Hub microRNAs and genes in the development of atrial fibrillation identified by weighted gene co-expression network analysis

Qiang Qu,Shan-Shan Li,Feng Li,Ru-Xing Wang,Jin-Yu Sun,Zhen-Ye Zhang,Yue Su

doi:10.1186/s12920-021-01124-5

Abstract

Co-expression network may contribute to better understanding molecular interaction patterns underlying cellular processes. To explore microRNAs (miRNAs) expression patterns correlated with AF, we performed weighted gene co-expression network analysis (WGCNA) based on the dataset GSE28954. Thereafter, we predicted target genes using experimentally verified databases (ENOCRI, miRTarBase, and Tarbase), and overlapped genes with differentially expressed genes (DEGs) from GSE79768 were identified as key genes. Integrated analysis of association between hub miRNAs and key genes was conducted to screen hub genes. In general, we identified 3 differentially expressed miRNAs (DEMs) and 320 DEGs, predominantly enriched in inflammation-related functional items. Two significant modules (red and blue) and hub miRNAs (hsa-miR-146b-5p and hsa-miR-378a-5p), which highly correlated with AF-related phenotype, were detected by WGCNA. By overlapping the DEGs and predicted target genes, 38 genes were screened out. Finally, 9 genes (i.e. ATP13A3, BMP2, CXCL1, GABPA, LIF, MAP3K8, NPY1R, S100A12, SLC16A2) located at the core region in the miRNA-gene interaction network were identified as hub genes. In conclusion, our study identified 2 hub miRNAs and 9 hub genes, which may improve the understanding of molecular mechanisms and help to reveal potential therapeutic targets against AF.

Highlights

Atrial fibrillation (AF), the most common sustained cardiac arrhythmia, is expected to affect one quarter of middle-aged adults worldwide, primarily at the age of > 70 years [1]
Microarray-based miRNA and mRNA expression data were obtained from the Gene Expression Omnibus (GEO) database if they met the following selection criteria [22]: (1) expression data were from atrial tissue samples in non-AF and AF patients; (2) study type was restricted to expression profiling by array; (3) organism was restricted to Homo sapiens; and (4) raw/processed expression data were public and available
Microarray data normalization and identification of the Differentially expressed miRNA (DEM) and Differentially expressed gene (DEG) In the dataset GSE28954, 10 AF and 18 non-AF samples were included, with no further information given; in the dataset GSE79768, 14 AF and 12 non-AF samples were included with a mean age of 55.4 years, comprising 10 (38.5%) men and 16 (61.5%) women

Summary

Introduction

Atrial fibrillation (AF), the most common sustained cardiac arrhythmia, is expected to affect one quarter of middle-aged adults worldwide, primarily at the age of > 70 years [1]. According to the epidemiological investigation, the overall prevalence was estimated to be 1–4% in Australia, Europe and the USA [1, 2]. The number continued to increase as the population is ageing, reaching 4. AF has been associated with electrophysiological and structural remodeling, characterized by rapid and uncoordinated atrial activity [6]. The mechanisms underlying AF are complex and multifactorial, and are broadly classified as trigger and substrate, which lead to an increased risk of stroke, heart failure, and premature death [7].

Objectives

Methods

Results

Discussion

Conclusion