Abstract
OBJECTIVE: Previous studies have shown that both pre- and post-natal adversities, the latter including exposures to stress during childhood and adolescence, explain variation in structural properties of white matter (WM) in the brain. While previous studies have examined effects of independent stress exposures within one developmental period, such as childhood, we examine effects of stress across development using data from a prospective longitudinal study. More specifically, we ask how stressful events during prenatal development, childhood, and adolescence relate to variation in WM properties in early adulthood in young men recruited from a birth cohort. METHOD: Using data from 393 mother-son pairs from a community-based birth cohort from England (Avon Longitudinal Study of Parents and Children), we examined how stressful life events relate to variation in different structural properties of WM in the corpus callosum and across the whole brain in early adulthood in men aged 18-21 years. We distinguish between stress occurring during three developmental periods: a) prenatal maternal stress, b) postnatal stress within the first four years of life, c) stress during adolescence (age 12-16 years). To obtain a comprehensive quantification of variation in WM, we assess structural properties of WM using four different measures, namely fractional anisotropy (FA), mean diffusivity (MD), magnetization transfer ratio (MTR) and myelin water fraction (MWF). RESULTS: The developmental model shows that prenatal stress is associated with lower MTR and MWF in the genu and/or splenium of the corpus callosum, and with lower MTR in global (lobar) WM. Stress during early childhood is associated with higher MTR in the splenium, and stress during adolescence is associated with higher MTR in the genu and lower MD in the splenium. We see no associations between postnatal stress and variation in global (lobar) WM. CONCLUSIONS: The current study found evidence for independent effects of stress on WM properties during distinct neurodevelopmental periods. We speculate that these independent effects are due to differences in the developmental processes unfolding at different developmental time points. We suggest that associations between prenatal stress and WM properties may relate to abnormalities in neurogenesis, affecting the number and density of axons, while postnatal stress may interfere with processes related to myelination or radial growth of axons. Potential consequences of prenatal glucocorticoid exposure should be considered in obstetric care.