Between-female variation in house sparrow yolk testosterone concentration is negatively associated with CYP19A1 (aromatase) mRNA expression in ovarian follicles

Maternally-derived yolk androgens influence the development and long-term phenotype of offspring in oviparous species. Between-female variation in the amounts of these yolk androgens has been associated with a number of social and environmental factors, suggesting that the variation is adaptive, but the mechanisms behind it are unknown. Using two different approaches, we tested the hypothesis that variation in yolk androgen levels across individuals is associated with variation in their capacity to synthesize androgens. First, we injected female house sparrows with exogenous gonadotropin-releasing hormone (GnRH) to maximally stimulate ovarian steroidogenesis. Second, we collected pre-ovulatory follicle tissue and quantified the mRNA expression of four key enzymes of the steroid synthesis pathway: steroidogenic acute regulatory protein (StAR), cytochrome P450-side chain cleavage enzyme (CYP11A1), 17β-hydroxysteroid dehydrogenase (HSD17B1), and aromatase (CYP19A1). Thirty minutes after GnRH injection, androgen concentrations in both the plasma and in the yolks of pre-ovulatory follicles were significantly elevated compared to controls. However, this measure of steroidogenic capacity did not explain variation in yolk testosterone levels, although physiological differences between house sparrows and more widely studied poultry models were revealed by this approach. Steroidogenic enzyme mRNA levels were detectable in all samples and were significantly lower in the most mature pre-ovulatory follicles. Of the four measured genes, CYP19A1 expression exhibited a significant negative relationship with yolk testosterone concentrations in laid eggs, revealing a key mechanism for between-female variation in yolk testosterone. Furthermore, this suggests that any factors which alter the expression of CYP19A1 within an individual female could have dramatic effects on offspring phenotype.

► Follicle expression of CYP19A1 is negatively associated with yolk testosterone. ► Steroidogenic enzyme mRNA expression decreases with follicle maturation. ► GnRH injection increased yolk testosterone levels in pre-ovulatory follicles. ► Plasma testosterone response to GnRH did not explain yolk testosterone variation. ► Yolk androgens appear to decline after ovulation and before oviposition.