Genetic Association Between Schizophrenia and Cortical Brain Surface Area and Thickness

Abstract

Importance: Schizophrenia is a complex heritable disorder associated with many genetic variants, each with a small effect. While cortical differences between patients with schizophrenia and healthy controls are consistently reported, the underlying molecular mechanisms remain elusive.

Objective: To investigate the extent of shared genetic architecture between schizophrenia and brain cortical surface area (SA) and thickness (TH) and to identify shared genomic loci.

Design, setting, and participants: Independent genome-wide association study data on schizophrenia (Psychiatric Genomics Consortium and CLOZUK: n = 105 318) and SA and TH (UK Biobank: n = 33 735) were obtained. The extent of polygenic overlap was investigated using MiXeR. The specific shared genomic loci were identified by conditional/conjunctional false discovery rate analysis and were further examined in 3 independent cohorts. Data were collected from December 2019 to February 2021, and data analysis was performed from May 2020 to February 2021.

Main outcomes and measures: The primary outcomes were estimated fractions of polygenic overlap between schizophrenia, total SA, and average TH and a list of functionally characterized shared genomic loci.

Results: Based on genome-wide association study data from 139 053 participants, MiXeR estimated schizophrenia to be more polygenic (9703 single-nucleotide variants [SNVs]) than total SA (2101 SNVs) and average TH (1363 SNVs). Most SNVs associated with total SA (1966 of 2101 [93.6%]) and average TH (1322 of 1363 [97.0%]) may be associated with the development of schizophrenia. Subsequent conjunctional false discovery rate analysis identified 44 and 23 schizophrenia risk loci shared with total SA and average TH, respectively. The SNV associations of shared loci between schizophrenia and total SA revealed en masse concordant association between the discovery and independent cohorts. After removing high linkage disequilibrium regions, such as the major histocompatibility complex region, the shared loci were enriched in immunologic signature gene sets. Polygenic overlap and shared loci between schizophrenia and schizophrenia-associated regions of interest for SA (superior frontal and middle temporal gyri) and for TH (superior temporal, inferior temporal, and superior frontal gyri) were also identified.

Conclusions and relevance: This study demonstrated shared genetic loci between cortical morphometry and schizophrenia, among which a subset are associated with immunity. These findings provide an insight into the complex genetic architecture and associated with schizophrenia.

Figure 1.. Polygenic Comparison of Schizophrenia and Global Cortical Measures in Discovery Study

Bar diagram visualizing the estimated unique and shared variants, shown in thousands. Error bars indicate standard errors. SA indicates surface area; TH, thickness.

Figure 2.. Shared Genetic Loci Between Schizophrenia and Global Cortical Measures in Discovery Study

The y-axis represents –log₁₀–transformed conjunctional false discovery rate (FDR) values for each single-nucleotide variant, and the x-axis reflects the chromosomal position. The dotted horizontal line is the cutoff for significance. Details for genomic loci and candidate single-nucleotide variants are provided in eTable 2 in Supplement 1, Supplement 3, and Supplement 4.

Figure 3.. Expression of Credible Genes of Shared Loci in Brain Tissue

Among the shared loci that were significantly associated with at least 1 phenotype in the independent cohorts, 31 credible genes were mapped by all 3 strategies: positional mapping, expression quantitative trait locus mapping, and chromatin interaction mapping. Corresponding expression patterns in brain tissue were extracted from the Genotype-Tissue Expression Portal. The expression pattern for has-mir-8072 was not found. The KMT5A is also named as SETD8. Transcripts per million (TPM) is used to measure gene expression levels. BA indicates Brodmann area.