Cytoarchitectural abnormalities have been described in the prefrontal cortex of subjects with schizophrenia, bipolar disorder and depression. However, little is known about the gene expression profiles associated with these abnormalities. Genome-wide expression profiling technology provides an unbiased approach to identifying candidate genes and biological processes that may be associated with complex biological traits such as cytoarchitecture. In this study, we explored expression profiles associated with the abnormalities by using publicly available microarray metadata and cytoarchitectural data from post-mortem samples of the frontal cortex from 54 subjects (schizophrenia, n = 14; bipolar disorder, n = 13; depression, n = 12 and controls n = 15). Correlation analysis between genome-wide expression levels and cytoarchitectural traits revealed that 818 genes were significantly correlated with a decrease in the number of perineuronal oligodendrocytes across all subjects. A total of 600 genes were significantly correlated with a decrease in density of calbindin-positive interneurons across all subjects. Multiple biological processes including cellular metabolism, central nervous system development, cell motility and programmed cell death were significantly overrepresented in both correlated gene lists. These findings may provide novel insights into the molecular mechanisms that underlie the cytoarchitectural abnormalities of perineuronal oligodendrocytes and calbindin-containing GABAergic interneurons in the prefrontal cortex of the major psychiatric disorders.

The Stanley Neuropathology Consortium Integrative Database (SNCID, http://sncid.stanleyresearch.org) is a data-mining tool that includes 379 neuropathology data sets from hippocampus, as well as RNA-Seq data measured in 15 well-matched cases in each of four groups: schizophrenia, bipolar disorder (BPD), major depression (MD) and unaffected controls. We analyzed the neuropathology data from the hippocampus to identify those abnormalities that are shared between psychiatric disorders and those that are specific to each disorder. Of the 379 data sets, 20 of them showed a significant abnormality in at least one disorder as compared with unaffected controls. GABAergic markers and synaptic proteins were mainly abnormal in schizophrenia and the two mood disorders, respectively. Two immune/inflammation-related co-expression modules built from RNA-seq data from both schizophrenia and controls combined were associated with disease status, as well as negatively correlated with the GABAergic markers. The correlation between immune-related modules and schizophrenia was replicated using microarray data from an independent tissue collection. Immune/inflammation-related co-expression modules were also built from RNA-seq data from BPD cases or from MD cases but were not preserved when using data from control cases. Moreover, there was no overlap in the genes that comprise the immune/inflammation response-related modules across the different disorders. Thus, there appears to be differential activation of the immune/inflammatory response, as determined by co-expression of genes, which is associated with the major psychiatric disorders and which is also associated with the abnormal neuropathology in the disorders.

The choroid plexus (CP) has a key role in maintaining brain homeostasis by producing cerebrospinal fluid (CSF), by mediating transport of nutrients and removing metabolic products from the central nervous system and by responding to peripheral inflammatory signals. Although abnormal markers of immune response and inflammation are apparent in individuals with schizophrenia, the CP of these individuals has not been characterized. We therefore sequenced mRNA from the CP from two independent collections of individuals with schizophrenia and unaffected controls. Genes related to immune function and inflammation were upregulated in both collections. In addition, a co-expression module related to immune/inflammation response that was generated by combining mRNA-Seq data from both collections was significantly associated with disease status. The immune/inflammation-related co-expression module was positively correlated with levels of C-reactive protein (CRP), cortisol and several immune modulator proteins in the serum of the same individuals and was also positively correlated with CRP, cortisol and pro-inflammatory cytokines in the frontal cortex of the same individuals. In addition, we found a substantial number of nodes (genes) that were common to our schizophrenia-associated immune/inflammation module from the pooled data and a module we generated from lippopolysaccharides-treated mouse model data. These results suggest that the CP of individuals with schizophrenia are responding to signals from the periphery by upregulating immune/inflammation-related genes to protect the brain and maintain the homeostasis but nevertheless fails to completely prevent immune/inflammation related changes in the brain.

Cytoarchitectural abnormalities have been described in the prefrontal cortex (PFC) of subjects with psychiatric disorders. We explored the possible genetic causalities that may underlie the cytoarchitectural abnormalities of calbindin-containing c-aminobutyric acid (GABA)ergic neurons and perineuronal oligodendrocytes in the PFC of subjects with psychiatric disorders by converging results from genome-wide single-nucleotide polymorphism (SNP) scans for the traits and expression SNP (eSNP) associations. In the initial genome-wide scans, we identified several development-and apoptosis-related genes associated with the cytoarchitectural traits. Moreover, the susceptibility gene for bipolar disorder, PPP2R2C, was found to be associated with the number of perineuronal oligodendrocytes. Further eSNP analyses indicated that two novel candidate genes, RAB2A and SLC38A1, were associated with the density of calbindinpositive neurons and the number of perineuronal oligodendrocytes, respectively. Our findings may provide novel insights into the genetic causalities associated with cytoarchitectural abnormalities in the PFC of subjects with major psychiatric disorders as well as into the etiology of such disorders.