Submicroscopic copy-number variations associated with 46,XY disorders of sex development

Background Mutations in known causative genes and cytogenetically detectable chromosomal rearrangements account for a fraction of cases with 46,XY disorders of sex development (DSD). Recent advances in molecular cytogenetic technologies, including array-based comparative genomic hybridization (aCGH) and multiplex ligation-dependent probe amplification (MLPA), have enabled the identification of copy-number variations (CNVs) in individuals with apparently normal karyotypes. Findings This review paper summarizes the results of 15 recent studies, in which aCGH or MLPA were used to identify CNVs. Several submicroscopic CNVs have been detected in patients with 46,XY DSD. These CNVs included deletions involving known causative genes such as DMRT1 or NR5A1, duplications involving NR0B1, deletions involving putative cis-regulatory elements of SOX9, and various deletions and duplications of unknown pathogenicity. Conclusions The results of recent studies highlight the significance of submicroscopic CNVs as the genetic basis of 46,XY DSD. Molecular cytogenetic analyses should be included in the diagnostic workup of patients with 46,XY DSD of unknown origin. Further studies using aCGH will serve to clarify novel causes of this condition.

Recent advances in molecular technology, including array-based comparative genomic hybridization (aCGH), multiplex ligation-dependent probe amplification (MLPA), and next-generation sequencing (NGS), have enabled high-throughput analysis of clinical samples. Of these, aCGH is highly useful to detect copy-number variations (CNVs) in the genome of individuals with apparently normal karyotypes [5,6], and MLPA can identify various copy-number alterations in specific disease-associated loci [7]. NGS primarily focuses on identification of nucleotide substitutions. Molecular cytogenetic analyses using aCGH or MLPA revealed the importance of CNVs as the cause of several genetic disorders, although accumulating evidence shows that submicroscopic CNVs can also occur as functionally neutral polymorphisms [6]. Here, we review recent reports on molecular cytogenetic analyses of patients with 46,XY DSD.

46,XY DSD-associated CNVs identified by molecular cytogenetic analyses
In this review, we summarize the results of 15 recent studies [8][9][10][11][12][13][14][15][16][17][18][19][20][21][22]. We found these papers through a PubMed search using the key words 'disorders of sex development', together with 'comparative genomic hybridization', 'multiplex ligation-dependent probe amplification', or 'copynumber variations'. We focused on original articles, in which aCGH or MLPA was used to identify submicroscopic CNVs in patients with 46,XY DSD. The 15 studies showed 28 deletions and 4 duplications as genetic causes of 46,XY DSD, as well as several other CNVs whose association with the disease phenotype remains uncertain (Table 1, parts a and b, and  [9]. These data suggest the significance of submicroscopic CNVs in the etiology of 46,XY DSD. All CNVs, except for those on the sex chromosomes, were detected in the heterozygous state. Parental samples of the CNV-positive patients were analyzed in some cases, confirming the de novo occurrence or maternal inheritance of all CNVs examined (Table 1, parts a and b).
Deletions encompassing known 46,XY DSD-causative genes Submicroscopic deletions encompassing known causative genes were identified in 18 patients (Table 1, part a). These deletions ranged from 74 kb to 18.0 Mb and caused haploinsufficiency of LHCGR, DMRT1, NR5A1, WT1, or HOXD cluster [8][9][10][11][12][13][14][15][16][17]. Of these deletions, those encompassing DMRT1 or NR5A1 were relatively common. Notably, CNVs in cases 3, 16, and 17 were 10 to 18 Mb in length. These results imply that in some cases, deletions ≥10 Mb can be missed by standard karyotyping, although this method is expected to detect CNVs ≥5 to 10 Mb [11]. Actually, the results of karyotyping are affected by the quality of samples and the genomic position of CNVs [11]. Submicroscopic deletions in the 18 patients were associated with both isolated and syndromic 46,XY DSD. Syndromic DSD resulting from these deletions were primarily ascribed to contiguous gene deletions. For example, a deletion in case 16 encompassing WT1 and PAX6 caused 46,XY DSD, Wilms tumor, aniridia, and mental retardation, which is collectively referred to as WAGR syndrome. This WAGR syndrome has been described in several patients with cytogenetically detectable deletions at 11p. The size of submicroscopic deletions in 18 patients roughly corresponded to patients' phenotypes (isolated or syndromic); 7 of the 9 patients with deletions ≥1.0 Mb manifested additional clinical features such as mental retardation, short stature, and facial dysmorphism, whereas such features were not reported in 9 patients with smaller deletions (Table 1, part a). On the other hand, syndromic 46,XY DSD was also caused by deletions of a single gene with complex functions. Indeed, finger anomalies observed in DSD patients with HOXD cluster-containing deletions are consistent with the fact that HOXD genes, particularly HOXD13 [23], control the development of limb and external genitalia [4]. Although submicroscopic deletions involving known causative genes were frequently identified in patients with syndromic 46,XY DSD (Table 1, part a), previous molecular cytogenetic analyses may be biased toward patients with complex phenotypes. Thus, further studies are necessary to clarify the precise frequency of pathogenic CNVs in patients with isolated and syndromic 46,XY DSD.

Deletions in the upstream regions of known 46,XY DSD-causative genes
Submicroscopic deletions that reside adjacent to known causative genes also underlie 46,XY DSD. To date, deletions involving the upstream regions of SOX9, GATA4, or NR0B1 have been reported to cause 46,XY gonadal dysgenesis ( Table 1, part b) [10,12,[18][19][20][21]. These deletions are predicted to disrupt the cis-regulatory machinery of genes involved in testis formation. For example, SOX9 is regulated by SRY and plays a critical role in the development of testis and long bones [1]. SOX9 abnormalities lead to 46,XY gonadal dysgenesis with or without campomelic dysplasia [1]. It is known that SOX9 expression is tightly regulated by multiple cis-acting enhancers in the upstream and downstream regions and that elimination of the enhancer(s) leads to tissue-specific dysregulation of SOX9 [10,[18][19][20]. Previous studies have mapped SOX9 enhancers for craniofacial tissues to a genomic interval >1.0 Mb apart from the coding region and a testis enhancer to a 32.5-kb region at a position 607 to 640 kb upstream from the start codon [10,[18][19][20]. These enhancer regions seem to contain transcription factor binding sites which are essential to maintain SOX9 expression during development. Since microdeletions in the SOX9 upstream region can result in complete gonadal dysgenesis similar to that observed in patients with SOX9 amorphic mutations, elimination of the distal enhancer seems to completely abolish SOX9 expression on the affected allele [10,[18][19][20]. Deletions in the upstream regions of GATA4 and NR0B1 are also likely to encompass cisacting enhancers of these genes [12,18,21]. Identification of deletions in the upstream or downstream regions of genes provides clues regarding cis-regulatory mechanisms of each gene.
NR0B1 is a transcription factor gene isolated from the dosage-sensitive sex reversal adrenal hypoplasia congenita critical region on Xp21 [24]. Overdosage of NR0B1 is a well-known cause of syndromic 46,XY DSD in individuals with large X chromosomal rearrangements [25]. Recently, molecular cytogenetic analysis by Smyk et al. has shown that submicroscopic duplications encompassing NR0B1 can lead to 46,XY DSD as a sole clinical manifestation [21]. Copy-number gains of NR0B1 were predicted to perturb testicular development by downregulating the protein expression of SF1, WT1, and SOX9 [18].

CNVs of unknown pathogenicity
Genome-wide copy-number analyses identified several submicroscopic CNVs that could be associated with the development of 46,XY DSD. Ledig

Clinical applications of molecular cytogenetic technologies
Molecular cytogenetic analyses using aCGH or MLPA would be beneficial for patients with 46,XY DSD, because identification of pathogenic CNVs could help to predict the disease outcome and possible complications of patients. For example, patients with deletions encompassing WT1 have a high risk of Wilms tumor and renal failure. Furthermore, detection of disease-associated CNVs significantly improves the accuracy of genetic counseling for patients' families. Molecular cytogenetic analyses, together with mutation screening using NGS, should be included in the diagnostic workup of patients with 46,XY DSD of unknown origin.

Summary and conclusions
Recent studies revealed that submicroscopic CNVs constitute a fraction of the genetic causes of both isolated and syndromic 46,XY DSD. ACGH and MLPA appear to be useful for molecular diagnosis of patients with 46,XY DSD. Furthermore, genome-wide copy-number analyses using aCGH will serve to identify novel causes of 46,XY DSD.