325 genes associated with DSD was customized, and a capture strategy was performed using the GenCap custom enrichment kit (MyGenostics Inc, Beijing, China).

Burrows-Wheeler Aligner was used to align each sample’s clean reads with the reference genome using default parameters. Alignment files were converted to BAM files using SAMtools software. Variant calling was performed for all samples by using the Haplotype Caller in GATK software.

Potentially pathogenic variants were verified with classic Sanger sequencing. Genomic DNA from all available family members was obtained for Sanger sequencing. An ABI 3730XL sequencer was used for Sanger sequencing of the purified PCR products. The sequencing results were analyzed using Mutation Surveyor software. A structural in silico analysis was performed to determine the potential pathogenicity of the variants by SIFT, PolyPhen2, CADD, and MutationTaster. The pathogenicity of variants was categorized according to the criteria recommended by the American College of Medical Genetics and Genomics (ACMG) guideline [Richards et al., 2015].

Trio-WES was performed in family 1. Exome capture and sequencing of genes known to be associated with human sex development were performed for case 2. Sanger sequencing of DHX37 was performed to confirm the variant for all family members in families 1 and 2. Case 1 and his unaffected mother were heterozygous for DHX37, c.1432G>A (p.G478R), but the variant was not detected in his father (shown in Fig. 1a). This variant is considered deleterious by SIFT (0.0), PolyPhen2 (0.995), CADD (24.3), and MutationTaster (disease causing). Case 2, her unaffected eldest sister and mother were heterozygous for DHX37, c.1879C>A (p.L627F), but the variant was not detected in other family members (shown in Fig. 1b). This variant is also considered deleterious by SIFT (0.0), PolyPhen2 (0.998), CADD (27.1), and MutationTaster (disease causing). These two variants were not found in 1000 Genomes, gnomAD, and ExAC database and were not reported in PubMed, ClinVar, HGMD, and LOVD. They were both located in RecA2 domain (within IV and Va motifs, respectively) of the DHX37 protein. The affected amino acids and the surrounding region are conserved up to zebrafish (shown in Fig. 1c). Both variants were classified as VUS (PM1 + PM2 + PP3) according to the recommendation of the ACMG guidelines.

Previous studies have demonstrated a significant genetic link between missense variants in or near the highly conserved motifs within RecA1 or RecA2 and two distinct types of DSD: 46,XY CGD/PGD and 46,XY TRS [da Silva et al., 2019; McElreavey et al., 2020; Buonocore et al., 2019; Zidoune et al., 2021; Wan et al., 2023]. While DHX37 variants account for roughly 10–15% of non-syndromic 46, XY CGD and 20% of TRS cases, only 13 unique variants have been reported [da Silva et al., 2019; McElreavey et al., 2020; Buonocore et al., 2019; Zidoune et al., 2021; Wan et al., 2023]. Among these, 12 are located in or adjacent to the RecA-like domains, with the remaining one, G1030, situated away from these domains [da Silva et al., 2019; McElreavey et al., 2020; Buonocore et al., 2019; Zidoune et al., 2021; Wan et al., 2023]. In our study, we identified two novel heterozygous missense DHX37 variants, p.G478R and p.L627F, in a 46,XY TRS boy and a 46,XY GD girl, located in the IV and Va motifs within the RecA2 domain. For this girl, the mild posterior labial fusion phenotype of the external genitalia, slightly increased testosterone levels after hCG test, and absence of Müllerian derivatives by pelvic US, all pointed toward a suspected PGD diagnosis. However, the final diagnosis hinges on the findings of surgical exploration and gonad biopsy to definitively establish whether it is CGD or PGD. Protein structural domain analysis, amino acid sequence homology alignment, and bioinformatics analyses all strongly support the pathogenicity of these variants. Although the exact mechanism by which DHX37 missense variants cause DSD remains unclear, recurrent variants in or near RecA-like domains establish a clear genotype-phenotype correlation between 46,XY disorders of gonadal development and DHX37 variants [da Silva et al., 2019; McElreavey et al., 2020; Buonocore et al., 2019; Zidoune et al., 2021; Wan et al., 2023].

Consistent with previous studies, these pathogenic/likely pathogenic variants were either maternally transmitted or de novo [da Silva et al., 2019; McElreavey et al., 2020; Buonocore et al., 2019; Zidoune et al., 2021; Wan et al., 2023]. These family cases suggest that the clinical manifestations of DHX37 variants can vary widely, ranging from TRS, atypical genitalia (primarily PGD), to complete 46,XY sex reversal (46,XY CGD). However, paternal inheritance has only been reported in two families, with one asymptomatic heterozygous carrier father confirmed and another predicted carrier father exhibiting unilateral testicular agenesis alongside typical male external genitalia [da Silva et al., 2019; Zidoune et al., 2021], which was necessary to authenticate these atypical phenotypes and inheritance patterns in a broader range of families. Intriguingly, the same variant can result in different phenotypes, leading to various sex assignments for 46,XY DSD family members [da Silva et al., 2019]. The proposed mechanism for 46,XY DSD caused by DHX37 involves disruption of testicular determination to varying degrees [McElreavey et al., 2022]. The exact mechanism responsible for failed testicular determination associated with DHX37 variant remains unknown. Mammalian sex determination is regulated by two opposing genetic pathways, with imbalances potentially leading to DSD [Gonen et al., 2018; Harris et al., 2018; Eozenou et al., 2020]. Our report of two cases with missense variants in the same domain of the DHX37 protein, resulting in completely contrasting sex phenotypes, adds to the complexity of understanding DHX37’s mechanism. The most compelling theory proposed so far is the nucleolar stress model [McElreavey et al., 2022], which suggests that nucleolar stress from DHX37 variants leads to a rapid, transient increase in WNT signaling and subsequent β-catenin stabilization [Dannheisig et al., 2021]. This disruption may interfere with testis determination and result in 46,XY gonadal dysgenesis (CGD/PGD). WNT signaling in XY somatic cells of the developing gonad could then trigger a p53-dependent pro-apoptotic response [Bursać et al., 2012], causing an absence of gonadal tissue and thus lead to TRS. This raises the possibility that DSD caused by DHX37 variants may be ribosomopathies [Boneberg et al., 2019; McElreavey et al., 2022], but the underlying mechanism requires further investigation.

In conclusion, we identified two novel heterozygous missense DHX37 variants, p.G478R and p.L627F, in a 46,XY TRS boy and a 46,XY GD girl. These variants, located in the IV and Va motifs within the RecA2 domain, expand the currently limited variant spectrum of DHX37 in 46,XY DSD.

Written informed consent was obtained from the patient and the parents for publication of the details of their medical case and any accompanying images. The study was approved by the Ethics Committee of Children’s Hospital, Zhejiang University School of Medicine (approval number: 2021-IRB-032).

The authors have no conflicts of interest to declare.

The authors acknowledge the financial support from the National Key Research and Development Programme of China (No. 2021YFC2701901, No. 2016YFC1305301), National Natural Science Foundation of China (No. 81570759 and No. 81270938), Zhejiang Provincial Key Disciplines of Medicine (Innovation Discipline, 11-CX24), Science and Technology Fund Project of Guizhou Provincial Health Commission (No. gzwkj2021-286).

Junfen Fu planned the study; Xiuqi Ma, Hongjuan Tian, Jinna Yuan, Dehua Wu, Guanping Dong, and Qian Liu treated the patient and contributed to the acquisition and interpretation of clinical data and images. Hao Yang carried out the genetic analysis and bioinformatic analysis. Hao Yang, Xiuqi Ma, and Junfen Fu drafted and reviewed the manuscript. All co-authors approved the manuscript. Hao Yang and Xiuqi Ma have contributed equally to this work.

Hao Yang and Xiuqi Ma contributed equally to this work.