Identification of predictors for bacterial meningitis diagnosis based on transcriptomics and genetic analysis

Abstract

Bacterial meningitis (BM) requires rapid intervention, especially in immunocompromised populations. Understanding early immune responses is crucial, as they precede clinical symptoms; however, comprehensive studies remain limited. This research investigates immune-related genes to improve BM diagnosis and treatment. Mendelian randomization, differential gene expression analysis, and co-expression network analysis identified key genes associated with BM. Immune cell ratio calculations and infiltration analyses demonstrated altered immune cell proportions. Spearman correlation analysis revealed relationships between gene expression and immune cell types. Single-cell RNA sequencing, gene set enrichment analysis, and pseudotime analysis explored changes in gene expression and cell proportions across disease stages, focusing on the roles of key genes in specific immune cells. Ring Finger Protein 144B (RNF144B) was identified as a risk gene predominantly expressed in monocytes and neutrophils. Conversely, FYN Proto-Oncogene (FYN) was identified as a protective gene primarily associated with NKT cells. During BM onset, increased RNF144B expression positively correlated with elevated neutrophil levels, while reduced FYN expression correlated with decreased NKT cell levels. During remission and recovery, RNF144B expression and neutrophil proportions decreased, whereas FYN expression and NKT cell proportions increased. NKT cells appeared to play a protective role, with FYN potentially modulating T-cell receptor function in these cells, thereby reducing BM risk. RNF144B and FYN expression exhibit opposing trends in peripheral blood across BM stages, suggesting their potential as biomarkers for diagnosis and monitoring. These findings provide a valuable reference for early intervention strategies and personalized treatment approaches tailored to specific disease stages in the clinic.

Keywords: Bacterial meningitis; FYN proto-oncogene; Gene expression; Immune cell infiltration; RING finger protein 144B.

Fig. 1

Identification differentially expressed mitochondrial gene in BM. A Visualization of differentially expressed genes (DEGs) between bacterial meningitis and control groups, with showing a heatmap. B-D Identification and merging of candidate genes using weighted gene co-expression network analysis (WGCNA). B Clustering of module eigengenes. A cut-line was selected for the module dendrogram, and some modules were merged according to the dissimilarity of estimated module eigengenes. C Clustering dendrograms of genes, with dissimilarity based on the topological overlap, together with assigned module colors. D) Heatmap of Pearson correlation between gene modules and clinical traits of bacterial meningitis. Data for panels A–D are based on the GSE40586 dataset. E The Venn diagram shows genes that intersect among meningitis DEGs, WGCNA module green genes, and mitochondrial-associated genes

Fig. 2

Causal impact of differential mitochondrial genes on bacterial meningitis and their expression profiles. A Forest plot of mendelian randomization analysis on the effect of RNF144B and FYN on bacterial meningitis. B Differential expression analysis of RNF144B and FYN between the meningitis group and the healthy control group. C Comparison of RNF144B and FYN gene expression levels following infection with different bacterial meningitis strains. D Comparison of the standardized ratio of RNF144B/FYN between the meningitis group and the healthy control group. Data for panels B–D are based on the GSE40586 dataset

Fig. 3

Differential immune cell proportions and their correlation with RNF144B and FYN in meningitis. A-B Violin plot A illustrating and heatmap B displaying immune cell scores using the ssGSEA algorithm. C Analysis of the correlation between infiltration levels of different immune cells in meningitis using the Spearman method. D Heatmap of the correlation analysis between key genes (RNF144B and FYN) and immune cells. E-F Radar charts presenting the correlation analysis between immune cells and FYN E and RNF144B F. Data for panels A-F are based on the GSE40586 dataset

Fig. 4

Dynamic changes in immune cell proportions and key gene expression during bacterial meningitis progression. A t-SNE clustering plot of immune cell subsets. B Bubble plot showing marker genes for each immune cell type. C–D Bar plot C and column plot D of immune cell proportions at various stages of bacterial meningitis progression. E–F Dot plot E and violin plot F of FYN and RNF144B gene expression in different immune cells. Data for panels A–F are based on the GSE163194 dataset

Fig. 5

NKT Cells as Protective Factors in Bacterial Meningitis and the Functional Implications of FYN Gene Expression. A Forest plot illustrating the mendelian randomization analysis of the effect of NKT cells on bacterial meningitis. B–C GSEA analysis of the involvement of the FYN gene in GO functions B and KEGG pathways C

Fig. 6

Differential expression of FYN in NKT cell differentiation in bacterial meningitis. A Pseudotime analysis of NTK cells. B Expression of the FYN across different branches. C Clustering analysis based on gene expression trends in pseudotime

Fig. 7

Functional implications of FYN in NKT cell differentiation in bacterial meningitis. GO and KEGG enrichment analyses for each cluster, covering GO cellular component (CC) (A), GO molecular function (MF) (B), GO biological process (BP) (C), and KEGG pathways (D)

Fig. 8

Differential expression of FYN and RNF144B in peripheral blood during healthy, onset, and recovery stages of meningitis. A Analysis of expression levels of the FYN and the RNF144B at different stages of bacterial meningitis progression, extracted from the GSE163196 dataset. B Comparative analysis of gene expression levels of FYN and RNF144B in bacterial versus viral meningitis. Data are based on the GSE248261 dataset