Fig 2: Functional validation of AMH variants.(a) AMH released in the medium of COS-7 cells transiently transfected either with lipofectamine alone (mock), or with a WT AMH or a variant AMH identified in CHH and KS probands. Bar graph illustrates the mean amount of AMH secreted in the conditioned medium of transfected COS-7 cells (n = 3 independent experiments per condition). Comparisons between treatment groups were performed using a one-way ANOVA followed by Tukey’s post hoc comparison test (F4,10 = 1193, Mock vs AMH WT p<0.0001, AMH WT vs p.Pro151Ser p<0.0001, AMH WT vs p.Asp238Glu p<0.0001, AMH WT vs p.Thr99Ser p<0.0001). No significant motility difference was detected between Mock, p.Thr99Ser and p.Pro151Ser mutated forms of AMH treatment, p>0.9 for all. (b) Transwell assay was performed on GN11 cells transiently transfected either with lipofectamine alone (mock), or with a WT AMH or a variant AMH identified in CHH and KS probands. Comparisons between treatment groups were performed using a one-way ANOVA followed by Tukey’s post hoc comparison test (F4,50 = 13.94, Mock vs AMH WT p<0.0001, AMH WT vs p.Pro151Ser p<0.0001, AMH WT vs p.Asp238Glu p<0.0014, AMH WT vs p.Thr99Ser p = 0.0218. No significant motility difference was detected between Mock and mutated forms of AMH treatment, p>0.9 for all. (c) Quantification of GnRH secretion from GT1-7 cells transfected with lipofectamine alone (mock), or with a WT AMH or the p.Pro151Ser AMH variant identified in a nCHH proband. GnRH mean concentration measured in the medium (n = 3, one-way ANOVA: F 2,6 = 43.84, p = 0.0003; followed by Tukey’s multiple comparison post hoc test, mock vs. AMH WT p = 0.0003, mock vs p.Thr99Ser p = 0.5220, AMH WT vs p.Thr99Ser p = 0.0007. (d) Transwell assay was performed on GN11 cells transiently transfected with the AMHR2 plasmid or with the AMHR2 variant and stimulated with either serum-free medium (SFM) or with recombinant AMH (50 ng/ml). Bar graph illustrates the mean number of migrated GN11 cells under different treatment conditions (SFM n = 10 for both WT and mutant AMHR2, AMH 50 ng/ml n = 12 for both WT and mutant AMHR2). Comparisons between treatment groups were performed using two-way ANOVA (F1,43 = 16.5 P = 0.0002; followed by Sidak’s multiple comparison post hoc test, AMHR2 WT SFM vs AMHR2 WT + AMH 50 ng/ml p<0.0001, p.Gly445_Leu453del SFM vs p.Gly445_Leu453del + AMH 50 ng/ml P = 0.1036). (e) Quantification of GnRH secretion from GT1-7 cells transfected with the same plasmids as in d (n = 3 independent experiments per condition). Experiments were replicated three times with comparable results. Two-way ANOVA, F1,8 = 1.927, p<0.02025; followed by Holm-Šídák multiple comparison post hoc test, AMHR2 WT SFM vs AMHR2 WT + AMH 50 ng/ml P = 0.0269, p.Gly445_Leu453del SFM vs p.Gly445_Leu453del + AMH 50 ng/ml P = 0.4652. (f) Initial three-dimensional models of WT and p.Gly445_Leu453del catalytic intracellular serine/threonine domains of AMHR2. The backbone of the WT and deleted proteins are shown in tan or white cartoon representations, respectively, with the deleted 445–453 residues colored in red. The activation loop is depicted in blue. (g–i) Root-mean-square fluctuations (RMSF) of the Ca atoms along the simulations for the AMHR2 WT and the p.Gly445_Leu453del models. (g) RMSF (in Å) for the WT (black line) and the p.Gly445_Leu453del models (red line, being the average over the three 100 ns simulations) are given for each residue of the protein. For a better comparison, residue numbers were kept the same for both models. Molecular representation of the WT (h) and p.Gly445_Leu453del (i) models colored by RMSF: the blue-green-red scale corresponds to low-medium-high RMSF values. The yellow spheres indicate the first residues after the p.Gly445_Leu453del deletion. The activation loop region is highlighted inside a blue frame (arrows).10.7554/eLife.47198.019Figure 7—source data 1.This spreadsheet contains the values used to generate the bar plots shown in Figure 7a–e.

Fig 3: GnRH migration and olfactory innervation are perturbed in Amhr2-/- mice.(a) Schematic representation depicting whole-body iDISCO experiments in E13.5 Amhr2+/+ and Amhr2-/- embryos. E13.5 embryos (n = 2 per genotype) were immunolabelled for Peripherin and GnRH, rendered optically transparent using iDISCO and imaged with a light-sheet microscope (LSM). (b, c) Frontal projection of the embryo heads, arrowheads indicate noticeable differences in Peripherin-positive fibers innervating the olfactory bulb (OB). Lateral projection views (d, e) showing defective GnRH migration and terminal nerve projections to the ventral forebrain (vFB, arrows). (f, g) Higher magnification photomicrographs depicting olfactory axon innervations of the right OB shown in b and c. Dotted circles define the anatomical border of the OB. (h, i) 3D rendering of figures in f and g. Arrowheads indicate observed differences in olfactory axon innervation between Amhr2+/+ and Amhr2-/- embryos. (j, k) 3D rendering of peripherin and GnRH staining observed from a lateral projection in a representative Amhr2+/+ and Amhr2-/- embryo. Cx: cortex; VNO: vomeronasal organ. Scale bars: (b) 400 µm; (d) 300 µm; (f) 130 µm.

Fig 4: In utero pharmacological invalidation of Amhr2 disrupts GnRH neuronal migration and the olfactory/terminal nerve targeting.(a) Schematic of in utero injections targeting the olfactory pits. Injections were performed at E12.5 and embryos harvested 48 hr later. (b) Representative coronal section of an embryo head at E14.5 showing that olfactory pit Fluorogold delivery at E12.5 was successful. GnRH immunoreactive neurons are shown in green. (c–f) Representative photomicrographs of sagittal sections of mouse embryos injected at E12.5 with either saline or a neutralizing antibody for Amhr2 (Amhr2-NA) and immunostained for GnRH (green) and Peripherin (magenta) at E14.5. (e, f) Higher magnification confocal photomicrograph of boxed areas in c and d. (g) Quantification of the total number of GnRH immunoreactive neurons in saline-injected (control) and Amhr2-NA injected embryos (n = 4 for both groups, harvested from two independent dams). Data are represented as mean ± s.e.m (n = 4, unpaired two-tailed Student’s t test: mean cell number, t6 = 0.3796, p = 0.7173). (h) Quantitative analysis of GnRH neuronal distribution throughout the migratory pathway in the two experimental groups. Data are represented as mean ± s.e.m (n = 4, two-way ANOVA, F3,24 = 15.09, p<0.0001; followed by Holm-Šídák multiple comparison post hoc test, **p<0.005; n.s., not significant; N/FB J Amhr2+/+ vs. N/FB J Amhr2-/-p = 0.99, CX Amhr2+/+ vs. CX Amhr2-/-p = 0.88). Cx: cortex; FB: forebrain; N/FBJ: nasal/forebrain junction; oe: olfactory epithelium; NMC: nasal mesenchyme. Scale bars: (b) 100 µm; (d) 2.5 mm; (f) 50 µm.10.7554/eLife.47198.005Figure 2—source data 1.This spreadsheet contains the values used to generate the bar plots shown in Figure 2g and h.

Fig 5: Amhr2 mutant mice show reduced GnRH cell number and impaired LH secretion and fertility.(a–h) Immunolabelling of GnRH (red staining) in adult wild type and Amhr2-/- adult female mice (P90–P120). The majority of GnRH cell bodies are located at the level of the organum vasculosum laminae terminalis (OVLT) in both Amhr2 +/+ and Amhr2 -/- mice, (arrows (c, d). (e–h) GnRH fiber projections at the level of the median eminence. (i) Total mean GnRH population in Amhr2+/+, Amhr2+/- and Amhr2-/- adult female mice brains (3–4 months old). Comparisons between groups were performed using one-way ANOVA followed by Tukey’s post hoc test (n = 7 for all groups, F2,18 <0.0001; Amhr2+/+ vs Amhr2+/- P = 0.4716; WT vs. Amhr2-/-p<0.0001, Amhr2+/- vs Amhr2-/-p = 0.0007). (j) Plasma LH levels in adult mature (4–6 months old) diestrous females (Amhr2+/+, n = 4; Amhr2+/-, n = 5; Amhr2-/- n = 3). Statistical analysis was performed by one-way ANOVA (F2,9 = 12.64, p = 0.0024) followed by Tukey’s multiple comparison post hoc test (Amhr2+/+ vs. Amhr2+/- P = 0.005; Amhr2+/+ vs. Amhr2-/-p = 0.046, Amhr2+/- vs. Amhr2-/-p = 0.8164). (k) Quantitative analyses of the mean number of corpora lutea (CL) in Amhr2+/+ (n = 5), Amhr2+/- (n = 4) and Amhr2-/- (n = 5) adult ovaries (4–6 months old). Statistical significance between groups was assessed using one-way ANOVA (F2,11 = 22.11, p = 0.0001) followed by Tukey’s multiple comparison post hoc test (Amhr2+/+ vs. Amhr2+/- P = 0.6259; Amhr2+/+ vs. Amhr2-/-p = 0.0002 and Amhr2+/- vs. Amhr2-/-p = 0.0012). (l) Representative graphs for LH pulsatility in female dioestrous adult mice of the corresponding genotype. Asterisks indicate the number of LH pulses per 2 hr interval. (m) Number of LH pulses in adult (P60) diestrous females (Amhr2+/+, n = 5; Amhr2+/-, n = 4; Amhr2-/- n = 3). Statistical analysis was performed by non-parametric Kruskal-Wallis test p = 0.0028 (Amhr2+/+ vs. Amhr2+/- P = 0.041; Amhr2+/+ vs. Amhr2-/-p = 0.038 and Amhr2+/- vs. Amhr2-/-p>0.999). (n) Bar graphs illustrating the results of the constant mating protocol performed over 90 days on the following groups: (?Amhr2+/+ x ?Amhr2+/+, n = 9; ?Amhr2+/- x ?Amhr2+/+, n = 12; ?Amhr2-/- x ?Amhr2+/+, n = 4; ?Amhr2+/+ x ?Amhr2+/-, n = 3; ?Amhr2+/+ x ?Amhr2+/-, n = 3. Female and male mice were 4–6 months-old). Comparisons between groups were performed using one-way ANOVA (fertility index, F4,26 = 51.47, p<0.0001; first litter, F4,26 = 88.82, p<0.0001; pups per litter, F4,26 = 29.67 P<0.0001) followed by Tukey’s multiple comparison post hoc test, *p<0.05; **p<0.005; ***p<0.0005; ****p<0.0001. Each cluster of data points represents a different mouse. Data were combined from three independent experiments. Throughout the figure, data are displayed as mean ± s.e.m. *p<0.05; **p<0.005; ***p<0.0005; ****p<0.0001. Scale bars: (a, b, e, f) 100 µm; (c, d, g, h) 50 µm.10.7554/eLife.47198.011Figure 4—source data 1.This spreadsheet contains the values used to generate the bar plots shown in Figure 4i j, k, m, n.