Abstract

Effect of chronic stress on whole blood transcriptome in mice: a systematic meta-analysis of publicly available transcriptional profiling datasets

Flandreau EI, Hagenauer MH, Duan TQ, Bader A, Watson SJ, Akil H
53rd Annual Meeting of the Society for Neuroscience. 2024.

Stress increases the risk for psychiatric disorders and animal models of stress-induced biological changes are important tools in translational research. However, successful translation is impeded by small sample size and lack of directly comparable cross-species measurements. As a meta-analysis, the present study addresses the issue of power. As an accessible tissue, blood allows direct comparison between mouse models and human subjects.Blood microarray and RNA-Seq datasets from rodent studies were systematically identified in Gemma, an open-source database of curated and re-analyzed transcriptional profiling studies. Three microarray datasets (GSE68076, GSE72262, and GSE84185) derived from whole blood samples met inclusion/exclusion parameters. Each study exposed eight-week old mice to chronic stress (5 or 10 days social defeat stress, 6 weeks chronic mild stress, or 8 weeks unpredictable chronic mild stress, respectively). When necessary, we subsetted to the relevant tissue (whole blood) and removed outlier samples and genes lacking variability. The final sample size was n=92 (n=45 Non-Stress/n=47 Stress: GSE68076: n=17/19 male C57Bl6 mice (1 outlier per condition eliminated), GSE72262: n=12/12 female C57Bl6/j mice, GSE84185: n=16/16 male Balb/c mice). Differential expression related to chronic stress in each dataset was quantified using the Limma pipeline followed by empirical Bayes moderation. Differential expression results were aligned across datasets by mouse Entrez Gene ID. For genes included in all three studies, we ran a meta-analysis of Log2FC values (with accompanying sampling variance) using a random effects model, and corrected for false discovery rate (FDR). Functional patterns in results were assessed with fast Gene Set Enrichment Analysis using the Brain.GMT (v.2) gene set database.Of 9,219 commonly identified genes, 23 transcripts were downregulated and 16 upregulated in stress-exposed mice (FDR<0.05). Enrichment analysis identified down-regulation in gene sets related to leukocytes, lymphocytes, T cells, B cells, and immune function, DNA/chromatin regulation, RNA processing and translation, cellular metabolic processes, and mitochondrial function and organelles. Upregulated gene sets related to erythrocytes and oxygen binding, synapses, cell junctions, transport, and cell signaling. Our results implicate specific genes and pathways in the widespread immune dysregulation observed following chronic stress and identify stress-induced changes in whole blood transcriptome as a potential tool for improved translation across species.