Mutations in NR5A1 associated with ovarian insufficiency

Abstract

Background: The genetic causes of nonsyndromic ovarian insufficiency are largely unknown. A nuclear receptor, NR5A1 (also called steroidogenic factor 1), is a key transcriptional regulator of genes involved in the hypothalamic-pituitary-steroidogenic axis. Mutation of NR5A1 causes 46,XY disorders of sex development, with or without adrenal failure, but growing experimental evidence from studies in mice suggests a key role for this factor in ovarian development and function as well.

Methods: To test the hypothesis that mutations in NR5A1 cause disorders of ovarian development and function, we sequenced NR5A1 in four families with histories of both 46,XY disorders of sex development and 46,XX primary ovarian insufficiency and in 25 subjects with sporadic ovarian insufficiency. None of the affected subjects had clinical signs of adrenal insufficiency.

Results: Members of each of the four families and 2 of the 25 subjects with isolated ovarian insufficiency carried mutations in the NR5A1 gene. In-frame deletions and frameshift and missense mutations were detected. Functional studies indicated that these mutations substantially impaired NR5A1 transactivational activity. Mutations were associated with a range of ovarian anomalies, including 46,XX gonadal dysgenesis and 46,XX primary ovarian insufficiency. We did not observe these mutations in more than 700 control alleles.

Conclusions: NR5A1 mutations are associated with 46,XX primary ovarian insufficiency and 46,XY disorders of sex development.

Figure 1. Pedigrees of Four Families with 46,XY Disorders of Sex Development or 46,XX Ovarian Insufficiency and Histologic Analysis of Samples from Two Affected Subjects in Family 2

In Panel A, squares represent male family members and circles represent female family members. Solid squares represent affected 46,XY subjects who were raised as boys, and solid circles represent affected 46,XX subjects. Squares containing solid circles represent affected 46,XY subjects who were raised as girls. Symbols containing a black dot represent apparently unaffected carriers of the mutation. The triangle in Family 1 represents miscarriage, and the symbol with a slash represents a deceased twin. Numbers within symbols indicate multiple siblings. The index patient is indicated with an arrow in each family. Genotyping information is provided for Family 2. The genotypes of the parents of the proband are inferred, whereas all others have been determined by molecular analysis. Panel B shows gonadal histologic analysis of affected members of Family 2. Subpanels a, b, and c show dysgenetic pubertal testis with Leydig-cell hyperplasia and aplasia of germ cells in the proband (Subject III-11), who had been raised as a girl. Subpanel a shows cells from the right testis, which are shown under higher magnification in subpanel b; cells from the left testis are shown in subpanel c. Subpanel d shows a sample from the proband’s sister (Subject III-8), who had 46,XX primary amenorrhea, showing a dysgenetic gonad with fibrovascular tissue and without germ cells.

Figure 2. Distribution of NR5A1 Mutations in Relation to the Protein

The functional domains of the NR5A1 protein are shown, indicating the position of known NR5A1 mutations, including the six mutations described in this study. Representative chromatograms are shown with each mutation. The DNA-binding domain (DBD) containing two zinc-finger motifs is indicated. The FtzF1 box stabilizes protein binding to DNA. The hinge region is important for stabilizing the ligand-binding domain and interacts with other proteins that control NR5A1 transcriptional activity. The AF2 domain recruits cofactors necessary for NR5A1 transactivating activity.

Figure 3. Comparison of Sequence Alignment of Portions of the Human NR5A1 Protein with Those of Other Mammals and 3-D Models of Wild-Type and Mutated NR5A1 Proteins

In Panel A, sequence alignment of the distal portion of the hinge and the ligand-binding domain (LBD) of human NR5A1 protein is compared with those of other mammals. The 12 predicted alpha helixes in the ligand-binding domain of NR5A1 are indicated as solid boxes, and amino acids in bold text. The position of the p.Asp293Asn missense mutations and the deletions of three amino acids (p.Leu231_Leu233) are highlighted. Both mutations fall either in the highly conserved Helix 1 (p.Leu231_Leu233) or in Helix 4 (p.Asp293Asn) of the ligand-binding domain. In Panel B, three-dimensional models of the ligand-binding domain of both the wild-type and p.Leu231_Leu233del mutated NR5A1 proteins were obtained with the use of the Web-based interface 3D-JIGSAW (for details, see the Supplementary Appendix). The hydrophobic helixes are shown in red, and the hydrophilic helixes in blue. Note the change of Helix 1 from hydrophobic to hydrophilic in the p.Leu231_Leu233 deletion mutant.

Figure 4. Assays of NR5A1 Transcriptional Activity

The transcriptional activity of wild-type NR5A1 (SF1) and variants associated with ovarian insufficiency was studied with the use of Cyp11a1 (Panel A) and Cyp19a1 (Panel B) promoters in tsa201 cells. NR5A1 activity is shown as the factor increase above baseline, which is defined as the activity observed in transfection with empty vector. Data represent the mean of three independent experiments, each performed in triplicate. The T bars represent standard errors.