Loss of anti-Müllerian hormone (AMH) immunoactivity due to a homozygous AMH gene variant rs10417628 in a woman with classical polycystic ovary syndrome (PCOS)

Abstract

Anti-Müllerian hormone (AMH) is produced by granulosa cells of pre-antral and small antral ovarian follicles. In polycystic ovary syndrome (PCOS), higher levels of serum AMH are usually encountered due to the ample presence of small antral follicles and a high AMH production per follicular unit which have led to the proposal of AMH as a serum diagnostic marker for PCOS or as a surrogate for polycystic ovarian morphology (PCOM). However, heterozygous coding mutations of the AMH gene with decreased in vitro bioactivity have been described in some women with PCOS. Such mutation carriers have a trend toward reduced serum AMH levels compared to noncarriers, although both types of women with PCOS have similar circulating gonadotropin and testosterone (T) levels. This report describes a normal-weight woman with PCOS by NIH criteria with severely reduced AMH levels (index woman with PCOS). Our objective was to examine the molecular basis for her reduced serum AMH levels and to compare her endocrine characteristics to similar-weight women with PCOS and detectable AMH levels. Twenty normoandrogenic ovulatory (control) and 13 age- and BMI-matched women with PCOS (19-35 years; 19-25 kg/m2) underwent transvaginal sonography and serum hormone measures including gonadotropins, sex hormone-binding globulin, total and free T, androstenedione, dehydroepiandrosterone sulfate, estrone, estradiol and AMH. The latter was measured by ELISA (Pico-AMH: Ansh Labs, Webster, TX, USA). Women with PCOS and detectable AMH had higher serum AMH (10.82 (6.74-13.40) ng/ml, median (interquartile range)), total and free T (total T: 55.5 (49.5-62.5) ng/dl; free T: 5.65 (4.75-6.6) pg/ml) levels and greater total antral follicle count (AFC) (46 (39-59) follicles) than controls (AMH: 4.03 (2.47-6.11) ng/ml; total T: 30 (24.5-34.5) ng/dl; free T: 2.2 (1.8-2.45) pg/ml; AFC 16 (14.5-21.5) follicles, P < 0.05, all values), along with a trend toward LH hypersecretion (P = 0.06). The index woman with PCOS had severely reduced serum AMH levels (∼0.1 ng/ml), although she also had a typical NIH-defined PCOS phenotype resembling that of the other women with PCOS and elevated AMH levels. All women with PCOS, including the index woman with PCOS, exhibited LH hypersecretion, hyperandrogenism, reduced serum estrogen/androgen ratios and PCOM. A homozygous Ala515Val variant (rs10417628) in the mature region of AMH was identified in the index woman with PCOS. Recombinant hAMH-515Val displayed normal processing and bioactivity, yet had severely reduced immunoactivity when measured by the commercial pico-AMH ELISA assay by Ansh Labs. In conclusion, homozygous AMH variant rs10417628 may severely impair serum AMH immunoactivity without affecting its bioactivity or PCOS phenotypic expression. Variants in AMH can interfere with serum AMH immunoactivity without affecting the phenotype in PCOS. This observation can be accompanied by discordance between AMH immunoactivity and bioactivity.

Keywords: anti-Müllerian hormone; gene mutations; hyperandrogenism; polycystic ovary syndrome; variants.

Figure 1.

DNA sequence analysis. (A) Sequence chromatograms of AMH of the woman with polycystic ovary syndrome (PCOS) and severely reduced serum anti-Müllerian hormone (AMH) levels and a control woman with PCOS. The arrow indicates the location of the mutation. (B) Nucleotide sequence alignment with predicted amino acid sequence corresponding to position g.7705 and flanking sequences.

Figure 2.

Western blot analysis and bioactivity of anti-Müllerian hormone (AMH) proteins AMH-⁵¹⁵Ala and AMH-⁵¹⁵Val. Western analysis of HEK293 cells expressing the hAMH variants without (RAQR) or with (RARR) the optimized cleavage site were analyzed. The mature region-specific 5/6A antibody recognizes the AMH precursor protein (75 kD) and the cleaved C-terminal mature protein (∼15 kD). The pro-region-specific 2/6A antibody recognizes the dimeric AMH precursor protein (150 kD) and the cleaved N-terminal pro-region (∼57 kD). The relative molecular masses (kD) of the protein marker are indicated on the left. No difference was observed in processing of the AMH variants. Cell lysates (A) Lane 1: hAMH-RAQR-⁵¹⁵Ala; lane 2: hAMH-RAQR-⁵¹⁵Val; lane 3: hAMH-RARR-⁵¹⁵Ala; lane 4: hAMH-RARR-⁵¹⁵Val. Supernatants (B) Lane 1: hAMH-RARR-⁵¹⁵Ala; lane 2: hAMH-RARR-⁵¹⁵Val. (C) Analysis of AMH-⁵¹⁵Ala and AMH-⁵¹⁵Val bioactivity. KK1/AMHR2 cells were transiently transfected with the AMH-responsive BRE-Luc reporter plasmid together with the hAMH variants without (RAQR) or with (RARR) the optimized cleavage site. Luciferase was measured after 24 and 48 h incubation. No differences were observed in luciferase stimulation by both variants. Data are presented as relative luciferase units (RLU) expressed relative to the empty vector control. Data points are the mean ± SEM of triplicates of a representative experiment, which was repeated at least three times.