Abstract

Neurotransmission-related gene expression in the frontal pole (Brodmann area 10) is altered in subjects with schizophrenia and bipolar disorder

Thew Forrester LC, Medina A, Hagenauer MH, Krolewski DM, Waselus M, Hughes E, Vawter MP, Walsh DM, Richardson E, Turner CA, Bunney WE, Myers RM, Barchas JD, Lee F, Schatzberg AF, Akil H, Watson SJ
50th Annual Meeting of the Society for Neuroscience, Virtual. 2021.

Abstract

BACKGROUND: Schizophrenia (SCZ) and Bipolar Disorder (BP) have long been associated with dysregulation in the frontal cortex. The gene expression underlying this dysregulation has been explored in detail using transcriptional profiling studies (microarray, RNA-Seq) in the dorsolateral prefrontal cortex, but the larger frontal pole region (Brodmann Area 10, BA10) has received less attention. Our study explored the association between SCZ and BP and the frontal pole using methods that provide a sensitive snapshot of neurotransmission-related gene expression.

METHODS: To ensure a high concentration of neuron-derived transcripts, post-mortem dissections focused on grey matter. Only samples with agonal factor=0, pH>6.5, and RNA integrity>6.0 were included to exclude brains with prolonged hypoxia before death. Gene expression was measured using two sets of TaqMan microfluidic qPCR cards targeting transcripts related to signaling via 1) GABA and glutamate (“GABAGLU” experiment), 2) dopamine and serotonin (“DA5HT” experiment). In each experiment, the subject samples were run on at least two separate cards. After quality control (QC), the sample size was 69 subjects (control (CTRL): n=26, SCZ: n=22, BP: n=21) and included measurements from 111 targeted gene transcripts (GABAGLU: 82, DA5HT: 29). We normalized measurements using a panel of reference gene transcripts (GABAGLU: 11 genes, DA5HT: 17 genes). Differential expression (DE) was assessed via a multi-level model that included biological (gender, age, pH, post-mortem interval [PMI]) and technical (RNA concentration, integrity, card) co-variates. The results from the top DE genes (false discovery rate <0.10) were explored in relation to symptom profiles, medication, and drug and alcohol exposure. These results were compared to the results from two publicly available microarray datasets examining the influence of psychiatric diagnosis on gene expression in BA10 (GSE12654; GSE17612). We re-analyzed both datasets using a custom chip definition file with updated annotation. Our DE models controlled for large sources of noise as determined by relationships with top principal components of variation (pH, age, RNA degradation, PMI, rate of death). After QC, the sample size for GSE12654 was 50 subjects (CTRL: n=15, SCZ: n=13, BP: n=11, Depression: n=11); and for GSE17612 the sample size was 51 subjects (CTRL: n=23, SCZ: n=28).

RESULTS AND CONCLUSION: Collectively, our results showed a similar pattern of DE gene expression in the frontal pole in both disorders, including genes directly involved in major neurotransmitter functions affecting the serotonergic, dopaminergic and GABAergic systems.