Early life adversity and social withdrawal behavior - identifying risk for adult psychopathology using a translational model and expression-based polygenic scores

Social withdrawal behavior has been linked to internalizing problems and increased risk of anxiety disorders and depression later in life. It is known that inhibition of pyramidal neurons in the ventral hippocampus impairs normal social behavior (Murugan et al., 2017; Okuyama et al., 2016; Brumback et al., 2017). To identify the biological underpinnings of social withdrawal behavior, we used ventral hippocampus RNA sequencing data from rats that naturally varied in their willingness to interact with a co-specific. Data was processed through weighted gene co-expression network analysis (WGCNA), and we identified one co-expression network that significantly correlated with social interaction behavior. We used functional enrichment analysis (Metacore®, Clarivate Analytics, https://portal.genego.com) to identify the significant gene ontology processes associated with the genes comprising this network. The network was associated with gene ontology processes related to immune system response and cytokine production. Human ortholog genes from these networks were used to inform the calculation of expression-based polygenic scores (ePRS) in adolescents (ALSPAC) and adults (UK Biobank). Variations in these ePRS scores were significantly associated with mood disorders in adults and emotional symptoms in adolescents in response to early life adversity. To test the specificity of ePRS for interacting with early adversity on these outcomes, we additionally calculated the traditional polygenic risk scores using the major depression GWAS 2018 (Wray, Ripke et al. 2018) and anxiety GWAS 2020 (Levey, Gelernter et al. 2020). PRSs were calculated to be comparable in size to ePRS in terms of SNPs for each cohort (ASLPAC and UK Biobank). Testing both PRSs in models of gene x environment interaction, no significant effects were observed. Functional genomics translational tools like the one used in this study can help in the identification of individual differences in child behavior that map onto adult psychopathology, especially in response to early adversity.