A MESOCORTICOLIMBIC INSULIN RECEPTOR GENE CO-EXPRESSION NETWORK MODERATES THE ASSOCIATION BETWEEN EARLY LIFE ADVERSITY AND FOOD APPROACH EATING BEHAVIOUR STYLE IN CHILDHOOD.

Abstract

Insulin receptors, located in brain regions associated with reward sensitivity and decision-making, facilitate insulin action in the brain, modulating intracellular signaling cascades, gene expression, and neural activity. Here, we tested if variations in the expression of the insulin receptor gene network in the prefrontal cortex (PFC) and striatum (STR) moderates the association between early life adversity and eating behaviour in childhood and if this moderation is sex-specific. Participants from the Maternal Adversity, Vulnerability and Neurodevelopment (MAVAN) and Basal Influences on the Baby´s Development (BIBO) were included as two independent cohorts. A biologically-informed polygenic score reflecting functional variation of the mesocorticolimbic insulin receptor gene network was created by using insulin receptor co-expression data from the PFC and STR in mice, and validated in humans through filtering by homologous expression in PFC using well-known databases. Early life adversity exposure was measured as a composite score. Eating behaviour was characterized using the Child Eating Behaviour Questionnaire administered to mothers of children aged 4 and 6 years in MAVAN, and 6 years in BIBO. We found that only in those with high expression of the mesocorticolimbic insulin receptor gene network a higher early adversity score associated with a higher desire to drink in 4-year boys and 6-year girls, as well as a higher food approach score and food approach/food avoidance ratio in 4-year girls. Also, a higher early life adversity was associated with higher food responsiveness, food approach score and food approach/food avoidance ratio at 6 years in the MAVAN full sample. The moderation observed on desire to drink was partially found in BIBO children aged 6 years. Identifying individual differences in response to early adversity may help to prioritize individuals at high risk for long-term disease and to design suitable interventions.