Diagnostic Application of Targeted Next-Generation Sequencing of 80 Genes Associated with Disorders of Sexual Development

Scientific Reports, Mar 2017

Disorders of sexual development (DSD) are estimated to occur in 1 of 4500 births. Since the genetic etiology of DSD is highly heterogeneous, obtaining a definitive molecular diagnosis by single gene test is challenging. Utilizing a high-throughput sequencing upfront is proposed as an efficient approach to aid in the diagnosis. This study aimed to examine the diagnostic yield of next-generation sequencing in DSD. 32 DSD patients that previously received clinical examinations and single gene tests were selected, with or without a diagnosis. Prior single gene tests were masked, and then samples went through targeted next-generation sequencing of 80 genes from which the diagnostic yield was assessed. A likely diagnosis, with pathogenic or likely pathogenic variants identified, was obtained from nine of the 32 patients (i.e., 28.1%, versus 10% by single gene tests). In another five patients (15.6%), variants of uncertain significance were found. Among 18 variants identified (i.e., 17 single nucleotide variants and one small deletion), eight had not been previously reported. This study supports the notion that next-generation sequencing can be an efficient tool in the clinical diagnosis and variant discovery in DSD.

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:

https://www.nature.com/articles/srep44536.pdf

Diagnostic Application of Targeted Next-Generation Sequencing of 80 Genes Associated with Disorders of Sexual Development

Abstract Disorders of sexual development (DSD) are estimated to occur in 1 of 4500 births. Since the genetic etiology of DSD is highly heterogeneous, obtaining a definitive molecular diagnosis by single gene test is challenging. Utilizing a high-throughput sequencing upfront is proposed as an efficient approach to aid in the diagnosis. This study aimed to examine the diagnostic yield of next-generation sequencing in DSD. 32 DSD patients that previously received clinical examinations and single gene tests were selected, with or without a diagnosis. Prior single gene tests were masked, and then samples went through targeted next-generation sequencing of 80 genes from which the diagnostic yield was assessed. A likely diagnosis, with pathogenic or likely pathogenic variants identified, was obtained from nine of the 32 patients (i.e., 28.1%, versus 10% by single gene tests). In another five patients (15.6%), variants of uncertain significance were found. Among 18 variants identified (i.e., 17 single nucleotide variants and one small deletion), eight had not been previously reported. This study supports the notion that next-generation sequencing can be an efficient tool in the clinical diagnosis and variant discovery in DSD. Introduction Sexual development is an orchestrated pathway that directs undifferentiated zygotes towards male or female destiny1. Disruption of the tightly regulated process of sex determination and differentiation can result in disorders of sexual development (DSD), defined as a state when the “development of chromosomal, gonadal or anatomical sex is atypical”2. The genetic causes of DSD are highly heterogeneous, complicated by the associated genetic alterations that include copy number variations (CNVs), single nucleotide variations (SNVs) and small insertions and deletions (InDels)3. Etiological diagnosis of DSD usually requires a wide spectrum of endocrinological tests, radiological imaging and genetic tests2. Traditionally, clinicians conduct physical examinations, endocrine tests and radiological imaging in the first instance, and then request single genetic tests to confirm the diagnosis. Thus far, no study has yet been published regarding the diagnostic yield of Sanger sequencing following this traditional routine in a large DSD cohort with mixed conditions. Sporadic studies on specific DSD condition hint that the yield might be low. For example, a study in Korean patients with Kallmann syndrome, which is a relatively common DSD seen in the endocrinology clinic, showed that molecular defects were found in 16% of the cohorts after a thorough endocrine/radiological workup and Sanger sequencing of 10 genes4. The actual yield in the practice is probably lower, considering the single gene tests requested for a specific condition are usually less than 10. This is also a lengthy process. It could be very stressful when medical treatments or surgical decisions are pending. To address these challenges, one proposal is to initially sequence all the current known genes of DSD5, and then conduct more specific endocrine and clinical tests guided by the obtained genetic results. Recent advances in next-generation sequencing have greatly accelerated the discovery of genetic variants. The Vilain lab conducted two studies, one of which, was a targeted panel sequencing of 35 DSD genes that revealed genetic defects in two of the seven patients not previously diagnosed, and confirmed the diagnosis in another seven patients with known genetic causes6. In another study, whole exome sequencing followed by analysis of selected DSD genes reached a diagnostic yield of 22.5% in 40 patients presenting with 46XY DSD7. A more recent study by the Liang lab showed a relatively higher yield by a targeted panel sequencing approach, in which a genetic diagnosis in eight out of 21 (38.1%) DSD patients was obtained, with patients harboring copy number variations included8. These studies demonstrated the diagnostic potential of a high-throughput approach like next-generation sequencing (NGS). In this study, we designed a targeted panel to sequence 80 genes, including known genes associated with human DSD and some recently identified genes that influence the pathways of sex determination or differentiation. Thirty-two Chinese patients with DSD were recruited, each of whom had been subjected to the traditional approach of endocrine analysis and radiological imaging followed by single genetic tests. Their prior genetic testing results were blinded in the sample selection, thus inclusion in this study was not predicated on whether a diagnosis was previously established. The diagnostic yield was compared between the single genetic test and high-throughput sequencing to evaluate the potential of NGS utilization in a clinical setting. This current study aimed to investigate the diagnostic potential of high-throughput sequencing in a mixed cohort of DSD, to provide the mutation spectrum of DSD patients in the Chinese population, and to potentially discover novel genetic variants in DSD. Materials and Methods Subject selection Subjects were selected from patients visiting the pediatric endocrinology clinic at Xinhua hospital or the Children’s medical center, Shanghai, China during August to December 2014. Most patients were from Eastern China. The recruitment was based on the clinical criteria of DSD diagnosis9. The inclusion required consent from patients and family, complete clinical records and access to prior results of biochemical, radiological and single genetic tests (if requested). Karyotype analysis was done in all patients as routine. Those patients with chromosomal abnormality known to cause the DSD phenotype, for example 45,XO Turner’s syndrome, Klinefelter’s syndrome, dup Xp21.3 (with the NR0B1 gene duplication known to cause 46,XY complete gonadal dysgenesis) and others were excluded. Subjects with a clear indication of classic congenital adrenal hyperplasia were also excluded. Subjects may have gone through biochemical assays, ultrasound and other clinical workup. A total of seven pediatric endocrinologists were involved in this study, and most of the 32 patients were seen by two clinicians. Genetic tests by Sanger sequencing of single genes were requested following the clinician’s professional judgment (at least one main phenotype matches the genetic condition) on the most likely genetic causes. In subject selection and subsequent analysis, the results of all prior genetic tests were not revealed to the investigator. All patients selected were derived from unrelated families. The study was reviewed and approved by the ethical committee at Xinhua hospital, all methods were performed in accordance with the relevant guidelines and regulations, and informed consent was obtained from the patients and parents (i.e., for patients under 18 years old). Selection of candidate genes The list of 80 DSD genes was generated by searching online database and the available literature including OMIM (http://www.omim.org), HGMD (http://www.hgmd.cf.ac.uk/) and Pubmed (http://www.ncbi.nlm.nih.gov/pubmed). The keywords that were used in this online search included: “gonadal”, “sex development”, “hypogonadism”, “testicular”, “ovarian”, “hormone” and their combinations with “disorder” or “disease”. The list was then curated to include those genes that were reported in at least one human DSD case, or that had multiple lines of evidence suggesting its role in sexual development based on the model organism (please see Table 1). Table 1: List of genes, related phenotype, inheritance and coverage in the DSD panel. Full size table Sample preparation, sequencing and analysis Approximately 2 ml of peripheral venous blood was drawn from patients to standard K3EDTA-tubes, and genomic DNA was extracted with the GentraPuregene Kit (Qiagen, Germany) or the Lab-Aid 820 kit (ZSandx, China). Coding and flanking regions of the 80 DSD genes were amplified using a customized Ion Ampliseq panel (Life Technologies). The preparation of each library was completed according to the manufacturer’s instructions, including automated template preparation by the Ion One-touch System. All sample preparations were then sequenced on the PGM with 316 or 318 chip (Ion PGM Sequencing 200 kit v2, Life technologies). Approximately 0.4G of raw data was obtained for each sample. Data analysis was performed on the Torrent server. Base calling, read filtering, alignment to the human genome (hg19), and variant calling were done with Ion Torrent Suites (Life technologies). Picard (http://picard.sourceforge.net/) and the Genome Analysis Tool Kit (GATK, http://www.broadinstitute.org/gatk/) were used for further QC analysis, coverage analysis and variant filtering, and snpeff/snpsift (http://snpeff.sourceforge.net/) were used to annotate the variants. For basic filtering, variants with a coverage depth less than five-fold or a Phred-like score below 30 were excluded. Subsequently, variants with a frequency >1% in the Exome Aggregation Consortium (ExAC, http://exac.broadinstitute.org/), 1000 Genome Project (http://www.1000genomes.org/) or Exome Sequencing Project (ESP, http://evs.gs.washington.edu/EVS/), or >20% in our in-house database were excluded (based on 72 whole exome datasets and 130 datasets from Agilent Clearseq inherited disease panel). It should be noted that a large portion of our in-house datasets were from patients with endocrine disorders, and thus a high threshold was applied to avoid over-filtration due to the enrichment of DSD samples. Quality metrics were generated and evaluated (with the description found in the Results section 3.2). All variants of interest were visually inspected in the Integrative Genomics Viewer (https://www.broadinstitute.org/igv) to evaluate the mapping and variant calling quality. The candidate variants were then classified into five main categories using the American College of Medical Genetics and Genomics guidelines - pathogenic, likely pathogenic, variants of uncertain clinical significance (VUS), likely benign, and benign10. Sanger sequencing was performed to validate those variants categorized as pathogenic, likely pathogenic or VUS. Segregation was analyzed when parental samples were available. Results Clinical profiles and the diagnostic yield of prior single-gene tests in the traditional approach We recruited 32 patients from the pediatric endocrinology clinic, including 27 patients with a 46,XY karyotype and five patients with a 46,XX karyotype. As listed in Table 2, the patients presented with a variety of DSD phenotypes. Among the 27 patients with 46,XY DSD, 20 patients presented primarily with micropenis and small testes, four patients with hypospadias, two patients with cryptorchidism, one patient with ambiguous genitalia, and one patient with 46,XY complete gonadal dysgenesis. Among the five patients with a 46,XX karyotype, the majority (4/5) of the patients presented with ambiguous genitalia. Anosmia was reported in three patients. Table 2: Clinical phenotype, karyotype, prior genetic tests and findings in the current study. Full size table Sanger sequencing of single genes was ordered depending on the clinician’s judgment on the most likely genetic cause, following endocrine and radiological workup. The results of these prior genetic tests reflected the molecular diagnostic rate following traditional routines in our clinical practice (listed in Table 2 column “prior tests”). A total of 41 single gene tests by Sanger sequencing were requested for 30 patients, and pathogenic or likely pathogenic variants were identified in three patients (i.e., P01, P27, P32), obtaining a diagnostic rate of 10% (3/30). Both NR0B1 and KAL1 were the most frequently requested genetic tests, and all three patients with a likely diagnosis by single gene tests indeed harbored mutations in those genes. In this current study of NGS, information obtained from prior genetic tests was masked during subject selection and subsequent data analysis. Quality control, diagnostic yield and genetic variants identified by NGS Key quality metrics were assessed to validate the performance of targeted panel sequencing. In this analysis, 98.6% of the reads were aligned to the reference genome, with an average sequencing depth of 185X on targeted regions. In addition, 94.6% of the targeted regions were covered by ten or more reads (please see Supplementary Table 1). All variants of interest were subsequently validated by Sanger sequencing. Variants suspected to relate to the patient’s condition are listed in Table 3. A likely molecular diagnosis - with pathogenic or likely pathogenic variants identified - was obtained from 9 of the 32 patients (28%). These include three patients (i.e., P09, P22, P28) with 5-alpha-reductase deficiency due to biallelic mutations of the SRD5A2 gene, and four patients with idiopathic hypogonadotropic hypogonadism due to mutations of KAL1 (i.e., P27, P32), PROKR2 (P03) or GNRHR (P13). In addition, one patient with hypogonadotropic hypogonadism and adrenal insufficiency (later treated by hydrocortisone) was found to harbor a hemizygous mutation of the NR0B1 gene (P01), and one patient with androgen insensitivity syndrome was found to harbor a hemizygous mutation of the AR gene (P21). Therefore, the diagnostic yield of NGS (28%) in our study was almost three-fold that of the prior single gene tests (10%) as requested by clinicians in practice. Table 3: Details of genetic variants detected in DSD patients. Full size table Discussion Increased diagnostic yield of sequencing by the DSD panel as compared with single genes Panel sequencing confirmed all three cases with a previous genetic diagnosis (3/3, 100%) and identified the genetic causes in six of the 29 patients that received negative results by prior analyses (6/29, 20.7%), reaching an almost three-fold increase in yield over the traditional single gene approach in our center’s practice (28.1% versus 10%). The detection rate of single gene test is largely influenced by the selection of target gene, which is limited by the clinicians’ knowledge and experience with each genetic condition. For disorders like DSD, the phenotypes do not differentiate well among different conditions, thus the selection of target gene could be subjective. Even though, the detection rate reflects the possible outcome of traditional approach in one center’s clinical practice. For this given mixed group of patients, NGS excels the single gene test in DSD diagnosis. The diagnostic rate of NGS in our study is similar to the two studies by the Vilain lab (28% and 22.5%)6,7. Compared with the targeted panel by the Vilain lab, we included some genes known to be essential regulators of sex development but only few variants have been reported in human samples, like WDR11, BMP4 and SEMA3A, and eventually identified novel variants of these genes in patients who remain undiagnosed (Table 3b). In a more recent study by the Liang lab in the Chinese population, the targeted panel included 219 genes. Excluding two patients with large copy number variations, likely genetic causes were identified in 6 out of 21 patients (~28.6%)8, which is close to our study. Obtaining a diagnosis largely depends on variants identification in the well-established DSD genes - those genes with multiple lines of evidence to support their roles in pathogenesis. These well-established genes were basically covered by all four studies, including Vilain’s (two studies)6,7, Liang’s8 and our study. These four studies shared a core set of established genes and eventually reached a similar diagnostic rate (i.e., Vilain lab:28% & 22.5%, Liang lab: 28.6%, our group:28.1%). This also suggests 28% might be a possible limit of diagnostic rate based on currently established genes in human DSD, and further understanding of DSD etiology may expand the pool of candidate genes and eventually increase the diagnostic yield. Compared with the other three studies, our study provides the direct comparison of diagnostic rate between single gene tests following traditional approach (endocrine and radiological workup) and next-generation sequencing. The advantage of next-generation sequencing shown in this study supports a shift of paradigm in the clinical practice, based on a single center’s experience. In a further analysis, we attempted to identify the overlapped target genes among those previously published studies6,7,8 and this current report. By deriving data that was obtained from these three studies (Vilain’s6, Liang’s8 and our study), we employed the website http://bioinformatics.psb.ugent.be/cgi-bin/liste/Venn to construct a Venn diagram (Please refer to Supplementary Figure S1) of these overlapped target genes and found 22 genes that were common between all three studies. In addition, we found 22 plus 34 overlapped genes when comparing our study and that study previously reported by Liang’s group8. Further, all of the detected mutations in our study (Table 3) were located in the 22 plus 34 overlapped gene group (please see Supplementary Figure S1). Most of these overlapped genes are generally agreed disease-causing genes in DSD. The observation that detected variants located in this group suggested that it would be suitable to select only the identified 22 plus 34 gene sets as an initial “quick and small” screening panel, for patients whose main phenotypes are micropenis, small testes, cryptorchidism or ambiguous genitalia. This panel could be continuously optimized with more NGS data collected and a broader mutation spectrum revealed. We recommend this approach over the previously applied approach of attempting to analyze a large number of target genes such as the 80 gene set in our study. Clearly, this would afford a significant technical work effort and financial cost saving usually invested in the diagnosis of DSD. It should be mentioned that copy number variations were not examined in this study. A more comprehensive genetic panel including copy number variations may reveal another 13–22% of disease etiology, based on previous reports of chromosome microarray analysis11,12. Frequent genetic causes in Chinese DSD patients The mutational spectrum of DSD genes has not been well characterized in a Chinese population. For those patients presenting under-virilization, the genetic tests most often prescribed in our clinic are NR0B1 and KAL1. This reflects the “hot” candidate genes in clinicians’ opinion, which could be biased by their previous experience with those genetic conditions. Our results revealed SRD5A2 variants were the most frequent in the study group - with five variants identified in 3 patients. This is followed by KAL1 with 2 variants identified. NR0B1, GNRHR, AR, and PROKR2 variants were identified in one patient each. 5-alpha-reductase deficiency appears to be the most common defect, though not always the top possible etiology our clinicians initially consider before Dihydrotestosterone/Testosterone measurements. Future studies regarding the prevalence of each DSD condition would offer the clinicians a prioritized list of endocrine tests in the practice. Possible genotype-phenotype correlation of SRD5A2 mutations An interesting observation was that the severity of undervirilization appears to correlate with the genotype of SRD5A2 in our three 46, XY DSD patients. Frameshift or nonsense variants usually exert more detrimental effects on the enzymatic function than missense variants. P09 harbors two missense variants (p.Arg227Gln and p.Gly203Ser) and presents a relatively mild phenotype - micropenis and small testes. The variant p.Arg227Gln is reported to retain ~3.2% enzyme activity and associate with micropenis in compound heterozygous state13. The phenotype of P09 is consistent with literature. However, P28, who harbors this missense variant (p.Arg227Gln) and a frameshift variant (p.Phe219fs), presents a moderate degree of undervirilization - hypospadias in addition to micropenis. P22, who harbors a homozygous nonsense variant (p.Gln6*), exhibits the most severe phenotype of these three patients - presenting a female appearance and having ambiguous genitalia with testis on the right and a vagina. This might hint an increasing degree of undervirilization correlates with the more deleterious variants of SRD5A2, though enzymatic function assay of steroid 5-alpha-reductase and larger sample numbers are needed to validate this correlation. Nonetheless, this example clearly shows the broad range of phenotypic variations exerted by mutations of the same DSD gene. Novel variants identified in DSD genes Genes involved in the fine-tuned process of sex determination and differentiation have been studied in many model organisms, while not all the genes have been reported with human mutations. Inclusion of these genes in the panel sequencing may reveal variants in those lesser-known DSD genes. DMRT1 is a gene exclusively expressed in male gonads14. This gene was originally identified in chromosome 9p, the critical region of which was lost in 9p monosomy patients with defective testis development and 46,XY complete gonadal dysgenesis15. So far, very few variants have been reported in this gene. Murphy et al. reported the first point mutation of DMRT1 (p.R111G) in a 46,XY complete gonadal dysgenesis patient, and showed the heterozygous mutant affects DNA binding affinity and results in a severe phenotype16. Similar to the patient with p.R111G mutation, our patient P19, who has a 46, XY karyotype with normal SRY, harbors a novel missense variant c.251 A > G (p.Try84Cys) and exhibits complete gonadal dysgenesis. This heterozygous variant was transmitted from the unaffected mother. As DMRT1 is only expressed in male-specific gonads, it is possible that the defects of DMRT1 do not affect ovarian development, thus female carriers are normal and fertile. WDR11 is considered to associate with hypogonadotropic hypogonadism with or without anosmia (OMIM#614858). Five missense variants were reported so far, all from one study17. Our patient P06 harbors a novel missense variant c.817C > G (p.Leu273Val, predicted to be deleterious) and presents the phenotype of idiopathic hypogonadotropic hypogonadism with small testes and normosmia. This is consistent with the phenotypes in literature17. However, unlike the missense variants previously reported, this variant is not located in the region of WD-domain, which is important for protein-protein interaction. Further evidence is needed to verify if the variant indeed affects the production of WDR11 or protein interaction in pubertal development. BMP4 is a member of the bone morphogenetic protein (BMP) family that plays a vital role in embryonic development. There were 4 missense variants of BMP4 reported in 46,XY DSD patients with hypospadias18, and BMP4 is thought to regulate the signaling cascades involved in urethral development. A missense variant c.806G > A (p.Arg269Gln) was detected in our patient P07, with a 46, XX karyotype. P07 has a male appearance with hypospadias and inguinal testes. The BMP4 association with 46, XX testicular DSD has not been reported before. It is possible that the BMP4 variant affects the urethral formation in our patient, and another unidentified genetic variant causes the sex reversal phenotype. Based on current evidence, the clinical significance of the BMP4 variant identified in our patient cannot be determined. Variations of SEMA3A and AKR1C4 are both implicated in DSD, but definite roles are not established. SEMA3A variants were reported in patients of hypogonadotropic hypogonadism (OMIM#614897). Based on the normal phenotype of heterozygous mutant mice, mono-allelic mutation of SEMA3A may not be sufficient to cause sex differentiation problems without synergistic effect of another mutant allele of DSD genes19. The heterozygous AKR1C4 variant has been previously reported in a Swiss family of 46,XY complete gonadal dysgenesis20. We found a missense variant of SEMA3A (c.487G > C, p.Glu163Gln) and a missense variant of AKR1C4 (c.773G > A, p.Arg258His) both present in P23, who showed ambiguous genitalia including labioscrotal hypertrophy, micropenis and inguinal testes. This patient had a 46,XX karyotype. If SEMA3Aand AKR1C4 variants were responsible, they might exert a dominant effect that leads to masculinization – an observation that warrants further investigation. NR5A1 (also named SF1) encoding the steroidogenic factor 1 is not only important for the formation of bipotential gonads and sex determination, but also involved in the pathogenesis of hypogonadism in mice21. A novel missense variant of NR5A1 (c.34C > A, p.Leu12Met) was identified in patient P31, who has a normal male karyotype with the main phenotype of small testes. However, this variant was transmitted from the unaffected mother, who is free of primary ovarian failure (OMIM#184757). The clinical significance of this missense variant is uncertain. In conclusion, high-throughput sequencing markedly increased the diagnostic yield and showed great potential in the discovery of novel variants. 6 novel variants were identified in WDR11, BMP4, DMRT1, SEMA3A, AKR1C4 and NR5A1. Some of these genes are not yet generally accepted as disease-causing genes in DSD, and only few variants have been reported in human DSD patients. The identification of novel variants in DSD patients expands the pool of candidates in disease pathogenesis. For utilization in clinical diagnosis, we recommended a core set of 56 genes as initial screening panel based on ours and previous studies from other labs, which could provide sufficient diagnostic yield while save the technical effort and financial cost. Additional Information How to cite this article: Fan, Y. et al. Diagnostic Application of Targeted Next-Generation Sequencing of 80 Genes Associated with Disorders of Sexual Development. Sci. Rep. 7, 44536; doi: 10.1038/srep44536 (2017). Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References 1. Sperling, M. Pediatric endocrinology. Fourth edition. edn, 111–112 (Elsevier, 2014). Google Scholar2. Hughes, I. A., Houk, C., Ahmed, S. F. & Lee, P. A. Consensus statement on management of intersex disorders. Arch Dis Child 91, 554–563, doi: adc.2006.09831910.1136/adc.2006.098319 (2006). CASPubMedArticleGoogle Scholar3. Ono, M. & Harley, V. R. Disorders of sex development: new genes, new concepts. Nat Rev Endocrinol 9, 79–91, doi: nrendo.2012.23510.1038/nrendo.2012.235 (2013). CASPubMedArticleGoogle Scholar4. Shin, S. J. et al. Clinical, endocrinological, and molecular characterization of Kallmann syndrome and normosmic idiopathic hypogonadotropic hypogonadism: a single center experience. Annals of pediatric endocrinology & metabolism 20, 27–33, doi: 10.6065/apem.2015.20.1.27 (2015). PubMedArticleGoogle Scholar5. Baxter, R. M. & Vilain, E. Translational genetics for diagnosis of human disorders of sex development. Annu Rev Genomics Hum Genet 14, 371–392, doi: 10.1146/annurev-genom-091212-153417 (2013). CASPubMedArticleGoogle Scholar6. Arboleda, V. A. et al. Targeted massively parallel sequencing provides comprehensive genetic diagnosis for patients with disorders of sex development. Clinical genetics 83, 35–43, doi: 10.1111/j.1399-0004.2012.01879.x (2013). CASPubMedArticleGoogle Scholar7. Baxter, R. M. et al. Exome sequencing for the diagnosis of 46,XY disorders of sex development. J Clin Endocrinol Metab 100, E333–344, doi: 10.1210/jc.2014-2605 (2015). CASPubMedArticleGoogle Scholar8. Dong, Y. et al. Targeted next-generation sequencing identification of mutations in patients with disorders of sex development. BMC Med Genet 17, 23, doi: 10.1186/s12881-016-0286-210.1186/s12881-016-0286-2 (2016). CASPubMedArticleGoogle Scholar9. Hughes, I. A., Nihoul-Fekete, C., Thomas, B. & Cohen-Kettenis, P. T. Consequences of the ESPE/LWPES guidelines for diagnosis and treatment of disorders of sex development. Best Pract Res Clin Endocrinol Metab 21, 351–365, doi: S1521-690X(07)00055-310.1016/j.beem.2007.06.003 (2007). PubMedArticleGoogle Scholar10. Richards, S. et al. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genetics in medicine: official journal of the American College of Medical Genetics 17, 405–424, doi: 10.1038/gim.2015.30 (2015). PubMedArticleGoogle Scholar11. Tannour-Louet, M. et al. Identification of de novo copy number variants associated with human disorders of sexual development. PLoS One 5, e15392, doi: 10.1371/journal.pone.0015392 (2010). CASPubMedArticleGoogle Scholar12. White, S. et al. Copy number variation in patients with disorders of sex development due to 46,XY gonadal dysgenesis. PLoS One 6, e17793, doi: 10.1371/journal.pone.0017793 (2011). CASPubMedArticleGoogle Scholar13. Sasaki, G. et al. Micropenis and the 5α-Reductase-2 (SRD5A2) Gene: Mutation and V89L Polymorphism Analysis in 81 Japanese Patients. The Journal of Clinical Endocrinology and Metabolism 88, 3431–3436 (2011). CASArticleGoogle Scholar14. Moniot, B., Berta, P., Scherer, G., Sudbeck, P. & Poulat, F. Male specific expression suggests role of DMRT1 in human sex determination. Mech Dev 91, 323–325, doi: S0925-4773(99)00267-1 (2000). CASPubMedArticleGoogle Scholar15. Guioli, S. et al. Molecular analysis of 9p deletions associated with XY sex reversal: refining the localization of a sex-determining gene to the tip of the chromosome. Am J Hum Genet 63, 905–908, doi: S0002-9297(07)61397-410.1086/302017 (1998). CASPubMedArticleGoogle Scholar16. Murphy, M. W. et al. An ancient protein-DNA interaction underlying metazoan sex determination. Nature Structural & Molecular Biology 22, 442–451 (2015). CASPubMedArticleGoogle Scholar17. Kim, H. G. et al. WDR11, a WD protein that interacts with transcription factor EMX1, is mutated in idiopathic hypogonadotropic hypogonadism and Kallmann syndrome. Am J Hum Genet 87, 465–479, doi: S0002-9297(10)00470-210.1016/j.ajhg.2010.08.018 (2010). CASPubMedArticleGoogle Scholar18. Chen, T. et al. Mutation screening of BMP4, BMP7, HOXA4 and HOXB6 genes in Chinese patients with hypospadias. Eur J Hum Genet 15, 23–28, doi: 520172210.1038/sj.ejhg.5201722 (2007). CASPubMedArticleGoogle Scholar19. Hanchate, N. K. et al. SEMA3A, a gene involved in axonal pathfinding, is mutated in patients with Kallmann syndrome. PLoS Genet 8, e1002896, doi: 10.1371/journal.pgen.1002896PGENETICS-D-12-00582 (2012). CASPubMedArticleGoogle Scholar20. Fluck, C. E. et al. Why boys will be boys: two pathways of fetal testicular androgen biosynthesis are needed for male sexual differentiation. Am J Hum Genet 89, 201–218, doi: S0002-9297(11)00262-X10.1016/j.ajhg.2011.06.009 (2011). CASPubMedArticleGoogle Scholar21. Zhao, L. et al. Steroidogenic factor 1 (SF1) is essential for pituitary gonadotrope function. Development 128, 147–154 (2001). CASPubMedGoogle Scholar22. Moss, S. B. & Gerton, G. L. A-kinase anchor proteins in endocrine systems and reproduction. Trends Endocrinol Metab 12, 434–440, doi: S1043-2760(01)00493-3 (2001). CASPubMedArticleGoogle Scholar23. Rizner, T. L. & Penning, T. M. Role of aldo-keto reductase family 1 (AKR1) enzymes in human steroid metabolism. Steroids 79, 49–63, doi: S0039-128X(13)00236-510.1016/j.steroids.2013.10.012 (2014). CASPubMedArticleGoogle Scholar24. Beleza-Meireles, A. et al. Activating transcription factor 3: a hormone responsive gene in the etiology of hypospadias. Eur J Endocrinol 158, 729–739, doi: 158/5/72910.1530/EJE-07-0793 (2008). CASPubMedArticleGoogle Scholar25. Ogino, Y. et al. Bmp7 and Lef1 are the downstream effectors of androgen signaling in androgen-induced sex characteristics development in medaka. Endocrinology 155, 449–462, doi: en.2013-150710.1210/en.2013-1507 (2014). CASPubMedArticleGoogle Scholar26. Lang, M. R., Patterson, L. B., Gordon, T. N., Johnson, S. L. & Parichy, D. M. Basonuclin-2 requirements for zebrafish adult pigment pattern development and female fertility. PLoS Genet 5, e1000744, doi: 10.1371/journal.pgen.1000744 (2009). CASPubMedArticleGoogle Scholar27. Kopanja, D. et al. Cul4A is essential for spermatogenesis and male fertility. Dev Biol 352, 278–287, doi: S0012-1606(11)00061-310.1016/j.ydbio.2011.01.028 (2011). CASPubMedArticleGoogle Scholar28. Nakamura, Y. et al. Human adrenal cells that express both 3beta-hydroxysteroid dehydrogenase type 2 (HSD3B2) and cytochrome b5 (CYB5A) contribute to adrenal androstenedione production. J Steroid Biochem Mol Biol 123, 122–126, doi: S0960-0760(10)00375-410.1016/j.jsbmb.2010.12.001 (2011). CASPubMedArticleGoogle Scholar29. Calatayud, N. E. et al. Ontogeny of the oestrogen receptors ESR1 and ESR2 during gonadal development in the tammar wallaby, Macropus eugenii. Reproduction 139, 599–611, doi: REP-09-030510.1530/REP-09-0305 (2010). CASPubMedArticleGoogle Scholar30. Bagheri-Fam, S. et al. Loss of Fgfr2 leads to partial XY sex reversal. Dev Biol 314, 71–83, doi: S0012-1606(07)01539-410.1016/j.ydbio.2007.11.010 (2008). CASPubMedArticleGoogle Scholar31. Geller, F. et al. Genome-wide association analyses identify variants in developmental genes associated with hypospadias. Nat Genet 46, 957–963, doi: ng.306310.1038/ng.3063 (2014). CASPubMedArticleGoogle Scholar32. Mansouri, M. R. et al. Molecular genetic analysis of a de novo balanced translocation t(6; 17) (p21.31; q11.2) associated with hypospadias and anorectal malformation. Hum Genet 119, 162–168, doi: 10.1007/s00439-005-0122-9 (2006). PubMedArticleGoogle Scholar33. Birk, O. S. et al. The LIM homeobox gene Lhx9 is essential for mouse gonad formation. Nature 403, 909–913, doi: 10.1038/35002622 (2000). CASPubMedArticleGoogle Scholar34. Baetens, D. et al. Extensive clinical, hormonal and genetic screening in a large consecutive series of 46,XY neonates and infants with atypical sexual development. Orphanet J Rare Dis 9, 209, doi: 10.1186/s13023-014-0209-2s13023-014-0209-2 (2014). PubMedArticleGoogle Scholar35. Quintero-Rivera, F. et al. Is the disruption of an N-myristoyltransferase (NMT2) associated with hypoplastic testes? Am J Med Genet A 143A, 1796–1798, doi: 10.1002/ajmg.a.31799 (2007). Google Scholar36. Wilhelm, D., Palmer, S. & Koopman, P. Sex determination and gonadal development in mammals. Physiol Rev 87, 1–28, doi: 87/1/110.1152/physrev.00009.2006 (2007). CASPubMedArticleGoogle Scholar37. Polanco, J. C., Wilhelm, D., Davidson, T. L., Knight, D. & Koopman, P. Sox10 gain-of-function causes XX sex reversal in mice: implications for human 22q-linked disorders of sex development. Hum Mol Genet 19, 506–516, doi: ddp52010.1093/hmg/ddp520 (2010). CASPubMedArticleGoogle Scholar38. Tziaferi, V., Kelberman, D. & Dattani, M. T. The role of SOX2 in hypogonadotropic hypogonadism. Sex Dev 2, 194–199, doi: 00015203510.1159/000152035 (2008). CASPubMedArticleGoogle Scholar39. Pillai, R. S. & Chuma, S. piRNAs and their involvement in male germline development in mice. Dev Growth Differ 54, 78–92, doi: 10.1111/j.1440-169X.2011.01320.x (2012). CASPubMedArticleGoogle Scholar40. Balasubramanian, R. et al. Expanding the phenotypic spectrum and variability of endocrine abnormalities associated with TUBB3 E410K syndrome. J Clin Endocrinol Metab 100, E473–477, doi: 10.1210/jc.2014-4107 (2015). CASPubMedArticleGoogle Scholar41. White, S. et al. A multi-exon deletion within WWOX is associated with a 46,XY disorder of sex development. Eur J Hum Genet 20, 348–351, doi: ejhg201120410.1038/ejhg.2011.204 (2012). CASPubMedArticleGoogle Scholar Download references Acknowledgements The authors thank all the patients and their families. We are also grateful for the revision and comments from Dr. Andrew Dauber at Cincinnati Children’s Hospital Medical Center. Funding for this project was from the Shanghai Municipal Science and Technology Commission (No. 15YF1409600, to Y.F.), the Shanghai Municipal Education Commission (No. 15CG14, to Y.F.), the National Natural Science Foundation of China (No. 81500972, to Y.F.; No. 81670812, to Y.G.Y.), and the Shanghai Jiao Tong University School of Medicine (No. 15ZH3003, to Y.G.Y.). Author information AffiliationsDepartment of Pediatric Endocrinology/Genetics, Xinhua Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai Institute for Pediatric Research, Shanghai, 200092, ChinaYanjie Fan, Xia Zhang, Lili Wang, Ruifang Wang, Zhuo Huang, Yu Sun, Jun Ye, Lianshu Han, Wenjuan Qiu, Huiwen Zhang, Lili Liang, Xuefan Gu & Yongguo YuDepartment of Endocrinology, Shanghai Children’s Medical Center, Shanghai Jiao Tong University School of Medicine, Shanghai, 200127, ChinaRuen Yao & Xiaodong Huang AuthorsSearch for Yanjie Fan in:Nature Research journals • PubMed • Google ScholarSearch for Xia Zhang in:Nature Research journals • PubMed • Google ScholarSearch for Lili Wang in:Nature Research journals • PubMed • Google ScholarSearch for Ruifang Wang in:Nature Research journals • PubMed • Google ScholarSearch for Zhuo Huang in:Nature Research journals • PubMed • Google ScholarSearch for Yu Sun in:Nature Research journals • PubMed • Google ScholarSearch for Ruen Yao in:Nature Research journals • PubMed • Google ScholarSearch for Xiaodong Huang in:Nature Research journals • PubMed • Google ScholarSearch for Jun Ye in:Nature Research journals • PubMed • Google ScholarSearch for Lianshu Han in:Nature Research journals • PubMed • Google ScholarSearch for Wenjuan Qiu in:Nature Research journals • PubMed • Google ScholarSearch for Huiwen Zhang in:Nature Research journals • PubMed • Google ScholarSearch for Lili Liang in:Nature Research journals • PubMed • Google ScholarSearch for Xuefan Gu in:Nature Research journals • PubMed • Google ScholarSearch for Yongguo Yu in:Nature Research journals • PubMed • Google Scholar Contributions X.G. and Y.Y. designed and supervised the study. Y.F. performed majority of the data analysis, wrote the main manuscript and prepared the figures. X.Z., L.W., R.W., Z.H. and R.Y. performed the experiments. Y.S. participated in the data analysis. X.H., J.Y., L.H., W.Q., H.Z., L.L., X.G. and Y.Y. recruited the patients. All authors reviewed the manuscript. Competing interests The authors declare no competing financial interests. Corresponding authors Correspondence to Xuefan Gu or Yongguo Yu. Supplementary information PDF files1.Supplementary Information Rights and permissions This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ About this article Publication history Received 26 August 2016 Accepted 10 February 2017 Published 15 March 2017 DOI https://doi.org/10.1038/srep44536


This is a preview of a remote PDF: https://www.nature.com/articles/srep44536.pdf

Yanjie Fan, Xia Zhang, Lili Wang, Ruifang Wang, Zhuo Huang, Yu Sun, Ruen Yao, Xiaodong Huang, Jun Ye, Lianshu Han, Wenjuan Qiu, Huiwen Zhang, Lili Liang, Xuefan Gu, Yongguo Yu. Diagnostic Application of Targeted Next-Generation Sequencing of 80 Genes Associated with Disorders of Sexual Development, Scientific Reports, 2017, DOI: 10.1038/srep44536