Analysis of transcriptome alterations in schizophrenia

Abstract

Schizophrenia is a major psychiatric disorder characterized by abnormal social behaviour and failure of recognition. A lot of studies are proposed, but underline mechanisms are not revealed yet. In previous studies, schizophrenia is a complex disease caused by multiple genetic and environmental factors. Genetic studies have yielded numerous candidate genes and genetic variations such as DNA, gene expression, viruses, toxins, nutrition, or birth injury. However, most candidates have not been well replicated or validated. Recently, next-generation sequencing technique and retrotransposon related studies reported that retrotransposon activation may contribute to gene expression regulations. In this study, using next-generation sequence analysis, we propose methods that identify robust biomarkers for schizophrenia and reveal the relationship of gene expressions and retrotransposon expressions. Retrotransposons constitute 40% of the human genome. These genetic elements can amplify themselves in a genome through copy and paste mechanism. Usually, retrotransposons are inactivated by hypermethylation, but they are activated in the human brain. Especially, retrotransposons are more activated in schizophrenia patient’s brain than normal. Therefore, the objectives of this study are identifying robust schizophrenia biomarkers and revealing underline mechanism of gene-repeat regulations. We identified immune/inflammation response are significantly related with schizophrenia using gene co-expression network analysis. We validated the results with qRT-PCR and pathological data and we found that the cell-types showed activated immune/inflammation response. Moreover, over expressed retrotransposon RNAs contribute to down-regulations of gene expression if the strand directions of retrotransposon and gene are same. We identified a co-expression module associated with schizophrenia that includes the majority of differentially expressed genes related to immune/inflammation response as well as with the density of parvalbumin-containing neurons in the hippocampus. Moreover, we found that over expressed intronic repeats down-regulate the gene expressions in schizophrenia. The results indicate that abnormal immune/inflammation response in the hippocampus may underlie the pathophysiology of schizophrenia and may be associated with abnormalities in the parvalbumin-containing neurons that lead to the cognitive deficits of the disease.