Genetic overlap between idiopathic scoliosis and schizophrenia in the general population

Abstract

Introduction: Adolescent idiopathic scoliosis (AIS) and schizophrenia (SCZ) are two distinct conditions with poorly understood aetiologies that both emerge in otherwise healthy young adolescents. One rare genetic condition associated with both phenotypic outcomes is the 22q11.2 deletion (22q11DS). This microdeletion, encompassing 47 genes, occurs in approximately 1 in 2,148 live births and confers a 20-fold higher risk for both AIS and schizophrenia compared to the general population. In the general population (non-22q11DS carriers), AIS and SCZ have also been reported to be related and genetic studies suggest the involvement of genetic variants implicated in the central nervous functioning. In this study, our objective was to further investigate genetic overlaps between these conditions in the general population. Specifically, we aimed to explore the role of genes within the 22q11.2 region, not only in terms of common variants but also their potential impact on gene networks and biopathways. Methods: We used summary statistics from three genome-wide association studies (GWAS): two focused on AIS (n = 11,210), and one on schizophrenia (n = 36,989). To explore potential overlaps between the two conditions, we conducted a comparative analysis on the significance-based ranked single nucleotide polymorphisms (SNPs) that are associated with both AIS and SCZ. Next, we employed in silico analyses to assess gene-networks enrichment for the most significant SNPs and investigate the contribution of genes within the 22q11.2 region. Post-hoc analysis was conducted to explore the biological pathways correlated with SNPs significantly associated with both AIS and SCZ. Results: The in silico analyses revealed a significant (adjusted-p < 0.05) genetic overlap between SCZ and both AIS cohorts. The top 3% of the most significant SNPs associated with both conditions exhibited a distinct enrichment cluster which is unlikely to be a result of chance (p < 3e-04). The gene-networks analyses showed a significant overlap of 26–41% with the ones involving genes in the 22q11DS region. However, there was no overlap between SNPs in this region and the most significant SNPs identified in the GWAS. Conclusion: This study revealed compelling evidence that beyond the shared association with 22q11DS as a rare genetic variant, AIS and SCZ exhibit common genetic risk variants and an overlap of important genes. The gene networks enriched by the most significant SNPs for both conditions also intersect with the ones involving genes in the 22q11DS region. However, SNPs within this region were not overrepresented among the most significant SNPs from GWAS for both conditions. Notably, gene networks linked to the risk for both conditions suggest an involvement of biopathways related to cellular signaling and neuronal development.