Background
Adolescent primary amenorrhea is a frequent reason for consultation in pediatric and gynecological endocrine clinics. Primary amenorrhea may result from congenital abnormalities in gonadal or genital tract development or from a defect in the hypothalamic-pituitary-ovarian axis. Failure to menstruate by the age of 15 years requires investigation to determine the cause and establish a treatment plan. The standard investigation includes detailed clinical evaluation [
1], endocrine assessment (gonadotropins, testosterone, AMH and inhibin B assays) and pelvic imaging. In addition, genetic exploration is crucial to classify the primary amenorrhea. Karyotyping, which discriminates normal from abnormal chromosomes (i.e., 45,X0 or 46,XY), is the first step. In the 46,XY disorders of sex differentiation (DSD) [
2], primary amenorrhea may be caused by a genetic defect in fetal testis determination, failure of the fetal testis to produce testosterone, or androgen resistance [
3]. The assessment of endocrine parameters thus often orients the exploration toward the most probable genetic cause [
4]. For example, when plasma testosterone (pl-T) is low, an abnormality in the genes involved in fetal testis determination, such as
SRY,
SF1,
WT1,
SOX9, DMRT, DHH, DAX1 and
WNT4, should be considered [
5].
Here we describe a two-year experience of genetic exploration in adolescents with primary amenorrhea due to 46,XY DSD and low pl-T concentration. We specifically focused on SRY, SF1 and WT1 because these genes have previously been reported to be implicated in adolescents presenting with primary amenorrhea due to 46,XY DSD in association with low pl-T, but no other signs. Our aim was thus to assess the frequency of mutations in these genes in a cohort of 15 adolescents with this profile. We identified eight unreported mutations in these genes responsible for testis differentiation and development: two new mutations in SRY and five new mutations in SF1. Moreover, in a patient with a specific biological profile of elevated LH and normal FSH concentrations, we identified a new LH receptor mutation. Thus far, we have been unable to determine a genetic cause for the primary amenorrhea in the seven remaining subjects.
Discussion
Primary amenorrhea is a frequent reason for consultation in pediatric endocrine and gynecology clinics. Three types of primary amenorrhea can be observed: (1) 46,XX amenorrhea with high FSH concentration, which usually corresponds to a reduction in the number of primary follicles, accelerated follicular atresia, or follicular dysfunction; (2) 45,X0 amenorrhea associated with Turner syndrome; and (3) amenorrhea due to 46,XY DSD. This last can be further classified into two subtypes. The first is defined by primary amenorrhea with normal or high testosterone concentration, suggesting androgen insensitivity syndrome or possibly 5-alpha reductase type 2 deficiency. The second is defined by primary amenorrhea with low testosterone concentration. The etiologies are multiple and testis determination genes such as SRY, SF1 or WT1 should be explored.
In our 31 adolescents with primary amenorrhea due to 46,XY DSD, 16 showed high testosterone concentrations and, as expected,
AR gene mutation was a major cause among those who presented with breast development (B5) contrasting with an absence of pubic hair (P1) (data not presented). For the group with low testosterone concentration and considered as 46,XY gonadal dysgenesis [
13], we systematically analyzed
SRY,
SF1 and
WT1 genes, beginning with
SRY. Gene analysis identified two new
SRY mutations (2/15). This frequency (13.3%) is similar to that usually reported in the literature, which varies from 10 to 15% [
14], and suggests that our cohort was representative of the population of adolescents with primary amenorrhea due to 46,XY DSD and low pl-T.
Patients without SRY gene mutations were analyzed for defects in WT1 or SF1. WT1 is normally associated with Frasier or Denys-Drash syndrome. In our experience with this gene analysis, we have identified a mutation three times in patients with no sign of kidney abnormality or cancer (unpublished). However, no cases of WT1 abnormality were noted in the present study group.
In contrast, we identified five new SF1 mutations in this cohort (5/15), which amounts to one third of the cases of primary amenorrhea due to 46,XY DSD with low pl-T level and 5/31 of the cases of primary amenorrhea due to 46,XY DSD. Among these new mutations, three nucleotide substitutions, one insertion and one deletion were identified.
Although
in vitro studies are needed to demonstrate the implication of these mutations, two of our
SF1 mutations and the
SRY insertion certainly abolished activity since an insertion or deletion creates a frame shift and premature stop codon. The p.M1V
SF1 mutant probably abolished the transcriptional initiation. An alternative initiation codon is located downstream at codon 78, after the DNA-binding domain, and probably altered the SF1 function. A similar mutation in the initiation codon (p.Met1Ile) was reported in a girl with hypertrophic clitoris, which confirms the impact of this abnormality [
15]. These three cases are similar to the cases of haploinsufficiency reported by our group and others [
16‐
18] in patients with complete sex reversal. For the two other
SF1 missense mutations and the
SRY substitution reported here, the functional effects predicted by the two types of software were concordant, with the same conclusion of affected protein mutants. These predictions agreed with the phenotype and hormonal data observed in our patients, and we can conclude that these mutations were probably the cause of the phenotype and the biological abnormalities.
We analyzed the LH receptor gene in one of the girls of our cohort because of high LH contrasting with normal FSH concentration [
19]. We were thus able to identify a previously unreported
LHCGR mutation leading to an inactive truncated LH receptor.
For the seven primary amenorrhea adolescents with low pl-T concentrations and no mutation in any of the studied genes, it is probable that one of the other genes implicated in DSD was mutated. The absence of associated signs was a supplementary difficulty for the orientation of genetic exploration, but further gene analysis is warranted to identify the etiology of the DSD observed in these patients. Any one of several genes could be involved, such as SOX9, DMRT, or DHH, as well as duplications of DAX1 or WNT4. However, these genes are not often found to be mutated, especially without specific associated signs.
Conclusions
To conclude, the genetic analysis of low-testosterone primary amenorrhea due to 46,XY DSD is complex since several factors may be involved, including SRY, SF1, WT1 and LH receptor. We confirmed that SRY was mutated in about 10 to 15% of the cases. More interestingly, we identified new SF1 mutations in five of our 15 patients. As this amounts to one third of our cohort, we suggest that the SF1 sequence should be systematically analyzed in girls with primary amenorrhea due to 46,XY DSD and low testosterone concentration, as well as in newborns with 46,XY DSD.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
PP participated in the design of the study and the molecular genetic analyses, carried out the sequence alignment and the in silico analyses, and drafted the manuscript. El collected the clinical data and participated in patients follow-up. DZ, ET, MP, AMF, JL and IR all managed patients. NS carried out a part of the molecular genetic study. FA carried out the other part of the molecular genetic study. FP managed patients and participated in patients follow-up. CS conceived the study, participated in its coordination, and helped to draft the manuscript. All authors read and approved the final manuscript.