Altered theca and cumulus oocyte complex gene expression, follicular arrest and reduced fertility in cows with dominant follicle follicular fluid androgen excess

Abstract

Aspiration of bovine follicles 12-36 hours after induced corpus luteum lysis serendipitously identified two populations of cows, one with High androstenedione (A4; >40 ng/ml; mean = 102) and another with Low A4 (<20 ng/ml; mean = 9) in follicular fluid. We hypothesized that the steroid excess in follicular fluid of dominant follicles in High A4 cows would result in reduced fertility through altered follicle development and oocyte maternal RNA abundance. To test this hypothesis, estrous cycles of cows were synchronized and ovariectomy was performed 36 hours later. HPLC MS/MS analysis of follicular fluid showed increased dehydroepiandrosterone (6-fold), A4 (158-fold) and testosterone (31-fold) in the dominant follicle of High A4 cows. However, estrone (3-fold) and estradiol (2-fold) concentrations were only slightly elevated, suggesting a possible inefficiency in androgen to estrogen conversion in High A4 cows. Theca cell mRNA expression of LHCGR, GATA6, CYP11A1, and CYP17A1 was greater in High A4 cows. Furthermore, abundance of ZAR1 was decreased 10-fold in cumulus oocyte complexes from High A4 cows, whereas NLRP5 abundance tended to be 19.8-fold greater (P = 0.07). There was a tendency for reduction in stage 4 follicles in ovarian cortex samples from High A4 cows suggesting that progression to antral stages were impaired. High A4 cows tended (P<0.07) to have a 17% reduction in calving rate compared with Low A4 cows suggesting reduced fertility in the High A4 population. These data suggest that the dominant follicle environment of High A4 cows including reduced estrogen conversion and androgen excess contributes to infertility in part through altered follicular and oocyte development.

Figure 1. Individual Aspirate (A–D) and Temporal (E–H) differences in Dominant Follicle Steroid Hormone Concentrations Following PG.

Androstenedione (A4) concentrations (A) naturally separate above (blue) or below (red) the average (43 ng/mL) with some samples indicated in the moderate region (20–40 ng/mL; grey). Estradiol (E2; B), Progesterone (P4; C) and dehydroepiandrosterone (DHEA; D) concentrations are similar regardless of A4 concentrations. At each time point measured (12–36 hours) androstenedione concentrations (E) are increased in High A4 cows; although E2 concentration (F), P4 concentration (G), and DHEA concentration (H) were similar for High and Low A4 cows.

Figure 2. HPLC/MS-MS Analysis of Dominant Follicle Sex Steroid Concentrations Collected after Ovariectomy.

Differences in concentration of androgens (A) and estrogens (B) was determined by HPLC/MS-MS (Biocrates Life Sciences AG) analysis for High A4 (black bars) and Low (white bars) A4 cows. *, P≤0.05.

Figure 3. HPLC/MS-MS Analysis of Dominant Follicle Mineralocorticoid and Glucocorticoid Concentrations Collected after Ovariectomy.

Difference in mineralocorticoid (A) and glucocorticoid (B) concentration in dominant follicle determined by HPLC/MS-MS (Biocrates Life Sciences AG) analysis for High A4 (black bars) and Low (white bars) A4 cows.*P≤0.05.

Figure 4. High A4 Cows Have Increased Steroidogenic Gene Expression in Theca Cells.

Quantitative RT-PCR results for StAR (A), LHCGR (B), CYP11A1 (C), CYP17A1 (D), and GATA6 (E) in theca cells collected from dominant follicles of High A4 (black bars, n≥12) and Low A4 (white bars, n≥19) cows. The geometric mean of GAPDH and RPL-15 was used as an endogenous control. Data for CYP11A1, CYP17A1, LHCRG, and GATA6 were log transformed to meet normal distribution assumptions. Graphs represent a fold change in mRNA abundance with Low A4 set as control (1). The mean ± SEM normalized values are presented. *, P≤0.05.

Figure 5. Maternal Effect Gene ZAR1 mRNA Abundance is Reduced in High A4 Compared with Low A4 Cows.

Quantitative RT-PCR results for ZAR1 (A), NLRP5 (B), DPPA3 (C), and DNMT1 (D); in cumulus oocyte complexes of High A4 (black bars, n≥3) and Low A4 (white bars, n≥5) cows. The geometric mean of GAPDH and RPL-15 was used as an endogenous control. Graphs represent a fold change in mRNA abundance with Low A4 set as control (1). Data for NLRP5 and DPPA3 were log transformed to meet normal distribution assumptions. The mean ± SEM normalized values are presented. *, P≤0.05.

Figure 6. High A4 Cows Tend to have Reduced Numbers of Tertiary Follicles per Section.

Follicles were classified and staged based on morphology (B) and counted on a per section basis to determine the effect of cow A4 classification on follicle growth and development. There was no difference in the number of primordial to secondary follicles (stage 0–3) between Low or High A4 cows (A). However, Low A4 cows tended to have a greater number of tertiary follicles (stage 4) compared with High A4 cows (A). ^†, P = 0.06.