Lack of Support for the Genes by Early Environment Interaction Hypothesis in the Pathogenesis of Schizophrenia

Abstract

Ursini et al reported recently that the liability of schizophrenia explained by a polygenic risk score (PRS) derived from the variants most associated with schizophrenia was increased 5-fold in individuals who experienced complications during pregnancy or birth. Follow-up gene expression analysis showed that the genes mapping to the most associated genetic variants are highly expressed in placental tissues. If confirmed, these findings will have major implications in our understanding of the joint effect of genes and environment in the pathogenesis of schizophrenia. We examined the interplay between PRS and obstetric complications (OCs) in 5 independent samples (effective N = 2110). OCs were assessed with the full or modified Lewis-Murray scale, or with birth weight < 2.5 kg as a proxy. In a large cohort we tested whether the pathways from placenta-relevant variants in the original report were associated with case-control status. Unlike in the original study, we did not find significant effect of PRS on the presence of OCs in cases, nor a substantial difference in the association of PRS with case-control status in samples stratified by the presence of OCs. Furthermore, none of the PRS by OCs interactions were significant, nor were any of the biological pathways, examined in the Swedish cohort. Our study could not support the hypothesis of a mediating effect of placenta biology in the pathway from genes to schizophrenia. Methodology differences, in particular the different scales measuring OCs, as well as power constraints for interaction analyses in both studies, may explain this discrepancy.

Obstetric Complications and Polygenic Risk Score: Which Role in Predicting a Severe Short-Term Outcome in Psychosis?

“…Furthermore, no evidence of gene–environment interaction was found in either group, remitting and non-remitting course. This finding is line with a recent large-scale study, which included our sample, that did not find evidence of a role of SZ-PRS caused by the interaction of OCs in the prediction of schizophrenia case-control [ 57 ]. Our report, although of an exploratory nature, is part of the research aiming to examine gene–environment interactions in general and in relation to a specific phenotype—remission.…”

Environmental Risk Factors for Schizophrenia and Bipolar Disorder and Their Relationship to Genetic Risk: Current Knowledge and Future Directions

“…One study reported a stronger association between a PRS derived from SZ-associated SNPs and SZ in those with OCs compared to those who did not experience OCs, with the most significantly associated SNPs mapping to genes highly expressed in placental tissue (Ursini et al, 2018). However, another study which applied this approach in five independent samples found no evidence that OCs interact with PRS or modulate the effect of the PRS on risk of SZ (Vassos et al, 2021). Therefore, there is currently insufficient evidence that OCs impact the association between SZ and genomic risk.…”

Polygenic association with severity and long-term outcome in eating disorder cases

“…65 As AN most likely is polygenic and may have diverse underlying causes, one may hypothesise that more strongly associated genetic variants (i.e., lower generation p value thresholds) act via different biological pathways than less associated ones. 66 However, ndings are mixed. 65,66 It is anticipated that increasing the sample size of eating disorders GWASs will lead to the identi cation of more AN-associated genetic variants, as seen in other psychiatric disorders such as SCZ, 67 boosting the statistical power, robustness, and clinical utility of the AN PGS.…”

“…66 However, ndings are mixed. 65,66 It is anticipated that increasing the sample size of eating disorders GWASs will lead to the identi cation of more AN-associated genetic variants, as seen in other psychiatric disorders such as SCZ, 67 boosting the statistical power, robustness, and clinical utility of the AN PGS. 27 Analyses like ours should be repeated when PGS with greater statistical power are available.…”