Effects of Prenatal Environment on Phenotype Are Revealed by Postnatal Challenges: Embryonic Hormone Exposure, Adrenocortical Function, and Food in Seabird Chicks

The interaction between prenatal environments and postnatal environments is an important source of phenotypic variability. We examined the ability of prenatal steroid exposure and postnatal energy restriction to explain adrenocortical function and fledging age in captive seabird chicks. We proposed and tested two hypotheses: (1) the strength of prenatal effects is attenuated by challenging postnatal environments (postnatal override) and (2) the strength of prenatal effects increases with the severity of postnatal challenges (postnatal reveal). We reared common murre (Uria aalge) chicks and measured prenatal exposure to corticosterone (CORT) and testosterone (T) from allantoic waste. Adrenocortical function was assessed after 10 d of ad lib. feeding and then after 5 and 10 d on controlled diets. Postnatal override predicts that prenatal steroids will explain more phenotypic variation before implementation of energy restriction; postnatal reveal predicts that the contribution of prenatal steroids will increase with duration and severity of energy restriction. Energy restriction increased secretion of baseline CORT and the adrenocortical response to the standardized stressor of handling and restraint. The ability of prenatal steroids to explain baseline CORT increased with duration of energy restriction, and for day 20 free baseline CORT, there was a significant interaction between kilo joules per day and prenatal CORT levels; severity of restriction strengthened the relationship between prenatal hormone levels and postnatal hormone levels. Both maximum CORT at day 20 and fledging age were best explained by diet treatment and day 15 or day 20 baseline CORT, respectively. Overall, prenatal CORT increased fledging age and baseline secretion of CORT, while prenatal T decreased them. However, prenatal effects on adrenocortical function were apparent only under the energy restriction conditions. Thus, we found some support for the postnatal reveal hypothesis; our results suggest that some prenatal effects on phenotype may be more likely to manifest in challenging postnatal environments