Human sex differentiation from bipotential gonads to testes or ovaries is a complex process involving the interaction of several genes [1,2]. In XY embryos, the expression of sex-determining region Y (SRY) initiates a cascade of testicular formation pathways via the upregulation of SRY-box-9 (SOX9), followed by other SOX family members, including SRY-box-3 (SOX3), SRY-box-8 (SOX8), and SRY-box-10 (SOX10) [1-3]. In XX embryos, in the absence of SRY, pro-ovarian factors, such as Wnt family member 4 (WNT4) and R-spondin 1 (RSPO1) stabilize β-catenin and antagonize SOX9, resulting in ovarian differentiation [2,3].

46,XX testicular or ovotesticular disorders of sex development (DSD) is a rare condition with a prevalence of approximately 1 per 20,000 newborn males [4]. Approximately 80-90% of these cases are caused by translocation of the SRY allele onto the X chromosome, known as SRY-positive [5]. Testicular development may also occur in SRY-negative conditions and is caused by increased expression of pro-testis genes or insufficient expression of pro-ovarian genes [1,3,6]. SRY-negative (ovo)testicular DSD has been reported in XX individuals with duplications of SOX3, SOX9, and SOX10 [2]. In addition, loss-of-function variants of WNT4 and RSPO1 have been identified in a few individuals with syndromic 46,XX (ovo)testicular DSD [2].

Nuclear receptor subfamily 5 group A member 1 (NR5A1), also known as steroidogenic factor-1 (SF1), is a nuclear receptor transcription factor that regulates adrenal and reproductive development [7]. NR5A1 mediates the initial step of gonadal development in both XY and XX embryos [2,7]. It upregulates SOX9 synergistically with SRY in XY embryos, and induces WNT4 and RSPO1 expression in XX embryos [2-4]. Loss-of-function variants of NR5A1 have been characterized in 46,XY DSD, gonadal dysgenesis and 46,XX primary ovarian insufficiency [2,3,7]. The case of 46,XY female who carried a missense variant of c.205C>G; p.R68G in NR5A1 has been reported in Korea [8]. Recently, six reports have described a missense variant of NR5A1 (c.274C>T; p.Arg92Trp) as the cause of 46,XX (ovo)testicular DSD [9-14] in 15 patients. Additional NR5A1 variants, c.275G>A; p.Arg92Gln and c.779C>T; p.Ala260Val, have also been reported in 46,XX individuals with (ovo)testicular DSD [12,15].

Herein, we report the rare Korean case of 46,XX testicular DSD with a heterozygous NR5A1 p.Arg92Trp variant, which presented with a small penis, bifid scrotum, and bilateral undescended testes.

1. Subject

A 27-month-old male was referred to the Seoul National University Hospital for evaluation of microphallus and small testicular size. He was the first child in dichorionic diamniotic twins of non-consanguineous healthy parents and was born through in vitro fertilization. On physical examination, his height, weight, and head circumference were 84.6 cm (6th percentile), 11 kg (10th percentile), and 44 cm (<3rd percentile), respectively. The patient had normal developmental milestones. He presented with male-type external genitalia with a micropenis, bifid scrotum, and bilateral undescended testes. Pelvic ultrasonography and magnetic resonance imaging showed bilateral normally shaped testes and epididymis in both supra-scrotal areas, and no uterus, ovary, or vagina was detected (Fig. 1). His gonadotropin levels were comparable to those of age- and sex-matched healthy controls: luteinizing hormone 0.07 mIU/mL and follicle stimulating hormone 0.03 mIU/mL. A proper testosterone response was observed after three days of stimulation with human chorionic gonadotropin (basal 0.02 ng/dL, peak 0.23 ng/dL). He underwent bilateral scrotal orchiopexy and scrotoplasty at the age of 2 years and 5 months. He also underwent testosterone replacement because of a small penis immediately after surgery (once at 4.8 years old) and later during childhood (four times during 8.1 and 9.0 years old) at the pediatric urology clinic. At the latest visit (at 9.2 years old), he showed an adequately sized penis, with a stretched penile length of 5 cm.

2. Cytogenetic and molecular genetic analysis

The research protocol was approved by the Institutional Review Board of Seoul National University Hospital (IRB no. 2206-190-1336). Written informed consent for genetic analysis was obtained from the parents of the patient.

Chromosome analysis of peripheral blood leukocytes revealed a 46,XX karyotype. Fluorescence in situ hybridization with an SRY-specific probe excluded SRY translocations on one X chromosome. Duplication of SOX9 was ruled out using real-time PCR assay. After a diagnosis of 46,XX (SRY-negative) testicular DSD, chromosome analysis of peripheral blood leukocytes from his younger twin sibling with a female phenotype was performed and revealed a normal female karyotype (46,XX). Considering the possible confined blood chimerism that can occur in diamniotic twins sharing a common placenta [16], short tandem repeat (STR) polymorphism analysis was performed. Skin fibroblasts of the proband and blood leukocytes from younger siblings and both parents were used to confirm the dizygosity of the twins. The twins were determined as dizygotic because of the nine discordant genotypes among the 16 STR markers. In the proband, karyotyping using skin fibroblasts also revealed 46,XX, and the patterns of all 16 STR markers were identical between the skin and peripheral blood samples, ruling out the possibility of mosaicism.

Whole exome sequencing was subsequently performed, and a previously reported heterozygous missense variant of NR5A1 (c.274C>T, p.Arg92Trp) was identified as the best candidate for the patient’s phenotype (likely pathogenic, PM1+PM2+PP3+PP4+PP5). There were no other pathogenic or likely pathogenic variants of sex differentiation-related genes, including SRY, SOX3, SOX9, and DAX1. Genetic analysis was also performed on his parents, and the parents did not harbor this variant, indicating his variant as de novo.

Human gonads are bipotential until about six weeks of gestation, then differentiate into testes or ovaries [1,2]. In developing testes, SRY leads to the upregulation of SOX9 and subsequent stimulation of testis-specific pathways. During normal ovarian development, ovary-specific genes, such as RSPO1 and WNT4, modulate ovary-specific signaling pathways via β-catenin by antagonizing SOX9, and NR0B1 represses the NR5A1-induced transactivation of SOX9 (Fig. 2) [1-3]. Dysregulation of these delicately balanced processes of testicular and ovarian differentiation can induce 46,XX (ovo)testicular DSD [4,10]. In the absence of SRY, gain-of-function variants of pro-testicular genes, such as SOX3, SOX9, and SOX10, or loss-of-function variants of pro-ovarian genes, such as RSPO1, WNT4, and NR0B1, can lead to SRY-negative 46,XX sex reversal [1,10].

NR5A1 plays an essential role in both male and female gonadal development and maintenance by upregulating SOX9 in the XY embryo, while upregulating pro-ovarian factors such as WNT4 and RSPO1, and anti-testis factors such as NR0B1 in the XX embryo [1-3]. To date, over 80 NR5A1 variants have been reported to be associated with various gonadal development disorders, ranging from primary ovarian failure in 46,XX individuals to (ovo)testicular 46,XX DSD and oligospermia in 46,XY individuals [7,17]. The p.Arg92Trp variant observed in our study has been previously reported in 15 patients from 13 families with 46,XX (ovo)testicular DSD (Table 1). A recurrent heterozygous p.Arg92Trp variant of NR5A1 was first reported in six individuals from four unrelated families of different ancestries who presented with ambiguous genitalia of variable degrees [9]. Five of them had 46,XX karyotype, but a sibling in one family raised as a girl was found to have 46,XY partial gonadal dysgenesis [9]. Another study reported three unrelated Caucasian patients with 46,XX (ovo)testicular DSD; two inherited NR5A1 p.Arg92Trp variant from their unaffected mothers, suggesting incomplete penetrance [10]. The same NR5A1 variant has been reported in two unrelated Japanese males with 46,XX (ovo)testicular DSD [11,13] and an Iranian infertile male with 46,XX testicular DSD [14]. An additional NR5A1 variant, c.779C>T; p.Ala260Val, has been reported in a patient with 46,XX ovotesticular DSD [12]. A paternally inherited NR5A1 c.275G>A; p.Arg92Gln variant in the same codon as c.274C>T; p.Arg92Trp variant, has also been reported in a patient with 46,XX ovotesticular DSD [15].

Several studies have suggested that the NR5A1 p.Arg92Trp variant activates testis development in the XX background. Human NR5A1 is expressed in ovarian somatic cells and can regulate cell fate at an early stage [9]. In silico analysis predicted that the NR5A1 p.Arg92Trp variant affects a highly conserved amino acid in the RGGR motif of the A-box, involving DNA-binding specificity and stability [9-11]. The NR5A1 p.Arg92Trp variant showed reduced ability to upregulate SOX9 and reduced synergy with β-catenin, resulting in reduced activation of the pro-ovary canonical WNT/RSPO1 signaling pathway [9,10]. In vitro assays demonstrated that the NR5A1 p.Arg92Trp variant was less reactive to NR0B1 [11]. In addition, the NR5A1 variant caused increased repression of ovarian-specific WNT/β-catenin signaling, resulting in reduced NR0B1 expression and decreased testis pathway inhibition [12]. Although the exact pathophysiology of NR5A1 variants causing 46,XX sex reversal has not been clearly elucidated, these findings provide evidence that NR5A1 is involved in sex differentiation and its variants lead to testis differentiation in the XX gonad.

The NR5A1 gene variants are also known to be related to primary adrenal insufficiency with or without sex reversal [18,19]. Adrenal insufficiency without sex reversal is extremely rare, and in only one report, a 46,XX girl with primary adrenal insufficiency has been reported to have a heterozygous NR5A1 p.R255L variant [20]. Accompanying adrenal insufficiency in NR5A1 variants also has been reported in 46,XY DSD, while not yet in 46,XX DSD [18,19]. Our case patient also had no symptom or sign of adrenal insufficiency, and we did not perform an ACTH stimulation test.

To the best of our knowledge, this is the first report of a Korean patient with 46,XX testicular DSD caused by a heterozygous NR5A1 p.Arg92Trp variant, although there has been a previous case report of 46,XY DSD with NR5A1 variant in Korea [8]. Our case report can be additional evidence that NR5A1 is involved in a series of gonadal development processes from bipotential gonads. NR5A1 gene variants need to be considered as important causative factors of SRY-negative non-syndromic 46,XX testicular DSD. Further investigations are necessary to demonstrate the exact molecular mechanism of NR5A1 variants in the development of 46,XX (ovo)testicular DSD.

The authors declare that they do not have any conflicts of interest.

Conception and design: JMK. Acquisition of data: LK, HYK. Analysis and interpretation of data: LK, HYK. Drafting the article: LK. Critical revision of the article: HYK, JMK. Final approval of the version to be published: all authors.