Bisphenol S Impairs Oestradiol Secretion during In Vitro Basal Folliculogenesis in a Mono-Ovulatory Species Model

Abstract

Bisphenol S (BPS) affects terminal folliculogenesis by impairing steroidogenesis in granulosa cells from different species. Nevertheless, limited data are available on its effects during basal folliculogenesis. In this study, we evaluate in vitro the effects of a long-term BPS exposure on a model of basal follicular development in a mono-ovulatory species. We cultured ovine preantral follicles (180−240 μm, n = 168) with BPS (0.1 μM (possible human exposure dose) or 10 μM (high dose)) and monitored antrum appearance and follicular survival and growth for 15 days. We measured hormonal secretions (oestradiol (at day 13 [D13]), progesterone and anti-Müllerian hormone [D15]) and expression of key follicular development and redox status genes (D15) in medium and whole follicles, respectively. BPS (0.1 µM) decreased oestradiol secretion compared with the control (−48.8%, p < 0.001), without significantly impairing antrum appearance, follicular survival and growth, anti-Müllerian hormone and progesterone secretion and target gene expression. Thus, BPS could also impair oestradiol secretion during basal folliculogenesis as it is the case during terminal folliculogenesis. It questions the use of BPS as a safe BPA substitute in the human environment. More studies are required to elucidate mechanisms of action of BPS and its effects throughout basal follicular development.

Keywords: bisphenols; endocrine disruptors; ewe; follicular growth; gene expression; hormonal secretions; ovary; plasticiser.

Figure 1

In vitro effects of bisphenol S (BPS) on ovine follicular survival. The survival of each follicle was assessed after 6, 13 and 15 days of treatment with or without BPS 0.1 µM or BPS 10 µM. A follicle was considered alive when there was no breakdown of the basal lamina and no extrusion of the oocyte. The oocyte had to be clear, and the follicular growth had to be measurable within 1 week. The results are representative of seven independent cultures with eight replicates per condition (n = 56 per condition). The results are expressed as percentage of alive follicles and the Kaplan–Meier survival curves obtained for the three conditions were compared using the Log-rank (Mantel–Cox) test. Differences were considered significant when p < 0.05.

Figure 2

In vitro effects of bisphenol S (BPS) on ovine follicular growth and antrum appearance. The diameter evolution (A) and the antral cavity appearance (B) for each alive follicle were assessed after 6, 13 and 15 days of treatment with or without BPS 0.1 µM or BPS 10 µM. The results are representative of seven independent cultures with eight replicates per condition (n = 56 per condition at day 0, n = 48–53 according to the conditions at day 6, n = 40–42 according to the conditions at day 13 and n = 38–39 according to the conditions at day 15). For the diameter evolution, the results are expressed as mean +/− SEM and were analysed with a Brown–Forsythe ANOVA test at each day of measure (A). For the antrum appearance, the results are expressed as the percentage of alive follicles with an antrum and were analysed with a logistic regression at each day of measure (B). Differences were considered significant when p < 0.05.

Figure 3

In vitro effects of bisphenol S (BPS) on ovine follicular hormonal secretions. Ovine follicles were cultured for 15 days with or without BPS (0.1 or 10 µM). Hormonal secretions were measured by ELISA in spent culture media of alive follicles after 13 days (D13) of treatment for oestradiol (A) and after 15 days (D15) for progesterone (B) and AMH (C). The results are representative of six independent cultures with eight replicates per condition at the beginning of the experiment for oestradiol (A), n = 34–37 alive follicles according to the conditions at day 13 and n = 32–33 according to conditions for progesterone (B) and AMH (C) at day 15. The data are expressed as mean +/− SEM relative to controls and were analysed with a Kruskal–Wallis test followed by a Dunn’s multiple comparison post hoc test. Bars without at least one common letter (a,b) are significantly different (p < 0.05).

Figure 4

In vitro effects of bisphenol S (BPS) on ovine basal stage follicular gene expression. Ovine follicles were cultured for 15 days with or without BPS (0.1 or 10 µM). At day 15, the culture was stopped, and 64 alive follicles were used to assess the expression of 19 gene; they were preserved in a lysis buffer for RNA extraction and stored at −80 °C until use. The results are representative of four independent cultures, with n = 18 for control, n = 22 for BPS 0.1 µM and n = 24 for BPS 10 µM. The geometric mean of two housekeeping genes (beta-actin (ACTB) and ribosomal protein L19 (RPL19)) was used to normalise gene expression. The data are expressed as mean +/− SEM and were analysed with the Brown–Forsythe ANOVA test followed by Dunnett’s T3 multiple comparison post-hoc test. In this figure, the results are presented for the 3 genes for which a p value ≤ 0.10 was obtained with the Brown–Forsythe ANOVA: one gene involved in follicular development, Cytochrome P450 Family 19 Subfamily A Member 1 (CYP19A1, p = 0.064, (A) and two genes involved in redox status, Glutathione Peroxidase 8 (GPX8, p = 0.068, (B) and NADH Ubiquinone Oxidoreductase Complex Assembly Factor (NDUFAF2, p = 0.097, (C). The p values ≤ 0.10 obtained with the Dunnett’s T3 multiple comparison post-hoc test are drawn between conditions on each graph.