F93. CELL-TYPE SPECIFIC TRANSCRIPTOMIC PROFILING IN SCHIZOPHRENIA IDENTIFIES CHANGES IN GABAERGIC NEURONS AND OLIGODENDROCYTES AT TRANSCRIPT LEVEL

Schizophrenia is a complex neuropsychiatric disorder which affects approximately 1% of the population. GWAS has enabled the discovery of risk genes associated with schizophrenia while functional studies have explored alterations in gene expression and regulation. There is an increasing focus on studying the molecular mechanisms of this disorder at cell-type resolution. Cell-type specific gene expression changes in schizophrenia are largely unexplored, particularly at the transcript level. As such, we aimed to investigate disease-specific changes in gene expression across a range of cell-types. RNA-seq was carried out to profile gene expression in prefrontal cortex tissue from schizophrenia cases (n=50) and controls (n=50). For each individual, four cell-types were isolated via fluorescence activated nuclear sorting (FANS), including GABAergic neurons (neuN+/sox6+), glutamatergic neurons (neuN+/sox6-), oligodendrocytes (neuN-/sox10+) and microglia/astrocytes (neuN-/sox10-). Differential analysis was carried out using DREAM and remaCor. MAGMA was used to test differentially expressed genes and transcripts for enrichment of common genetic variants associated with schizophrenia while gene-set enrichment analysis was used to identify perturbed pathways and biological processes. The cell-type specific profiles were also used as a reference to identify cell-type proportions in bulk RNA-seq data (n=870 samples) and subsequently impute expression profiles for the four cell-types using bMIND. Differential analysis at both gene and transcript levels yielded marked differences in terms of genes implicated. A large majority of the significant genes identified in the transcript analysis were not observed in the gene level analysis, indicating the need to study differences in transcript expression at cell-type resolution. Different isoforms of KMT5A, a known schizophrenia risk gene, were implicated in GABAergic neurons and oligodendrocytes, and were not observed in traditional gene or transcript level analysis. CACNA1C isoforms were also differentially expressed only in oligodendrocytes, again highlighting the need to explore these changes at the cell-type level. Enrichment analysis for the gene sets identified in both analyses returned differing results with the gene-level analysis implicating synaptic density and dysfunction. Cell-type specific imputed gene and transcript expression profiles for 870 individuals were created using the FANS data as a reference. This greatly increased our power to identify disease associated expression changes in comparison to the FANS analysis and enabled a QTL analysis to link expression changes to genetic variants. Overall, these analyses explored altered biology in schizophrenia on a cell-type specific basis and placed particular emphasis on transcript changes which have been understudied at the cell-type level to date. The identification of altered transcript expression may further our understanding of the neurobiology of schizophrenia.