Analysis of gene expression and DNA methylation in a cognitive phenotype model of schizophrenia

The underpinning hypothesis of this study is that the environmental insults implicated in schizophrenia cause epigenetic changes that trigger deleterious gene expression, resulting in deviations from normal neurodevelopment. The behavioural abnormalities in schizophrenia can be grouped into the three common classes of symptoms: positive, negative, and cognitive. Cognitive symptoms are symptoms that impair cognitive processing and have detrimental effects on individuals with schizophrenia. Maternal immune activation refers to a rat model that stimulates a maternal immune system with an infection or infectious-like stimulus resulting in adverse phenotypes. A cognitive phenotype, maternal immune activation (MIA) model of schizophrenia was employed to use epigenetic markers to discover what deleterious genes drive the cognitive deficits phenotype. Previous work has discerned many changes in gene expression that are implicated in schizophrenia. A hypothesis-driven approach was utilized to determine whether previously studied candidate genes are relevant in the cognitive symptoms of schizophrenia in this cognitive-phenotype model. It was found that prenatal treatment of lipopolysaccharide (LPS) (which is the major outer membrane component of gram-negative bacteria and mimics bacterial infection) on prenatal day 10 and 11 led to changes in mRNA levels in the prefrontal cortex of adolescent rats. Typically, an increase in the amount of transcript in the LPS condition compared to the saline condition, or a greater variability in the amount of transcript between replicates in the LPS condition than the saline condition, was observed. Statistical analysis revealed that these changes did not met statistical significance. To build towards a whole genome DNA methylation analysis, two discrete approaches were used. The first utilized bisulfite modification and investigated changes in candidate genes as a precursor to genome-wide BS-sequencing. DNA methylation was measured across CpG rich regions and an absence of DNA methylation was detected in these regions in both the LPS and saline conditions in the candidate genes. The second approach utilized a long-read sequencing platform to establish the feasibility of a bisulfite conversion-free method for whole-genome DNA methylation approach within our lab. Through the establishment of this method factors that affect the reliability, quality, and accuracy of the final sequencing product were explored. Many of which were in the downstream-from-sequencing, data analysis component of the process. Discoveries were also made regarding how much data would be needed to make direct DNA methylation detection feasible. The data presented here demonstrated that the cognitive-phenotype MIA model had altered gene expression correlating with previously measured behavioural cognitive deficits in the prefrontal cortex in genes that were known to be associated with schizophrenia. To extend this further, a whole genome approach would be needed to discover novel drivers of the phenotype. In the current study, headway was made towards the development and establishment of a whole genome DNA methylation detection method to further this continued aim