Identification of common mechanisms and biomarkers for dermatomyositis and atherosclerosis based on bioinformatics analysis

Abstract

Background: Dermatomyositis (DM) manifests as an autoimmune and inflammatory condition, clinically characterized by subacute progressive proximal muscle weakness, rashes or both along with extramuscular manifestations. Literature indicates that DM shares common risk factors with atherosclerosis (AS), and they often co-occur, yet the etiology and pathogenesis remain to be fully elucidated. This investigation aims to utilize bioinformatics methods to clarify the crucial genes and pathways that influence the pathophysiology of both DM and AS.

Method: Microarray datasets for DM (GSE128470, GSE1551, GSE143323) and AS (GSE100927, GSE28829, GSE43292) were retrieved from the Gene Expression Omnibus (GEO) database. The weighted gene co-expression network analysis (WGCNA) was used to reveal their co-expressed modules. Differentially expression genes (DEGs) were identified using the "limma" package in R software, and the functions of common DEGs were determined by functional enrichment analysis. A protein-protein interaction (PPI) network was established using the STRING database, with central genes evaluated by the cytoHubba plugin, and validated through external datasets. Immune infiltration analysis of the hub genes was conducted using the CIBERSORT method, along with Gene Set Enrichment Analysis (GSEA). Finally, the NetworkAnalyst platform was employed to examine the transcription factors (TFs) responsible for regulating pivotal crosstalk genes.

Results: Utilizing WGCNA analysis, a total of 271 overlapping genes were pinpointed. Subsequent DEG analysis revealed 34 genes that are commonly found in both DM and AS, including 31 upregulated genes and 3 downregulated genes. The Degree Centrality algorithm was applied separately to the WGCNA and DEG collections to select the 15 genes with the highest connectivity, and crossing the two gene sets yielded 3 hub genes (PTPRC, TYROBP, CXCR4). Validation with external datasets showed their diagnostic value for DM and AS. Analysis of immune infiltration indicates that lymphocytes and macrophages are significantly associated with the pathogenesis of DM and AS. Moreover, GSEA analysis suggested that the shared genes are enriched in various receptor interactions and multiple cytokines and receptor signaling pathways. We coupled the 3 hub genes with their respective predicted genes, identifying a potential key TF, CBFB, which interacts with all 3 hub genes.

Conclusion: This research utilized comprehensive bioinformatics techniques to explore the shared pathogenesis of DM and AS. The three key genes, including PTPRC, TYROBP, and CXCR4, are related to the pathogenesis of DM and AS. The central genes and their correlations with immune cells may serve as potential diagnostic and therapeutic targets.

Keywords: GEO database; atherosclerosis; bioinformatics analysis; dermatomyositis; expression; gene; immune infiltration analysis.

FIGURE 1

The flowchart for bioinformatic analysis.

FIGURE 2

Screening of genes in the GSE143323 (DM) and GSE28829 (AS) datasets using the WGCNA algorithm. (A), (B) The Cluster dendrogram in GSE143323 (DM) and in GSE28829 (AS). (C), (D) Heatmap illustrating the module‐trait relationships in GSE143323 (DM) and GSE28829 (AS). AS, atherosclerosis; DM, dermatomyositis; WGCNA, weighted gene coexpression network analysis.

FIGURE 3

Venn plot of common genes between DM and AS and the PPI network of the intersecting genes. (A) The overlapped genes between the turquoise module in GSE1551 (DM) and the green module in GSE28829 (AS). (B)–(E) A PPI network for the 192 common genes and three clusters extracted using MCODE. (F) Bubble plot of GO enrichment analysis of three gene clusters. (G) Bubble plot of KEGG enrichment analysis of three gene clusters. AS, atherosclerosis; DM, dermatomyositis; GO, Gene Ontology; IPF, idiopathic pulmonary fibrosis; KEGG, Kyoto encyclopedia of genes and genomes; MCODE, minimal common oncology data elements; PPI, protein‐protein interaction.

FIGURE 4

Volcano plot and Heatmap of the DEGs identified from GSE128470 and GSE100927. (A), (B) Volcano map of DEGs from GSE128470 and GSE100927. (C), (D) Heatmap of DEGs from GSE128470 and GSE100927. DEGs, differentially expression gene.

FIGURE 5

Verification and analysis of shared DEGs in DM and AS. (A) Venn diagram for shared DEGs in the GSE128470 and GSE100927 datasets. (B) The PPI network of the shared DEGs. (C) One cluster extracted by MCODE. (D) Bubble plot of GO enrichment analysis of common DEGs. (E) Bubble plot of KEGG enrichment analysis of common DEGs. AS, atherosclerosis; DEGs, differentially expression gene; DM, dermatomyositis; GO, Gene Ontology; IPF, idiopathic pulmonary fibrosis; KEGG, Kyoto encyclopedia of genes and genomes; MCODE, minimal common oncology data elements.

FIGURE 6

Screening out of common hub genes through the intersection of the top 15 genes derived from DEG and WGCNA. DEGs, differentially expressed genes; WGCNA, weighted gene coexpression network analysis.

FIGURE 7

The expression levels and diagnostic efficacy of the common genes in DM datasets. (A) PTPRC, TYROBP and CXCR4 expression levels in two DM databases, with blue violin plots representing DM, red representing controls. Students’t‐test with p < 0.05 was used to determine statistical significance. ^* p < 0.05; ^** p < 0.01;^*** p < 0.001; ^**** p < 0.0001. (B) The ROC curves showing AUC values of PTPRC, TYROBP and CXCR4 in DM. DM, dermatomyositis; AUC, area under curve.

FIGURE 8

The expression levels and diagnostic efficacy of the common genes in AS datasets. (A) PTPRC, TYROBP and CXCR4 expression levels in two AS databases, with yellow violin plots representing AS, red representing controls. Students’t‐test with p < 0.05 was used to determine statistical significance. ^* p < 0.05; ^** p < 0.01;^*** p < 0.001; ^**** p < 0.0001. (B) The ROC curves showing AUC values of PTPRC, TYROBP and CXCR4 in AS. AS, Atherosclerosis; AUC, area under curve.

FIGURE 9

Immune infiltration analysis of DM. (A) Barplot of the 22 immune cells proportion. (B) Comparison of immune cell proportion between DM and controls (Willcoxon's test). ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001. DM, dermatomyositis.

FIGURE 10

Immune infiltration analysis of AS. (A) Barplot of the 22 immune cells proportion. (B) Comparison of immune cell proportion between AS and controls (Willcoxon's test). ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001. AS, atherosclerosis.

FIGURE 11

Correlation analysis between hub genes and immune cells. (A) Correlation analysis between hub genes and immune cells in DM. (B) Correlation analysis between hub genes and immune cells in AS. AS, atherosclerosis; DM, dermatomyositis.

FIGURE 12

GSEA analysis of hub genes (PTPRC,TYROBP and CXCR4) in DM. DM, dermatomyositis; GSEA, Gene set enrichment analysis.

FIGURE 13

GSEA analysis of hub genes (PTPRC,TYROBP and CXCR4) in AS. AS, atherosclerosis; GSEA, Gene set enrichment analysis.

FIGURE 14

TF‐shared hub gene regulatory network. The red circle represents a shared hub gene, while the blue diamond represents transcription factors. TFs, transcription factors.