Critical Roles of ELVOL4 and IL-33 in the Progression of Obesity-Related Cardiomyopathy via Integrated Bioinformatics Analysis

Abstract

The molecular mechanisms underlying obesity-related cardiomyopathy (ORCM) progression involve multiple signaling pathways, and the pharmacological treatment for ORCM is still limited. Thus, it is necessary to explore new targets and develop novel therapies. Microarray analysis for gene expression profiles using different bioinformatics tools has been an effective strategy for identifying novel targets for various diseases. In this study, we aimed to explore the potential genes related to ORCM using the integrated bioinformatics analysis. The GSE18897 (whole blood expression profiling of obese diet-sensitive, obese diet-resistant, and lean human subjects) and GSE47022 (regular weight C57BL/6 and diet-induced obese C57BL/6 mice) were used for bioinformatics analysis. Weighted gene co-expression network analysis (WGCNA) of GSE18897 was employed to investigate gene modules that were strongly correlated with clinical phenotypes. Gene Ontology (GO) functional enrichment analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis were performed on the co-expression genes. The expression levels of the hub genes were validated in the clinical samples. Yellow co-expression module of WGCNA in GSE18897 was found to be significantly related to the caloric restriction treatment. In addition, GO functional enrichment analysis and KEGG pathway analysis were performed on the co-expression genes in yellow co-expression module, which showed an association with oxygen transport and the porphyrins pathway. Overlap analysis of yellow co-expression module genes from GSE18897 andGSE47022 revealed six upregulated genes, and further experimental validation results showed that elongation of very-long-chain fatty acids protein 4 (ELOVL4), matrix metalloproteinase-8 (MMP-8), and interleukin-33 (IL-33) were upregulated in the peripheral blood from patients with ORCM compared to that in the controls. The bioinformatics analysis revealed that ELOVL4 expression levels are positively correlated with that of IL-33. Collectively, using WGCNA in combination with integrated bioinformatics analysis, the hub genes of ELVOL4 and IL-33 might serve as potential biomarkers for diagnosis and/or therapeutic targets for ORCM. The detailed roles of ELVOL4 and IL-33 in the pathophysiology of ORCM still require further investigation.

Keywords: ELOVL4; IL-33; ORCM; WGCNA; yellow co-expression module.

FIGURE 1

Workflow of the study strategies using datasets including GSE18897 and GSE47022.

FIGURE 2

Selection of the proper soft-threshold power β for WGCNA. (A) Outliers were determined using sample clustering method, and the analysis was performed according to the expression data of DEGs from whole blood of well-matched obesity and lean individuals. (B) The scale-free fit index of network topology was determined by soft-thresholding power analysis. (C) Heatmap plot of the adjacencies in the hub gene network. The trait weight was included. Each column and row correspond to one co-expression module hub gene (labeled by color) or weight. In the heatmap, red represents high adjacency (positive correlation) and blue represents low adjacency (negative correlation). Red squares along the diagonal are the meta-modules.

FIGURE 3

Construction of the co-expression network and identification of the most related co-expression modules. (A) Clustering of genes together with assigned module colors. The dissimilarity was based on the topological overlap. The y-axis is the distance determined by the extent of the topological overlap. (B) Heatmap with each cell containing the p-value correlation from linear mixed effects model. Row corresponds to different co-expression modules; column corresponds to traits of caloric restriction treatment. (C) The gene correlation heatmap of the genes in yellow co-expression module. Each color circle represents the distance between different genes. (D) GO enrichment and KEGG pathway analysis of 633 genes in the yellow co-expression module. (E) PPI network construction of the top 50 genes in the yellow co-expression module.

FIGURE 4

Overlap analysis of hub genes from human and mouse DEGs. (A) KEGG pathway analysis of genes from yellow co-expression module and DEGs from mice. (B) Venn diagram shows common hub genes between yellow co-expression module and DEGs from mice. (C) Volcano plot of yellow co-expression module and DEGs from murine datasets.

FIGURE 5

Validation of hub genes in the clinical human samples. (A) Gene expression levels of the hub genes (ELOVL4, CYPB, SNCA, ZFN383, MMP8, and IL33) in the human peripheral blood from ORCM patients and healthy controls were determined by qRT-PCR. (B) Protein levels of ELOVL4, MMP8, and IL33 in the human peripheral blood from ORCM patients and healthy controls were determined by ELISA assay. N = 6; ns = non-significant; **P < 0.01.

FIGURE 6

Correlation analysis of ELOVL4 and IL33 gene expression levels. (A) KEGG pathway analysis of the hub genes. (B) The correlation between ELOVL4 and IL33 expression levels were analyzed using Pearson correlation analysis. (C) Analysis of ELOVL4 and IL33 expression levels in lean controls, obese patients, and obese patients received 3- or 6-month period of caloric restriction treatment.