Integrative Identification of Hub Genes Associated With Immune Cells in Atrial Fibrillation Using Weighted Gene Correlation Network Analysis

Abstract

Background: Atrial fibrillation (AF) is the most common tachyarrhythmia in the clinic, leading to high morbidity and mortality. Although many studies on AF have been conducted, the molecular mechanism of AF has not been fully elucidated. This study was designed to explore the molecular mechanism of AF using integrative bioinformatics analysis and provide new insights into the pathophysiology of AF. Methods: The GSE115574 dataset was downloaded, and Cibersort was applied to estimate the relative expression of 22 kinds of immune cells. Differentially expressed genes (DEGs) were identified through the limma package in R language. Weighted gene correlation network analysis (WGCNA) was performed to cluster DEGs into different modules and explore relationships between modules and immune cell types. Functional enrichment analysis was performed on DEGs in the significant module, and hub genes were identified based on the protein-protein interaction (PPI) network. Hub genes were then verified using quantitative real-time polymerase chain reaction (qRT-PCR). Results: A total of 2,350 DEGs were identified and clustered into eleven modules using WGCNA. The magenta module with 246 genes was identified as the key module associated with M1 macrophages with the highest correlation coefficient. Three hub genes (CTSS, CSF2RB, and NCF2) were identified. The results verified using three other datasets and qRT-PCR demonstrated that the expression levels of these three genes in patients with AF were significantly higher than those in patients with SR, which were consistent with the bioinformatic analysis. Conclusion: Three novel genes identified using comprehensive bioinformatics analysis may play crucial roles in the pathophysiological mechanism in AF, which provide potential therapeutic targets and new insights into the treatment and early detection of AF.

Keywords: WGCNA; atrial fibrillation; bioinformatics; hub genes; immune cells.

Figure 1

The relative expression of 22 immune cell subtypes in each sample estimated using Cibersort. The relative expression was higher from blue to red.

Figure 2

Correlation matrix of 13 immune cell subtype compositions. Blue dots represent negative correlation, and red dots represent positive correlation. The size of the dot is positively correlated with the correlation coefficient. ^*p < 0.05 and ^***p < 0.001.

Figure 3

Weighted genes correlation network analysis to cluster genes into different modules. (A) The selection of the soft-thresholding power β. (B) Dendrogram of all differentially expressed genes. (C) The clustering heat map between modules. Red means closer similarity, and blue means farther similarity.

Figure 4

The heatmap showing module-trait correlations. Blue represent negative correlation, and red represent positive correlation. The magenta module had the strongest correlation with M1 macrophages.

Figure 5

GO and KEGG enrichment analyses. (A) Biological process. (B) Cellular component. (C) Molecular function. (D) KEGG pathways.

Figure 6

The PPI network of genes in the magenta module.

Figure 7

A Venn diagram between five algorithms of CytoHubba. The coincident part represents the three genes (CTSS, CSF2RB, and NCF2) identified by all five algorithms.

Figure 8

The relative expression of three hub genes between atrial fibrillation (AF) and sinus rhythm (SR). (A–C) The relative expression of CSF2RB, CTSS, and NCF2 in left atriums. (D–F) The relative expression of CSF2RB, CTSS, and NCF2 in blood samples. ^****p < 0.0001.

Figure 9

The relative expression of three hub genes verified by three datasets. (A) The relative expression of CSF2RB. (B) The relative expression of CTSS. (C) The relative expression of NCF2. ^*p < 0.05, ^**p < 0.01.