Network State of Mind : Gene expression profiling of neuropsychiatric traits in human and mouse

Neuropsychiatric disorders like schizophrenia are likely caused by a large number of genes with a small effect that are difficult to identify using traditional genetic association studies. Therefore, we have applied system biology approaches in order to examine patterns of gene expression in healthy tissue and neuropsychiatric disorders. In the first chapters of this thesis we describe patterns of gene expression in healthy tissue of human and mouse. We first observe in two inbred strains of mice that two brain regions, amygdala and hippocampus, show strong differences in expression. Effects related to genetic background are for a large part due to a hybridization artefact. Next, we study the effect of common genetic variation on expression in blood and brain of healthy human subjects. This inversion on chromosome 17q21.31 has been associated with higher levels of MAPT, a gene linked with neurodegenerative diseases. However, we find the expression of other genes near the inversion to be influenced as well, in a tissue-specific manner. Finally, we study genetic and epigenetic regulation of gene expression and determine causality between SNPs, DNA methylation and gene expression in whole blood of healthy human subjects. We find that both the direction of association and the causal relationships are more complex than initially expected. In the second part of this thesis we examine gene expression profiles in relation to neuropsychiatric disorders. We first use gene expression in a population of mice to narrow down a region on chromosome 15 previously found to be involved in exploration behavior. We identify a gene, Ly6a, of which expression is both regulated by our region of interest and is part of a gene-network associated with exploration behavior. The behavioral phenotype of a Ly6a knock-out strain suggests functional involvement of this gene. Next, we examine an human whole blood schizophrenia sample. We find a gene-network related to schizophrenia, independent of possible confounding effects of antipsychotic medication. In addition, we find this network to be enriched for brain-expressed genes. Also, the results suggest that the MHC region, a region implicated in schizophrenia before, may increase disease susceptibility via altered gene expression of regulatory genes in this network. In the next study we aimed to complement findings from a large schizophrenia meta-analysis. We identified regulation of gene expression by the top 6,000 SNPs of this analysis. We subsequently tested these for differential expression in schizophrenia and identified new potential candidate genes. Finally, we study gene expression profiles of several human brain regions. A combination of system biology approaches allowed us to both discriminate brain regions based on their expression profiles and identify a gene-network in the cerebellum different between patients with schizophrenia or bipolar disorder and unaffected controls. These results suggest that genetic pathology of neuropsychiatric disorders can be different across brain regions and cell types. In conclusion, neuropsychiatric disorders are complex traits and the vast increase in the amount of available genomic information makes systems biology approaches indispensible to further unravel the network state of mind.