Identification of a Potential MiRNA-mRNA Regulatory Network for Osteoporosis by Using Bioinformatics Methods: A Retrospective Study Based on the Gene Expression Omnibus Database

Abstract

Introduction: As a systemic skeletal dysfunction, osteoporosis (OP) is characterized by low bone mass, impairment of bone microstructure, and a high global morbidity rate. There is increasing evidence that microRNAs (miRNAs) are associated with the pathogenesis of OP. Weighted gene co-expression network analysis (WGCNA) is a systematic method for identifying clinically relevant genes involved in disease pathogenesis. However, the study of the miRNA-messenger RNA (mRNA) regulatory network in combination with WGCNA in OP is still lacking.

Methods: The GSE93883 and GSE7158 microarray datasets were downloaded from the Gene Expression Omnibus (GEO) database. Differentially expressed miRNAs (DE-miRNAs) and differentially expressed genes (DEGs) were analyzed with the limma package. OP-related miRNAs from the most clinically relevant module were identified by the WGCNA method. The overlap of DE-miRNAs and OP-related miRNAs was identified as OP-related DE-miRNAs. Both upstream transcription factors and downstream targets of OP-related DE-miRNAs were predicted by FunRich. An intersection of predicted target genes and DEGs was confirmed as downstream target genes of OP-related DE-miRNAs. With the use of clusterProfiler in R, Gene Ontology (GO) annotation and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment were performed on target genes. Finally, both the protein-protein interaction (PPI) network and miRNA-mRNA network were constructed and analyzed.

Results: A total of 79 OP-related DE-miRNAs were obtained, most of which were predicted to be regulated by specificity protein 1 (SP1). Subsequently, 197 downstream target genes were screened out. The target genes were enriched in multiple pathways, including signaling pathways closely related to the onset of OP, such as Ras, PI3K-Akt, and ErbB signaling pathways. Through the construction of the OP-related miRNA-mRNA regulatory network, a hub network that may play a prominent role in the formation of OP was documented.

Conclusion: By using WGCNA, we constructed a potential OP-related miRNA-mRNA regulatory network, offering a novel perspective on miRNA regulatory mechanisms in OP.

Keywords: WGCNA; bioinformatics analysis; miRNAs; miRNA–mRNA regulatory network; osteoporosis.

Figure 1

Workflow of research routine to construct the miRNA–mRNA regulatory network in OP. miRNA, microRNA; mRNA, messenger RNA; OP, osteoporosis; DE-miRNAs, differentially expressed miRNAs; DEGs, differentially expressed genes.

Figure 2

Identifying significant gene modules associated with OP. (A) β = 13 was selected to establish a scale-free network in the GSE93883 dataset. (B) Gene clustering based on topological dissimilarity and module colors in GSE93883 dataset. (C) Correspondence between gene modules with osteoporosis participants or normal ones among the GSE93883 dataset. (D) The correlation between gene significance and module membership in the GSE93883 dataset was plotted. OP, osteoporosis.

Figure 3

Screening of DE-miRNAs, DEGs, and OP-related DE-miRNAs. (A) Heatmap of miRNA expression in GSE93883 dataset. (B) GSE93883 miRNA expression volcano plot. (C) Gene expression heatmap of GSE7158 dataset. (D) GSE7158 gene expression volcano plot. Black color indicates non-significant genes, while red/green color represents upregulated/downregulated DE-miRNAs or DEGs. (E) Venn diagram of DE-miRNAs and miRNAs closely related to OP in the turquoise module. (F) Venn diagram of predicted targets of OP-related DE-miRNAs and DEGs of GSE7158 dataset. The number of miRNAs or genes in each group is displayed in Venn diagrams. N, normal group; T, osteoporosis group; OP, osteoporosis; miRNA, microRNA; mRNA, messenger RNA; DE-miRNAs, differentially expressed miRNAs; DEGs, differentially expressed genes.

Figure 4

Upstream transcription factors for OP-related DE-miRNAs. OP, osteoporosis; DE-miRNAs, differentially expressed miRNAs.

Figure 5

GO functional enrichment of downstream targets of OP-related DE-miRNAs. The x-axis shows enriched gene numbers and the color represents significance. GO terms are shown on the y-axis. p < 0.05. OP, osteoporosis; DE-miRNAs, differentially expressed miRNAs; DEGs, differentially expressed genes; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; BP, biological process; CC, cellular component; MF, molecular function.

Figure 6

KEGG pathway enrichment for downstream target genes of OP-related DE-miRNAs. The enriched gene ratio is shown on the x-axis, color represents significance, and the pathway terms are displayed on the y-axis. p < 0.05. OP, osteoporosis; DE-miRNAs, differentially expressed miRNAs; KEGG, Kyoto Encyclopedia of Genes and Genomes.

Figure 7

PPI network of downstream targets of OP-related DE-miRNAs. miRNA, microRNA; PPI, protein–protein interaction; OP, osteoporosis; DE-miRNAs, differentially expressed miRNAs.

Figure 8

OP-relevant miRNA–mRNA regulatory network. Ellipse represents target genes of miRNAs; triangle represents miRNAs. Red dots represent miRNAs or targets that are upregulated, while green dots represent those that are downregulated. (A) MiRNA–mRNA regulatory network of OP. (B) A subnetwork identified by the MCODE algorithm in Cytoscape. OP, osteoporosis; miRNA, microRNA; mRNA, messenger RNA.