Early-life exposure to air pollution alters resting-state functional connectivity patterns in late childhood

BackgroundAdverse environmental exposures such as air pollution can disrupt brain development, contributing to long-term risk of neurodevelopmental and psychiatric disorders. Resting-state functional connectivity (rsFC) captures large-scale brain alterations seen in such disorders, and growing evidence links air pollution exposure to adverse cognitive and mental health outcomes. MethodsWe used connectome-based predictive modeling (CPM), a data-driven approach that uses brain states to predict individual variables, in this case, air pollution exposures, to investigate whether individual patterns of rsFC at 10-13 years in children from the NeuroSmog study are associated with prior exposure to particulate matter (PM10) and nitrogen dioxide (NO2). The tested exposure periods comprised prenatal, from birth until the 2nd birthday, 2nd to 4th birthday, the year prior to MRI data acquisition, and the week prior to MRI data acquisition. ResultsFunctional connectivity at 10-13 years successfully predicted exposures to PM10 in two subsequent early-life periods, from birth until the age of 2, and at the age of 2-4. Predictions for other exposure periods, as well as for all tested periods of exposure to NO2, were not significant. Higher PM10 exposures in both periods were associated with a decrease in network segregation and heightened functional connectivity, in particular between the default mode, ventral attention, cerebellar, and salience networks. ConclusionsSince segregation between brain networks normally decreases with age, our findings suggest that early-life exposure to PM10 may disrupt this developmental trajectory. Exposures during a period of increased neuroplasticity can thus exert enduring effects on functional network architecture.