This article is mentioned in:

Introduction

Coronary artery disease (CAD), which is the most common type of cardiovascular disease, is the most prevalent cause of mortality worldwide (1). Recently, the incidence of CAD was demonstrated to have increased in China (2). Typically, the underlying mechanism of CAD involves a section of the coronary artery inter wall developing atherosclerosis. There are a variety of risk factors associated with CAD, including hypertension, obesity, smoking, family history of the disease (3), diabetes, lack of exercise, depression (4) and high blood lipids (5,6). Although a number of advances in the understading of associated pathogeny, diagnosis and treatment of CAD have been described previously, current knowledge of the molecular mechanisms underlying CAD is insufficient for the development of improved treatment and diagnostic strategies for patients with CAD, and thus requires further investigation. Identification of the potential therapeutic targets for CAD remains clinically important.

DEGs have important roles in complex human diseases. MicroRNA (miRNAs) are a class of small noncoding RNA molecules that serve a crucial role in RNA silencing and the post-transcriptional regulation of gene expression by binding to the 3′-untranslated region of their target genes (7,8). The identification of DEGs and DEMs associated with CAD may help to elucidate the underlying molecular mechanisms as well as discover novel biomarkers and therapies.

Recently, the present study performed a comprehensive analysis of global mRNA and miRNA expression profile datasets, including data from normal individuals and patients with CAD. A total of 350 mRNAs and 66 miRNAs were dysregulated in patients with CAD. Using these data, 2 networks of miRNA-gene associations were established. Through the identification of 26 miRNA-target DEGs, potential underlying molecular mechanisms and therapeutic targets for CAD were investigated.

Materials and methods

mRNA and miRNA expression datasets

In the present study, the GSE28829 mRNA (9) and GSE59421 miRNA (10) expression datasets were downloaded from the Gene Expression Omnibus (GEO) database (www.ncbi.nlm.nih.gov/geo). In the GSE28829 dataset, 16 atherosclerotic plaque samples from patients and 13 control samples were included and analyzed using the Affymetrix Human Genome U133 Plus 2.0 Array. In the GSE59421 dataset, the expression levels in platelets were identified in 33 male patients with premature CAD and 37 age- and sex-matched healthy controls, and were quantified using the agilent-021827 Human miRNA Microarray (V3) (miRBase release 12.0 miRNA ID version). These 2 datasets were indexed and loaded into Entrez GEO Profiles and Entrez GEO DataSets, which allows users to query and analyze the data using simple Boolean queries, and provides weblinks to free public raw data, accurate information, reliable data platforms and other National Center for Biotechnology Information resources wherever possible.

Microarray data mining

Analysis of differentially expressed genes (DEGs) was conducted for the mRNA and miRNA microarray data using GEO2R Online Analytical tool (https://www.ncbi.nlm.nih.gov/geo/geo2r/). DEGs and differentially expressed miRNAs (DEMs) were identified using an adjusted P-value <0.05 and a |log fold change (FC)| >1 as criteria.

Establishment of the miRNA-gene regulatory network

miRWalk2.0 (http://zmf.umm.uni-heidelberg.de/apps/zmf/mirwalk2/) is a web-based tool developed by Dweep et al (11,12) that supplies the largest available collection of predicted and experimentally verified miRNA-target interactions with various novel and unique features. In the present study, target genes of the DEMs were identified using miRWalk2.0. Furthermore, Cytoscape version 3.5.1 software (13) was used to establish the miRNA-gene regulatory network.

miRNA-target DEG protein-protein interactions

Search Tool for the Retrieval of Interacting Genes/Proteins (STRING; http://string-db.org/) is a database of known and predicted protein-protein interactions (14). In the present study, the protein-protein interactions were screened using STRING.

Functional enrichment analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis

Based on the Database for Visualization, Annotation and Integrated Analysis (DAVID; david.abcc.ncifcrf.gov/.) (15), the enriched Gene Ontology (GO) terms for miRNA-target DEGs were identified (only the GO terms that were enriched in ≥5 genes were included). GO analysis is a commonly used approach for functional studies of large-scale transcriptomic data (16). The KEGG pathway database (17) contains information on networks of molecules or genes.

miRNA-target DEGs chromosomal location

The Map Viewer (https://www.ncbi.nlm.nih.gov/projects/mapview/) provides a wide variety of genome mapping and sequencing data (18). The locations of the miRNA-target DEGs on the chromosomes were identified using this resource.

Results

A total of 350 DEGs and 66 DEMs are identified

The available numerical mRNA and miRNA expression values were used to identify DEGs and DEMs. A total of 59 upregulated and 291 downregulated DEGs, and 30 upregulated and 36 downregulated DEMs were identified in patients with CAD compared with the normal controls. The top 10 DEGs and DEMs are presented in Table I (P<0.05; |logFC| >1).

Table I. Top 10 most differentially expressed genes and differentially expressed miRNAs.

Table I.

Top 10 most differentially expressed genes and differentially expressed miRNAs.

Gene	P-value	logFC	miRNA	P-value	logFC
C2	1.6×10−10	−1.25626406	hsa-miR-221	1.13×10−4	−1.2728544
DENND2D	5.16×10−10	−1.15465237	hsa-miR-1246	2.52×10−4	−1.6266646
ADAP2	2.45×10−9	−1.22371715	hsa-miR-425	8.73×10−4	−1.1111774
JAK3	3.09×10−9	−1.09039561	hsa-miR-505	1.54×10−3	−1.1639686
SLAMF8	4.26×10−9	−2.07254568	hsa-miR-598	1.93×10−3	−1.2433251
CD68	4.51×10−9	−1.13556898	hsa-miR-431*	3.26×10−3	1.0564151
LINC01094	6.25×10−9	−1.10497076	hsa-miR-25	3.85×10−3	−1.0901694
SERPINA1	8.24×10−9	−1.73250228	hsa-miR-30e	4.66×10−3	−1.1414787
VAMP8	8.69×10−9	−1.82538528	hsa-miR-370	5.2×10−3	1.2500556
C3AR1	1.14×10−8	−2.22628842	hsa-miR-1274a	5.36×10−3	−1.27226

[i] FC, fold-change; hsa, homo sapiens; miRNA, microRNA.

miRNA-gene regulation network and protein interactions of 57 overlapping DEGs

Using miRWalk2.0, 3,588 target genes of the DEMs were identified, and 57 of these target genes were demonstrated to overlap with the identified DEGs. In the group of 57 overlapping DEGs, a protein-protein interaction network of 26 DEGs was established using STRING. Furthermore, 26 miRNA-gene pairs were obtained among the 26 aforementioned DEGs and 19 DEMs. The 19 DEMs were demonstrated to regulate multiple genes, thereby forming a considerable network (Fig. 1).

Figure 1. DEMs regulate multiple genes. Red squares represent DEMs and blue squares denote DEGs. Each line (edge) is a pairwise connection between 2 genes, or between DEMs and DEGs. The figure was plotted using Cytoscape version 3.5.1 software. hsa, homo sapiens; miR, microRNA; DEMs, differentially expressed miRNAs; DEGS, differentially expressed genes.

GO and KEGG pathway analysis

Using DAVID, 40 enriched GO process terms were obtained for miRNA-targeting DEGs. The five most significantly enriched GO terms, including ‘regulation of cell communication’, ‘regulation of signaling’, ‘regulation of response to stimulus’, ‘positive regulation of biological process’ and ‘positive regulation of cellular process’, are presented in Fig. 2A. A total of 4 enriched GO component terms were obtained for miRNA-target DEGs, including ‘low-density lipoprotein particle’, ‘plasma membrane’, ‘cell periphery’ and ‘membrane’ (Fig. 2B). KEGG pathway analysis identified 3 genes [hexokinase 2 (HK2), phosphatidylinositol-4,5-biphosphate 3-kinase catalytic subunit γ (PIK3CG), and suppressor of cytokine signaling 3 (SOCS3)] that were present in Type II diabetes mellitus network and 3 other genes, ATPase Na+/K+ transporting subunit α 2, HK2 and PIK3CG that were present in the carbohydrate digestion and absorption network.

Figure 2. Enriched GO process terms and component terms of miRNA-target differentially expressed genes. (A) Enriched GO process terms and (B) enriched GO component terms of miRNA-target differentially expressed genes obtained from the Database for Visualization, Annotation and Integrated Analysis. GO, gene ontology; miRNA, microRNA.

Furthermore, the miRNA-gene regulatory network of the 26 miRNA-gene pairs was established using Cytoscape version 3.5.1 software (Fig. 3). The 26 regulatory pairs between DEMs and their target genes were identified, including hsa-miR-199a-5p-apolioprotein E (APOE), hsa-miR-106a-branched chain amino acid transaminase 1 (BCAT1), hsa-miR-1234-cytohesin 1 interacting protein (CYTIP), hsa-miR-1249-macrophage scavenger receptor 1 (MSR1), hsa-miR-142-3p-ecotropic viral integration site 2B (EVI2B), hsa-miR-142-3p-myristoylated alanine rich protein kinase C substrate (MARCKS), hsa-miR-142-3p-Phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit gamma isoform (PIK3CG), hsa-miR-152-MAF basic leucine zipper (BZIP) transcription factor B (MAFB), hsa-miR-215-DEP domain containing MTOR interacting protein (DEPTOR), hsa-miR-215-HOP homeobox (HOPX), hsa-miR-22-colony stimulating factor 1 receptor (CSF1R), hsa-miR-221-coronin 1A (CORO1A), hsa-miR-221-Suppressor of cytokine signaling 3 (SOCS3), hsa-miR-221-T cell lymphoma invasion and metastasis 1 (TIAM1), hsa-miR-30b-immunoglobulin superfamily member 6 (IGSF6), hsa-miR-30c-C-X-C motif chemokine ligand 16 (CXCL16), hsa-miR-30e-Serpin family E member 2 (SERPINE2), hsa-miR-323a-3p-phospholipase A2 group VII (PLA2G7), hsa-miR-329-ecotropic viral integration site 2A (EVI2A), hsa-miR-33a-chromosome 15 open reading frame 48 (C15orf48), hsa-miR-33a-FYN binding protein 1 (FYB), hsa-miR-376-death associated protein kinase 1 (DAPK1), hsa-miR-484-hematopoietic cell-specific Lyn substrate 1 (HCLS1), hsa-miR-484-HK2, hsa-miR-486-3p-netrin 1 (NTN1) and hsa-miR-616-lysozyme (LYZ).

Figure 3. miRNA-gene regulatory network of 26 miRNA-gene pairs. Green circles represent genes and red circles denote miRNA. Each line (edge) is a pairwise connection between two genes, or between differentially expressed miRNAs and differentially expressed genes. The figure was plotted using Cytoscape version 3.5.1 software. hsa, homo sapiens; miR, microRNA.

miRNA-target DEG chromosomal location

The 26 DEGs in the miRNA-gene pairs were located on the chromosomes by Map Viewer (Fig. 4). The chromosomal locations of these genes were inhomogeneous, and they were primarily distributed on chromosomes 17, 2, 5, 6, 12, 16, 1, 3, 4, 7, 8, 9, 15, 19, 20, and 21. None of the associated genes were distributed on sex chromosomes.

Figure 4. microRNA-target differentially expressed genes chromosomal location. Black rectangles represent chromosomes and white spots represents centromeres. The arrows signify the specific locations of the genes.

Discussion

miRNAs are crucial and evolutionarily conserved components of gene regulation (19). Up- or downregulated expression levels of specific miRNAs were observed in diseased human hearts, implying their involvement in cardiomyopathies (20). CAD is an ischemic heart disease that occurs when part of the arterial internal wall develops atherosclerosis, which is associated with lipid retention and inflammation. Murine miRNA-712 is a potential biomarker for atherosclerosis (21). However, the detailed mechanism underlying the mechanism by which miRNAs regulate the expression of CAD-associated genes and gene products requires thorough investigation.

In the present study, a total of 59 upregulated and 291 downregulated DEGs, and 30 upregulated and 36 downregulated DEMs were identified in patients with CAD compared with normal controls. The GO analysis indicated that several significantly enriched terms from the categories of regulation of cell communication, regulation of signaling, regulation of response to stimulus, positive regulation of biological process, and positive regulation of cellular process were markedly over-represented in patients with CAD, suggesting that these biological processes are important topics for future study.

The present study also identified the interactions and associations among proteins, which were encoded by 26 DEGs in the miRNA-gene pairs. In addition, the 19 miRNAs (which target the 26 DEGs) also regulate a large quantity of genes.

Apolipoprotein E is encoded by the APOE gene, and serves as a shuttle lipid in the bloodstream (22). Maintaining constant levels of cholesterol is very important for avoiding cardiovascular disease. Carriers of the APOEe4 allele exhibit increased risks of developing atherosclerosis; an abnormal expression of APOE is associated with fatty deposition, which increases the risk of heart attack and stroke (23). In addition, previous studies have identified that mutated APOE causes an increased risk of Alzheimer's disease in individuals; however, the mechanism is unclear (24). Thus, it may be suggested that the APOE gene is associated with Alzheimer's disease and CAD. The BCAT1 gene encodes a form of transaminase to catalyze the branched-chain amino acid transamination, which when defective is prone to 2 types of disease; hypervalinemia and hyperleucine-isoleucinemia (25,26). The coding products of the MSR1 gene are membrane glycoproteins, which are associated with atherosclerosis, Alzheimer's disease and host defense (27). DEPTOR has been considered as an endogenous regulator overexpressed in multiple myeloma cells (28). Increased levels of CSF1R were identified in microglia in Alzheimer's disease (29). A number of types of cytokines [interleukin (IL)6, IL10, interferon (IFN)-γ] may induce the expression of SOCS3 gene. Knocking out the SOCS3 gene prevents insulin resistance in obesity (30). The IGSF6 gene is associated with inflammatory bowel disease (31). HK2 is the first and rate-limiting enzyme in the Embden-Meyerhof-Parnas pathway (32). DAPK1 is a positive mediator of γ-interferon-induced programmed cell death. Diseases identified to be associated with HCLS1 include spherocytosis, type 1 and congenital hemolytic anemia (33). The product of the HCLS1 gene serve a role in antigen receptor signaling for clonal expansion and deletion in lymphoid cells (34). Platelet-activating factor acetylhydrolase deficiency may be caused by the defective PLA2G7 gene, in which the associated pathways are the phospholipase-C pathway and sweet taste receptor signaling (35). Oxidation of low-density lipoproteins is an initial step of atherogenesis that generates pro-inflammatory phospholipids, including platelet-activating factor and its analogs (36). GO annotations associated with EVI2A include transmembrane signaling receptor activity (37). Mutations in SERPINE2 may cause a range of diseases; the α-antitrypsin deficiency is one of the most common hereditary diseases associated with this gene (38). CXCL16 expression is induced by the inflammatory cytokines IFN-γ and tumor necrosis factor-α (39). In patients with systemic lupus erythematosus, the differential expression of the FYB gene encodes a protein that participates in the T-cell signaling cascade and in IL-2A expression modulation (40). The C15orf48 gene was first identified in a study of human esophageal squamous cell carcinoma tissues (41). The HOPX gene is involved in the malignant conversion of placental trophoblasts (42). The protein encoded by CYTIP is weakly expressed in resting natural killer and T cells, which contain 2 leucine zipper domains and a putative C-terminal nuclear targeting signal (43). CYTIP is associated with amplification and expansion of oncogenic pathways, and exhibits metastatic traits (43). The gene product of the EVI2B gene has been studied extensively regarding its role in neurofibromatosis, leukemia, and myeloid leukemia (44). The MARCKS gene product serves important roles in cell shape, cell motility, secretion, transmembrane transport, regulation of cell cycle and neural development (45). The PIK3CG gene product is an enzyme that phosphorylates phosphoinositides on the 3-hydroxyl group of the inositol ring, PIK3CG SNPs (rs1129293 and rs17398575) increase the risk of ischemia in patients suffering from coronary heart disease (44). MAFB is a bZIP transcription factor that serves an important role in the regulation of lineage-specific hematopoiesis, and rs2902940A alleles of the MAFB gene result in decreased serum ApoAI levels in controls and an increased risk of CAD (46). The encoded nuclear protein represses ETS1-mediated transcription of erythroid-specific genes in myeloid cells (47). CORO1A has been implicated in T-cell mediated immunity and mitochondrial apoptosis (48). TIAM1 has been identified as a specific activator of the Rho-like GTPase Rac and is implicated in the regulation of different cell biological functions, including cell polarity, adhesion, migration, invasion, metastasis, and carcinogenesis (49). NTN1 is included in a family of laminin-associated secreted proteins (50). The LYZ protein exhibits antibacterial activity against certain bacterial species. Missense mutations in the LYZ gene have been identified in heritable renal amyloidosis. miRNAs serve as stimulating factors at various stages of atherosclerosis to regulate a number of genes, thereby regulating endothelial cells, vascular smooth muscle cells, and the function of macrophages, which control the occurrence of atherosclerosis (51).

The chromosomal locations of these aforementioned 26 genes are inhomogeneous, and they are primarily distributed on chromosomes 17, 2, 5, 6, 12, 16, 1, 3, 4, 7, 8, 9, 15, 19, 20 and 21. No miRNA-target DEGs were identified on the sex chromosomes, which suggested that CAD may be independent of sex. in CAD. Finally, the results of the study may be a useful resource for understanding the functions of miRNA and potential therapeutic targets in the pathophysiology of CAD, but additional molecular analyses are required. The affected interactions of the 26 miRNA-gene pairs identified in the present study may be validated in vitro by dual-luciferase reporter gene assays, and the functions of the 26 miRNAs and their target genes will be confirmed in vivo and in vitro, so that potential therapeutic targets may be identified and effective targeting of the affected biological pathways may be achieved for the treatment and improved prognosis of CAD.

Acknowledgements

Not applicable.

Funding

The present study was supported by a grant from the Key Programs of the Educational Commission of Hebei Province of China (grant no. ZD2018076) and Youth Science and Technology Project of Health and Municipal Commission of Population and Family of Hebei Province of China (grant no. 20180811).

Availability of data and materials

The datasets generated and/or analyzed during the current study are available in the Gene Expression Omnibus repository [GSE28829 dataset, https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE28829; GSE59421 dataset, https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE59421].

Authors' contributions

WW performed the bioinformatics analyses. ZX mined and downloaded the data from the Gene Expression Omnibus database. XZ analyzed and interpreted the data. XC identified differentially expressed genes targeted by microRNAs associated with coronary heart disease and analyzed the results. WW was a major contributor in the writing of the manuscript. All authors read and approved the final manuscript.

Ethics approval and consent to participate

Not applicable.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

References