m6A Regulator-Mediated RNA Methylation Modification Patterns are Involved in the Pathogenesis and Immune Microenvironment of Depression

Abstract

Depression is a genetical disease characterized by neuroinflammatory symptoms and is difficult to diagnose and treat effectively. Recently, modification of N6-methyladenosine (m6A) at the gene level was shown to be closely related to immune regulation. This study was conducted to explore the effect of m6A modifications on the occurrence of depression and composition of the immune microenvironment. We downloaded gene expression profile data of healthy and depressed rats from the Gene Expression Omnibus. We described the overall expression of m6A regulators in animal models of depression and constructed risk and clinical prediction models using training and validation sets. Bioinformatics analysis was performed using gene ontology functions, gene set enrichment analysis, gene set variation analysis, weighted gene co-expression network analysis, and protein-protein interaction networks. We used CIBERSORT to identify immune-infiltrating cells in depression and perform correlation analysis. We then constructed two molecular subtypes of depression and assessed the correlation between the key genes and molecular subtypes. Through differential gene analysis of m6A regulators in depressed rats, we identified seven m6A regulators that were significantly upregulated in depressed rats and successfully constructed a clinical prediction model. Gene Ontology functional annotation showed that the m6A regulators enriched differentially expressed genes in biological processes, such as the regulation of mRNA metabolic processes. Further, 12 hub genes were selected from the protein-protein interaction network. Immune cell infiltration analysis showed that levels of inflammatory cells, such as CD4 T cells, were significantly increased in depressed rats and were significantly correlated with the depression hub genes. Depression was divided into two subtypes, and the correlation between hub genes and these two subtypes was clarified. We described the effect of m6A modification on the pathogenesis of depression, focusing on the role of inflammatory infiltration.

Keywords: N6-methyladenosine; biomarker; depression; epigenetics; immune microenvironment.

FIGURE 1

Analysis flow chart.

FIGURE 2

Adjusted for depression data. (A,B): GSE63377 chip data before and after background correction. (C,D): GSE124387 chip data before and after correction. (E,F): GSE86392 gene chip data before and after correction.

FIGURE 3

Overall expression of m6A methylation regulators in animal models of depression. (A) Difference in expression of the m6A regulator between the experimental and control groups. (B) Chromosome localization map of m6A regulators. (C) Correlation network diagram of m6A regulators. (D) Correlation heat map of m6A regulators. (E) Volcano plot of the results of the differential genetic analysis of m6A regulators.

FIGURE 4

Predictive power analysis of neural network models for depression. (A,B) LASSO regression analysis identified key genes for m6A regulators. (C) Receiver operating characteristic (ROC) curve of the neural network model test set; the abscissa represents the specificity and ordinate represents the sensitivity. (D) ROC curve of the validation set of the neural network model; the abscissa represents the specificity and ordinate represents the sensitivity.

FIGURE 5

Functional enrichment analysis of m6A regulator. (A) First 15 items enriched for biological processes, molecular functions, and cellular components the abscissa represents the GO term and ordinate is the -log (adj p-value). Band colors: blue represents downregulation and red represents upregulation. (B) Significantly enriched KEGG pathway; rno03040: spliceosome. (C) m6Aregulation of the first eight items of biological processes; red represents upregulation, and blue represents downregulation. (D) KEGG enrichment results; the outer circle represents the KEGG pathway and inner circle represents the log fold-change size.

FIGURE 6

Gene set enrichment analysis (GSEA) and gene set variation analysis (GSVA) of experimental and control groups. (A) Mountain map of the first five items of the GSEA Gene Ontology (GO) analysis results; the abscissa is normalized enrichment score (NES), ordinate shows the GO terms, size of the mountain represents the number of genes, and color represents the p-value. (B,C) Clustering of the first two items in GSEA GO. (D) Mountain map of the first five items of the GSEA KEGG results; the abscissa is NES, ordinate shows the KEGG terms, size of the mountain represents the number of genes, and color represents the p-value. (E,F) GSEA KEGG of clustering of the first two items. (G) Interaction network diagram of GSEA GO. (H) Interaction network diagram of GSEA KEGG. (I) Heat map of differential expression of GSVA; red represents upregulation and blue represents downregulation.

FIGURE 7

Weighted gene co-expression network analysis (WGCNA). (A) Scale-free network verification graph (R² > 0.8, slope < 0), conforming to the scale-free network standard. (B) Dynamic clipping tree clustering diagram; the abscissa is the clustering module and ordinate is the tree height. (C) TOM network clustering heatmap. (D) Heat map of correlations between WGCNA network modules and depression.

FIGURE 8

m6A regulator protein-protein interaction (PPI) network. (A) PPI network of the m6A regulators; the number of edges indicate the credibility of the evidence. (B) Network diagram of hub genes. (C) Venn diagram of hub genes and m6A regulators.

FIGURE 9

Analysis of immune infiltration in depression. (A) Differential expression of infiltrating immune cells in experimental and control groups. (B) Overall expression of infiltrating immune cells. (C) Correlation heat map of hub genes and immune-infiltrating cells.

FIGURE 10

Relevant molecular subtypes and correlations of depression. (A) Cumulative distribution function (CDF) curve of consensus clustering of depression-related molecules; the abscissa is the consensus index and the ordinate is the CDF index. (B) Relative change in the area under the CDF curve; the results show that it is divided into two types, and the change trend is the most stable. (C) Cluster heat map of depression-associated molecular subtypes. (D) Principal coordinate analysis plot of depression-related molecular subtypes. (E) Heat map of the correlation between hub genes and isoform 1. (F) Heat map of the correlation between hub genes and isoform 2.