Investigating the mechanisms mediating the outcomes of prenatal stress

Abstract

Excessive stress during pregnancy can strongly impact the developing offspring, leading to long-term changes in the brain that persist into adulthood, in a process known as “prenatal programming”. In a rat model, where pregnant dams were exposed to five days of social stress from gestational day (GD) 16 to 20, the prenatally stressed (PNS) offspring exhibit anxious behaviour, cognitive deficits and exaggerated hypothalamic-pituitary-adrenal (HPA) axis responses to stress, with outcomes often dependent on the sex of the offspring. However, the mechanisms mediating these outcomes are not completely understood. Firstly, it is not completely clear what underlies the “programmed” phenotypes in adulthood, and secondly, it is not known how the stress signals are transmitted from the mother to the foetus during gestation to result in foetal “programming”. Steroids are a large class of signalling molecules produced by endocrine organs, and include the glucocorticoids (e.g. corticosterone), sex hormones (e.g. progesterone and testosterone) and neuroactive steroids (such as allopregnanolone). Steroids play key roles in modulating stress responses, and are also involved in coordinating pregnancy adaptations and foetal development, therefore may be a common factor in the mechanisms underlying the “programmed” phenotypes observed in PNS offspring, and in the process of foetal “programming” during pregnancy. In order to investigate a role for steroids in foetal programming, it was first necessary to develop and validate a liquid chromatography-mass spectrometry method (LC-MS) that could reliably quantify a panel of steroids in complex tissues such as the brain. Neuroactive steroids are known to increase following an acute stressor and can rapidly modulate neuronal excitability. It was hypothesised that alterations in neuroactive steroid concentrations in the brain may underlie the “programmed” phenotype, such as dysregulated HPA axis activity, in adult PNS offspring. Using the LC-MS method developed, no obvious differences were found in central neuroactive steroid concentrations between control and PNS offspring at baseline; however, following an acute stressor (swim stress), the production of certain neuroactive steroids which modulate inhibitory neurotransmission seem compromised in the brains of PNS offspring as compared to controls, which could be further investigated with other acute stress paradigms. The possible mechanisms involved in the transmission of stress signals from the mother to foetus during gestation to result in “programming” were then investigated at GD20, focusing on glucocorticoid metabolism and the steroidal milieu in the mother and the foetus. Glucocorticoid overexposure has been proposed as one of the primary mechanisms of foetal programming. However, it was found that despite stressed dams having greater circulating corticosterone, stressed foetuses did not exhibit greater corticosterone concentrations in the brain. The steroidogenic profile was also generally similar in the brains of control and stressed foetuses. The direct contribution of altered corticosterone and neuroactive steroid concentration in “programming” of the foetal brain, and possibly behaviour, is therefore minimal. However, stressed placentae, exhibited greater mRNA expression for 11βhydroxysteroid dehydrogenase 2 (11β-HSD2), which inactivates corticosterone, indicating that the placenta plays an active role in regulating the transmission of gestational stress signals from the mother to the foetus during late pregnancy. The role of the placenta was then further investigated, where it was hypothesised that placental oxidative stress may play a role in foetal programming. Elevated oxidative stress status was observed in the placentae of stressed pregnancies at GD20, which was prevented by the administration of an antioxidant drug to the pregnant dam before the onset of chronic social stress. Notably, this maternal antioxidant treatment, which targets the maternal compartment and placenta but does not reach the foetus, rescued the anxiety phenotype of PNS offspring when examined in adulthood. Several other physiological markers in the brain linked to anxiety and inhibitory neurotransmission, which were altered in PNS offspring as compared to controls, were also normalised as a result of the maternal antioxidant treatment, and the effects were sex specific. Additionally, both secretions from placental explants and foetal plasma from stressed pregnancies contained unknown factors that altered neuronal growth in vitro, but this was not observed in stressed pregnancies with antioxidant treatment. Together, this suggests that the stressed placenta may transmit stress signals to the foetus through secreting, as yet unidentified, damaging “factors”. Hence, targeting placental oxidative stress during pregnancy could provide a therapeutic option to prevent the transmission of such signals, and therefore the adverse outcomes observed in the offspring later in life.