Leptin receptor co-expression gene network moderates the effect of early life adversity on eating behavior in children

Abstract

Leptin influences eating behavior. Exposure to early adversity is associated with eating behaviour disorders and metabolic syndrome, but the role of the leptin receptor on this relationship is poorly explored. We investigated whether individual differences in brain region specific leptin receptor (LepR) gene networks could moderate the effects of early adversity on eating behavior and metabolism. We created an expression-based polygenic risk score (ePRS) reflecting variations in the function of LepR gene network in prefrontal cortex and hypothalamus to investigate the interactions between a cumulative index of postnatal adversity on eating behavior in two independent birth cohorts (MAVAN and GUSTO). To explore whether variations in the prefrontal cortex or hypothalamic genetic scores could be associated with metabolic measurements, we also assessed the relationship between LepR-ePRS and fasting blood glucose and leptin levels in a third independent cohort (ALSPAC). We identified significant interaction effects between postnatal adversity and prefrontal-based LepR-ePRS on the Child Eating Behavior Questionnaire scores. In MAVAN, we observed a significant interaction effect on food enjoyment at 48 months (β = 61.58, p = 0.015) and 72 months (β = 97.78, p = 0.001); food responsiveness at 48 months (β = 83.79, p = 0.009) satiety at 48 months (β = −43.63, p = 0.047). Similar results were observed in the GUSTO cohort, with a significant interaction effect on food enjoyment (β = 30.48, p = 0.006) food fussiness score (β = −24.07, p = 0.02) and satiety score at 60 months (β = −17.00, p = 0.037). No effects were found when focusing on the hypothalamus-based LepR-ePRS on eating behavior in MAVAN and GUSTO cohorts, and there was no effect of hypothalamus and prefrontal cortex based ePRSs on metabolic measures in ALSPAC. Our study indicated that exposure to postnatal adversity interacts with prefrontal cortex LepR-ePRS to moderate eating behavior, suggesting a neurobiological mechanism associated with the development of eating behavior problems in response to early adversity. The knowledge of these mechanisms may guide the understanding of eating patterns associated with risk for obesity in response to fluctuations in stress exposure early in life.

Expression based polygenic scores - A gene network perspective to capture individual differences in biological processes

“…We first describe the methods for the identification of tissue-specific gene networks, which are the essential feature of the ePGS approach. Researchers can use both co-expression 6–16 and differential expression 17 data, from publicly available or their own datasets, see Supplementary Figure 2 . Publicly available data sets include RNAseq databases for both rodents (e.g.…”

“…Importantly, these analyses use a variety of measures of environmental quality and phenotypic outcomes. For example, De Lima et al (2022) described that PFC ePGS based on the leptin receptor gene moderated the effect of postnatal adversity on child eating behaviour 41 . This was an example of a hypothesis-driven analysis based on prior knowledge of leptin receptor activity in appetite regulation.…”

“…The possibility of reflecting functional genomics information in a tissue specific manner is one of the strengths of the ePGS, demonstrated here by the uniqueness of the SLC6A3 PFC gene network in comparison to the SLC6A3 Striatum gene network. As a consequence of these above-mentioned features, the ePGS is suited to test gene by environment effects, evidenced by previous published studies 9,16,42–44 .…”

Effects of prenatal maternal immune activation and exposure to circadian disruption during adolescence: exploring the two-hit model of neurodevelopmental disorders

“…The topological properties of the gene network were calculated using the CytoscapeSoftware® data. As previously described, 'degree' was calculated by summing the number of adjacent nodes, which describes the local topology, and 'betweeness' considers the number of shortest paths linking two nodes and passing through a node n (49). Nodes with 'degree' values higher than +1 SD above the mean were considered hubs, and nodes with 'betweenness' values higher than +1 SD above the mean were considered bottlenecks.…”

Striatal dopamine gene network moderates the effect of early adversity on the risk for adult psychiatric and cardiometabolic comorbidity

“…The ePGS was calculated considering genes co-expressed with the SLC6A3 gene in the striatum. As described previously [41][42][43][44][45][46][47][48] , we began by using brain region-specific RNA sequencing data from mice available at GeneNetwork (http://genenetwork.org/, HBP Rosen Striatum M430V2 (Apr05) RMA Clean) 49 to identify SLC6A3 co-expressed genes (absolute value of co-expression correlation with SLC6A3 gene greater or equal to r= 0.5). GeneNetwork was used to obtain gene expression from rodents since our previous findings demonstrated multiple effects of early life adversities, especially poor fetal growth, on dopaminergic mesocorticolimbic system in rodents [28][29][30][50][51][52][53][54] .…”