Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2018 Jun;24(6):834-846.
doi: 10.1038/s41591-018-0035-5. Epub 2018 May 14.

Elevated prenatal anti-Müllerian hormone reprograms the fetus and induces polycystic ovary syndrome in adulthood

Brooke Tata  1   2 , Nour El Houda Mimouni  1   2 , Anne-Laure Barbotin  1   3 , Samuel A Malone  1   2 , Anne Loyens  1   2 , Pascal Pigny  2   4 , Didier Dewailly  1   2   5 , Sophie Catteau-Jonard  1   2   5 , Inger Sundström-Poromaa  6 , Terhi T Piltonen  7 , Federica Dal Bello  8 , Claudio Medana  8 , Vincent Prevot  1   2 , Jerome Clasadonte  1   2 , Paolo Giacobini  9   10
Affiliations

Elevated prenatal anti-Müllerian hormone reprograms the fetus and induces polycystic ovary syndrome in adulthood

Brooke Tata et al. Nat Med. 2018 Jun.

Abstract

Polycystic ovary syndrome (PCOS) is the main cause of female infertility worldwide and corresponds with a high degree of comorbidities and economic burden. How PCOS is passed on from one generation to the next is not clear, but it may be a developmental condition. Most women with PCOS exhibit higher levels of circulating luteinizing hormone, suggestive of heightened gonadotropin-releasing hormone (GnRH) release, and anti-Müllerian hormone (AMH) as compared to healthy women. Excess AMH in utero may affect the development of the female fetus. However, as AMH levels drop during pregnancy in women with normal fertility, it was unclear whether their levels were also elevated in pregnant women with PCOS. Here we measured AMH in a cohort of pregnant women with PCOS and control pregnant women and found that AMH is significantly more elevated in the former group versus the latter. To determine whether the elevation of AMH during pregnancy in women with PCOS is a bystander effect or a driver of the condition in the offspring, we modeled our clinical findings by treating pregnant mice with AMH and followed the neuroendocrine phenotype of their female progeny postnatally. This treatment resulted in maternal neuroendocrine-driven testosterone excess and diminished placental metabolism of testosterone to estradiol, resulting in a masculinization of the exposed female fetus and a PCOS-like reproductive and neuroendocrine phenotype in adulthood. We found that the affected females had persistently hyperactivated GnRH neurons and that GnRH antagonist treatment in the adult female offspring restored their neuroendocrine phenotype to a normal state. These findings highlight a critical role for excess prenatal AMH exposure and subsequent aberrant GnRH receptor signaling in the neuroendocrine dysfunctions of PCOS, while offering a new potential therapeutic avenue to treat the condition during adulthood.

PubMed Disclaimer

Conflict of interest statement

Competing Financial Interests

The authors have no financial competing interests.

Figures

Figure 1
Figure 1. AMH levels during the second trimester of gestation are higher in women with PCOS than controls.
(a) Blood samples were derived from control and pregnant women with PCOS at gestational week 16-19 and AMH concentration was measured by ELISA. (b) Circulating AMH levels in control pregnant women (n = 63) and in pregnant women with PCOS (n = 66). Statistics by unpaired two-tailed Mann-Whitney U test, ***P ≤ 0.0001. (c) Circulating AMH levels in control pregnant women and in pregnant individuals with PCOS stratified by their body mass index (BMI) and classified into lean (Control lean n = 30, PCOS lean n = 32) and obese subjects (Control obese n = 33, PCOS obese n = 34). (d) Circulating AMH levels in pregnant women with PCOS stratified by their BMI and androgen levels (PCOS lean normoandrogenic n = 15, PCOS lean hyperandrogenic n = 16, PCOS obese normoandrogenic n = 16, PCOS obese hyperandrogenic n = 18). (e) Circulating AMH levels in control pregnant women and in PCOS pregnant subjects stratified by their age (Control 27-34 years old n = 42, PCOS 27-34 years old n = 43, Control > 34 years old n = 21, PCOS > 34 years old n = 23). The horizontal line in each plot corresponds to the median value from two technical replicates. The vertical line represents the 25th – 75th percentile range. Statistics in c-e were computed with one-way ANOVA (c: F(3, 125) = 7.534, P = 0.0001; d: F(3, 61) = 3.922, P = 0.0126; e: F(3, 125) = 6.282, P = 0.0005) followed by Bonferroni post hoc test, * P < 0.05, ** P ≤ 0.005, *** P ≤ 0.0005; n.s. = not significant.
Figure 2
Figure 2. Prenatal AMH treatment disrupts estrous cyclicity, ovarian morphology and fertility in adult offspring.
(a) Schematic of experimental design whereby pregnant dams were subjected to different treatments of intraperitoneal (i.p.) injections during the late gestational period (embryonic days (E) 16.5 - E18.5). Pregnant dams (P90-P120; n = 34) were split into four treatment groups: PBS-treated (n = 8), AMH-treated (AMHc, n = 10), AMH+GnRH antag-treated (AMHc plus Cetrorelix acetate, n = 8), GnRH antagonist-treated (Cetrorelix acetate alone, 0.5 mg/Kg, n = 8). The offspring were designated as follows: Control, (PBS-treated); PAMH (Prenatal recombinant AMHc-treated); PAMH+GnRH antag, (PAMH plus Cetrorelix acetate); GnRH antag (Cetrorelix acetate alone). (b) Quantitative analysis of ovarian cyclicity in adult (P60-P90) offspring mice (Control, n = 15; PAMH, n = 19; PAMH+GnRH antag, n = 13; GnRH antag, n = 11). Vaginal cytology was assessed for 16 days. The horizontal line in each scatter plot corresponds to the median value. The vertical line represents the 25th – 75th percentile range. Comparisons between treatment groups were performed using Kruskal-Wallis test followed by Dunn’s post hoc analysis test; *** P < 0.0001. Data were combined from three independent experiments. (c) Representative estrous cyclicity of 10 mice/treatment group during 16 consecutive days. (d) Quantitative analysis of corpora lutea, late antral follicles and atretic follicles in the ovaries of Control (n = 7, age: P90) and PAMH mice (n = 8, age: P90). Statistics were performed with unpaired two-tailed Student’s t-test (corpora lutea, t(13) = 4.879, **P = 0.0003; late antral follicles, t(13) = 4.637, ** P = 0.0005; atretic follicles, t(13) = 0.226, P = 0.8243, n.s. = not significant). Data are represented as mean ± s.e.m. and were combined from two independent experiments. (e) Fertility tests of the adult offspring mice (P90). Mating was performed for 90 days. Control females were paired with Control males (n = 7), PAMH females were paired with PAMH males (n = 7 for each sex), PAMH+GnRH antag females were paired with PAMH+GnRH antag males (n = 6 for each sex), and GnRH antag females were paired with GnRH antag males (n = 8 for each sex). Data are represented as mean ± s.e.m. Statistics in e were computed with one-way ANOVA (First litter, F(3,24) = 18.14, P < 0.0001; Fertility Index, F(3,24) = 7.647, P = 0.0009; Number of pups, F(3,24) = 24.26, P < 0.0001) followed by Tukey’s multiple comparison post hoc test, * P < 0.05, ** P ≤ 0.005, *** P ≤ 0.0005. Data of fertility tests were combined from two independent experiments.
Figure 3
Figure 3. Prenatal AMH treatment leads to hyperandrogenism and elevation in LH secretion/pulsatility.
(a) Anogenital distance (AGD) measured over post-natal days (P) indicated in female Control mice (P30, n = 16; P35, n = 16; P40, n = 16; P50, n = 10; P60, n = 10), PAMH mice (P30, n = 29; P35, n = 32; P40, n = 32; P50, n = 24; P60, n = 24), PAMH+GnRH antag mice (n = 13/age), and GnRH antag (n = 11/age) mice. Statistics were computed with one-way ANOVA (P30, F(3,65) = 108.2, P < 0.0001; P35 F(3,68) = 155, P < 0.0001; P40, F(3,68) = 234.5, P < 0.0001; P50, F(3,54) = 143.4, P < 0.0001; P60, F(3,54) = 165.4, P < 0.0001) followed by Tukey’s multiple comparison post hoc test. Data were pooled from 3 independent experiments. (b) Plasma testosterone concentration in adult females (P60) in diestrus: control group Control (n = 5), PAMH mice (n = 5) and PAMH+GnRH antag (n = 5) littermates. Statistics were computed with One-way ANOVA (F(2,12) = 28.67, P < 0.0001), followed by Tukey’s multiple comparison post hoc test. (c) Plasma LH levels in adult diestrous females (P60-P90). Number of mice: Control, n = 8; PAMH, n = 10; PAMH+GnRH antag, n = 9. One-way ANOVA (F (2, 24) = 257.3, ***P < 0.0001), followed by Tukey’s multiple comparison post hoc test. Data in b and c were combined from two independent replicates. (d) Schematic representation of tail-tip blood sampling in adult diestrous female mice (3-4 months old). (e) Number of LH pulses in adult (P60) diestrous females (Control, n = 8; PAMH, n = 10; PAMH+GnRH antag, n = 9). Statistics were computed with one-way ANOVA (F(2,24)= 57.06, P < 0.0001) followed Tukey’s multiple comparison post hoc test. Data were combined from three independent experiments. (f) Representative graphs for LH pulsatility of three independent experiments. Asterisks in (f) indicate the number of LH pulses/2-hr. (g) Hormone levels in trunk blood measured by ELISA at E19.5 in dams injected from E16.5 to E18.5 with either PBS (vehicle, n = 9-11), AMH (n = 8-10), AMH + GnRH antag (n = 5). Statistics were computed with one-way ANOVA (LH, F(2,22) = 11.48, P = 0.0004; Testosterone, F(2,22) = 8.095, P = 0.0023; Estradiol, E2, F(2,14) = 31.13, P < 0.0001; Progesterone F(2,14) = 32.08, P < 0.0001) followed by Tukey’s multiple comparison post hoc test. Data were pooled from at least two independent experiments. Statistical significance in a, b, c, e, g: * P < 0.05, ** P < 0.001, *** P = 0.0001. (h) Real-time PCR analysis for Amhr2, Cyp191a (cytochrome P450 family 19, subfamily a, polypeptide 1), Cyp11a1 (cytochrome P450, family 11, subfamily a, polypeptide 1), Hsd3b1 (hydroxy-delta-5-steroid dehydrogenase, 3 beta- and steroid delta-isomerase 1) mRNA in the placenta of E19.5 dams. Dams were injected i.p. from E16.5 to E18.5 with either PBS (vehicle; Amhr2, n = 16; Cyp191a, n = 17; Cyp11a1, n = 12; Hsd3b1, n = 9), AMH (Amhr2, n = 16; Cyp191a, n = 11; Cyp11a1, n = 9; Hsd3b1, n = 8), or AMH + GnRH antag (Amhr2, n = 16; Cyp191a, n = 13; Cyp11a1, n = 10; Hsd3b1, n = 10). Comparisons between treatment groups were performed using Kruskal-Wallis test followed by Dunn’s post hoc analysis test, *P < 0.05, ***P < 0.0001; n.s, not significant (P > 0.05). Data were combined from three independent experiments. Throughout the figure, data are displayed as mean ± s.e.m.
Figure 4
Figure 4. Prenatal AMH treatment increases perinatal T levels in females and masculinizes their brain.
(a) Plasma T and LH levels measured in pups 2 hours after birth (Males, n = 5; Control females, n = 5; PAMH females, n = 4; PAMH females + GnRH antag, n = 6). Statistics were computed with one-way ANOVA (Testosterone, F(3, 16) = 227.6, P < 0.0001; LH, F(3, 16) = 163.3, P < 0.0001) and Tukey’s multiple comparison post hoc test. Error bars represent s.e.m. of three technical replicates. (b) Schematic representation depicting the analyzed sexual dimorphic neuroanatomical regions of the brain (denoted in red: anteroventral periventricular nucleus, AVPV; bed nucleus of the stria terminalis, BNST; medial amygdala, MeA) expressing Vasopressin (VP) and Tyrosine Hydroxylase (TH). (c) Representative photomicrographs of three independent experiments showing TH-immunoreactive neurons in AVPV of a male, control female and PAMH female at P60. TH immunoreactive (-ir) neurons in the AVPV were quantified in these animal groups (n = 3 per sex and treatment). The number of AVPV TH-ir neurons are represented as the mean ± s.e.m. Statistics were computed with one-way ANOVA (F(2,6) = 231.1, P < 0.0001) and Tukey’s multiple comparisons post-hoc test. (d) Representative photomicrographs of three independent experiments showing VP-immunoreactive neurons in the BNST and MeA of a male, control female and PAMH female at P60. VP–ir neurons in the BNST and MeA were quantified in these animal groups (n = 3 per sex and treatment). Data are displayed as mean ± s.e.m. Statistics were computed with one-way ANOVA (BNST, F(2,6) = 185.2, P < 0.0001; MeA, F(2,6) = 351.7, P < 0.0001) followed by Tukey’s multiple comparison post hoc test. Statistical significance in a, c, d: * P < 0.05, ** P < 0.005, *** P < 0.0001. Scale bars, 50 μm.
Figure 5
Figure 5. PAMH/GnRH-GFP mice exhibit higher GnRH dendritic spine density, increased GABAergic appositions to GnRH neurons and elevated firing frequency of GnRH neurons in adulthood.
(a-d) Representative projected confocal images and 3D reconstruction of three independent experiments showing GnRH-GFP neurons from an adult (P60) female diestrous control (a, b) and a PAMH (c, d) female diestrous mouse. (e) The GnRH dendritic spine density was analyzed in adult (P60) diestrous Control and PAMH/GnRH-GFP females. Spine density was quantified at the soma (Control, n =10 neurons from three animals; PAMH, n = 10 neurons from three animals) and each 15 µm portion of the primary GnRH neuronal dendrite (Control, n =12 neurons from three animals; PAMH, n =12 neurons from three animals). Statistics were performed with a paired two-tailed Student’s t-test (soma, t(18) = 7.5434, ***P < 0.0001; 15 µm from soma, t(22) = 9.08, ***P < 0.0001; 30 µm from soma, t(22) = 6.92, ***P < 0.0001; 45 µm from soma, t(22) = 5.252, ***P < 0.0001). (f, g) Representative projected confocal images of three independent experiments showing a GFP-immunoreactive GnRH neuron (green) surrounded by vGaT-immunoreactive (red) puncta in an adult diestrous Control (f) and PAMH female mouse (g). Points where the vGaT signal was considered to be immediately adjacent to the GnRH neuron are indicated by arrowheads. (h) Quantification of the number of vGaT-immunoreactive puncta adjacent to GnRH neuron soma (Control, n =10 neurons from three animals; PAMH, n = 10 neurons from three animals) and along the primary GnRH neuronal dendrite is expressed as vGaT appositions/µm (Control, n =10 neurons from three animals; PAMH, n =10 neurons from three animals). Statistics were performed with a paired two-tailed Student’s t-test (soma, t(18) = 16.4, ***P < 0.0001; 15 µm from soma, t(18) = 7.341; ***P < 0.0001, 30 µm from soma t(18) = 9.929, ***P < 0.0001; 45 µm from soma; t(18) = 16.4, ***P < 0.0001). (i-m) Spontaneous electrical activity and membrane properties of GnRH neurons recorded in acute brain slices from control and PAMH GnRH-GFP female diestrous mice. (i) Whole-cell current-clamp recording showing the typical spontaneous burst firing of a GnRH neuron from a control mouse. The bottom trace shows an expanded time scale of that recording. (j) Same experiment as in (i) but in a GnRH neuron from PAMH mouse. (k) Average firing rate of GnRH neurons recorded from Control (n = 9 cells from 4 animals, age P90-P120) and PAMH mice (n = 8 cells from 4 animals, age P90-P120). Statistics were performed with unpaired two-tailed Student’s t-test (t(15) = 2.98, *P < 0.05). (l) Average resting membrane potential (RMP) of GnRH neurons recorded from control (n =10 cells from 4 animals, age P90-P120) and PAMH mice (n = 8 cells from 4 animals, age P90-P120). Two-tailed Student’s t-test, t(16) = 1.143, P = 0.27, n.s. not significant. (m) Average input resistance (Rin) of GnRH neurons recorded from control (n = 6 cells from 4 animals, age P90-P120) and PAMH mice (n = 7 cells from 4 animals, age P90-P120). Unpaired two-tailed Student’s t-test, t(11) = 0.3685, P = 0.7195, n.s. not significant. Throughout, data are displayed as mean ± s.e.m. Experiments were replicated three times with comparable results. Scale bars, 50 μm.
Figure 6
Figure 6. Postnatal GnRH antagonist treatment of PAMH mice restores the PCOS-like neuroendocrine phenotype.
(a) Schematic of experimental design whereby estrous cyclicity in PAMH adult mice was analyzed during 3 months, before and after postnatal intraperitoneal (i.p.) injections with 0.05, 0.5 and 5 mg/Kg Cetrorelix acetate. The Y axis refers to the different stages of the estrous cycle: Metaestrus/Diestrus (M/D), Estrus (E) and Proestrus (P). The X axis represents the time-course of the experiments (days). PAMH female mice (P60, n = 6) were injected i.p. for 12 days with Cetrorelix acetate at 0.05 mg/Kg (every second day) followed by Cetrorelix acetate at 0.5 mg/Kg (every second day) and finally by Cetrorelix acetate at 5 mg/Kg (every day). Tail-blood samples were collected for LH measurements twice before the beginning of the treatments, and at day 2 and 6 of each treatment as well as 4 days after the last injection (no treatment), that followed each administration period. (b) Quantitative analysis of the % of completed estrous cycles in Control (PBS-treated, n = 19), PAMH (n = 13), and PAMH mice (n = 6) postnatally treated with the three doses of GnRH antagonist (Cetrorelix 0.05 mg/Kg, Cetrorelix 0.5 mg/Kg, Cetrorelix 5 mg/Kg). The horizontal line in each scatter plot corresponds to the median value. The vertical line represents the 25th – 75th percentile range. Comparisons between treatment groups were performed using Kruskal-Wallis test followed by Dunn’s post hoc analysis test. Data were combined from two independent experiments. (c) Time course of serum LH concentration in PAMH mice (n = 6) before the beginning of the treatment (day 0), 2 and 4 days after the first injection of each dose of Cetrorelix and after discontinuation of the drug (recovery time). Control animals (P60 females, n = 3) were injected with PBS and tail-blood was collected during the same temporal windows as for Cetrorelix treatment. Error bars represent s.e.m. from two independent experiments with two technical replicates each. Statistics were computed with one-way ANOVA (before treatment, F(3,21) = 10.6, P = 0.0004; day 2, F(3,21) = 45.72, P < 0.0001; day 6, F(3,21) = 67.10, P < 0.0001, recovery, F(3,21) = 32.62, P < 0.0001) followed by Bonferroni comparison post hoc test. (d) Percentage (%) of time spent in each estrous cycle in adult mice (Control or PAMH) injected for 12 days with either PBS or Cetrorelix at 0.5 mg/Kg every two days. The horizontal line in each scatter plot corresponds to the median value. The vertical line represents the 25th – 75th percentile range. The percentage (%) of time spent in each estrous cycle was quantified in Control (n = 19), PAMH (n = 11) and PAMH+Cetrorelix 0.5 mg/Kg (n = 11) mice (age: P60-P90). Comparisons between treatment groups were performed using Kruskal-Wallis test followed by Dunn’s post hoc analysis test. Data were combined from three independent experiments. (e) Mean T levels measured in diestrus, at the end of the treatment time, in adult Control (n = 7), PAMH (n = 5) and PAMH+Cetrorelix-treated (0.5 mg/Kg; n = 5) animals. Statistics was computed with one-way ANOVA (F(2,14) = 29.24, P < 0.0001) followed by Tukey’s multiple comparison post hoc test. (f) Mean LH levels were measured in diestrus, at the end of the treatment time, in adult (P60-P120) Control mice (n = 9), PAMH (n = 16) and PAMH+Cetrorelix 0.5 mg/Kg (n = 11). Data were combined from three independent experiments. Statistics was computed with one-way ANOVA (F(2,33) = 72.2, P < 0.0001) followed by Tukey’s multiple comparison post hoc test. (g) LH pulsatility in adult (P90) diestrous females (Control, n = 6; PAMH, n = 6; PAMH+Cetrorelix 0.5 mg/Kg, n = 6). Statistics was computed with one-way ANOVA (F(2,15) = 40.71, P < 0.0001) followed by Tukey’s multiple comparison post hoc test. (h) Representative photomicrographs of ovaries stained with haematoxylin–eosin from adult (P60-P90) control, PAMH and Cetrorelix 0.5 mg/Kg mice (n = 3 each group). Right panels show the quantitative analyses for the mean number of corpora lutea (CL) and antral follicles (AF) in each treatment group. Comparisons between treatment groups were performed using one-way ANOVA (CL, F(2,6) = 23.38, P = 0.0015; AF, F(2,6) = 21, P = 0.002) followed by Tukey’s multiple comparisons post-hoc test. Values in e-h are represented as the mean ± s.e.m. Statistical significance in b-h: * P < 0.05, ** P < 0.005, *** P < 0.0001; n.s., not significant (P > 0.5). Scale bars, 200 μm.
See this image and copyright information in PMC

Comment in

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Norman RJ, Dewailly D, Legro RS, Hickey TE. Polycystic ovary syndrome. Lancet. 2007;370:685–697. - PubMed
    1. Goodarzi MO, Dumesic DA, Chazenbalk G, Azziz R. Polycystic ovary syndrome: etiology, pathogenesis and diagnosis. Nature reviews. Endocrinology. 2011;7:219–231. - PubMed
    1. Jayasena CN, Franks S. The management of patients with polycystic ovary syndrome. Nature reviews. Endocrinology. 2014;10:624–636. - PubMed
    1. March WA, et al. The prevalence of polycystic ovary syndrome in a community sample assessed under contrasting diagnostic criteria. Human reproduction. 2010;25:544–551. - PubMed
    1. Wild RA, et al. Assessment of cardiovascular risk and prevention of cardiovascular disease in women with the polycystic ovary syndrome: a consensus statement by the Androgen Excess and Polycystic Ovary Syndrome (AE-PCOS) Society. The Journal of clinical endocrinology and metabolism. 2010;95:2038–2049. - PubMed

Publication types

MeSH terms

Substances