Differential gene expression analysis and network construction of recurrent cardiovascular events
- Authors:
- Published online on: December 22, 2015 https://doi.org/10.3892/mmr.2015.4707
- Pages: 1746-1764
Abstract
Introduction
With the significant advances in medication, reperfusion therapy, cardiac rehabilitation and organ transplantation, cardiovascular disease remains one of the major causes of mortality worldwide (1). Evaluation of cardiovascular disease based on risk factors is important in the clinical prevention and treatment of cardiovascular disease, which may alter the risk stratification and guide the treatment and prognosis (2,3). More and more indexes are included in the risk stratification as clinical and experimental research develops, including brain natriuretic peptide, C reactive protein and blood homocysteine. However, the prediction of cardiovascular disease is not so satisfying (4), particularly in personalized prevention and treatment. Sensitivity of risk factors varies in different individuals, and clinical doctors must be aware of this and objective to the current risk factors and stratification (5). More superior and systematic algorithms for stratification remain to be elucidated (6).
The evaluation and stratification of cardiovascular diseases depend more on primary cardiovascular events, which elevate the stratification and enhance the treatment once they occur. However, recurrent cardiovascular events are also vital, which indicate that the current intervention is not marked enough to prevent disease progression. Although patients receive standard treatment based on the risk factors stratification, recurrent cardiovascular events still occur, which indicates that certain individuals are more prone to recurrent cardiovascular events. These patients may require more aggressive therapies, involving susceptibility screening and personalized treatment (7). With the development and application of clinical genomics technology and bioinformatics, novel biomarkers are used in the diagnosis and prognosis of cardiovascular disease (8,9). Previous research revealed that the expression of different genes varies in different stages of cardiovascular diseases, and these genes are involved in the pathological process, and may even predict the cardiovascular events (10). With the help of genomics and bioinformatics, patient susceptibility to recurrent cardiovascular events may be screened out, and personalized treatment can be made. This may reduce the recurrence of cardiovascular events and improve the prognosis. The present study used genomics and bioinformatics technology, and associated software to analyze the differentially expressed genes (DEGs) associated with recurrent cardiovascular events. The present study also aimed to identify the key genes in the pathological process and provide alternative guidance in the preventions, and personalized treatment of recurrent cardiovascular events.
Materials and methods
Microarray data and clinical characteristics
The microarray dataset, GSE48060 with GPL570 [HG-U133_Plus_2] Affymetrix Human Genome U133 Plus 2.0 Array platform, was obtained from the Gene Expression Omnibus (GEO) database (11). The data samples were peripheral blood samples collected from patients with acute myocardial infarction (AMI) 48 h within the primary AMI. All 27 samples were divided into two groups, according to the recurrence of cardiovascular events in the 18 month follow-up. A total of five patients exhibited recurrent cardiovascular events and 22 did not. The definition of recurrent cardiovascular events is recurrent myocardial infarction, re-vascularization, evidence of restenosis, hospitalization for unstable angina or heart failure, cardiovascular mortality, stroke or transient ischemic attack, or amputation due to peripheral vascular disease.
Raw data processing
All 27 sample files were downloaded from the GEO database and were reanalyzed using R software (version 3.1.1; http://www.r-project.org/). The Affy package was applied to read the probe set data from the CEL files (12). Robust Multiarray Averaging was used to normalize the original data. Following standardization, a total of 54,675 probe set IDs' expression levels in different samples were obtained.
Screening and annotation of the DEGs
The limma package in the R software was used to compare the expression levels of the probe sets between the two groups (13). The threshold was set as P<0.05 or a fold change >1.5. The annotate package was used to annotate the DEGs.
Enrichment analysis of DEGs
GeneCodis online tools (http://genecodis2.dacya.ucm.es/) were used to annotate and analyze the DEGs (14,15). The annotation and analysis were predominantly focussed on the molecular function, the biological process and the cellular component of Gene Ontology (GO) enrichment and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis. The threshold was set as P<0.05.
Protein-protein interaction (PPI) network analysis
Cytoscape (version 3.1.1; The Cytoscape Consortium, San Diego, CA, USA) and reactome plugin were applied to analyze the DEGs (16), and to construct and visualize the PPI network. Further analysis of the key nodes in the PPI network were processed.
Results
Clinical characteristics of the two group samples
According to Suresh et al (11), the clinical characteristics between the recurrence and no recurrence groups, including age, sex, body mass index, cardiovascular risk factor score, lipid profile and severity of AMI, were similar with the exception of the usage of beta-blockers. The clinical characteristic details are listed in Table I.
Table IBaseline clinical characteristics of AMI patients with or without recurrent events following primary AMI, who underwent whole-genome blood gene expression microarray analysis. |
Recurrent cardiovascular event-associated DEGs
By comparing the two group samples of with or without recurrent cardiovascular events in the 18 month follow-up following primary AMI, 1,329 genes (2.43% of total probe set) were identified to be differentially expressed and annotatable. A total of 1,023 DEGs (76.98%) were upregulated and 306 DEGs (23.02%) were downregulated in the recurrent cardiovascular events group. The top 10 markedly up or downregulated genes with a fold change >1.5 are listed in Tables II and III, respectively.
Significant GO enrichment
To gain insights into the biological roles of the DEGs in recurrent cardiovascular events, a GO categories enrichment analysis was performed using GeneCoDis. GO categories are predominantly in three groups: Biological process, cellular component and molecular function. The significantly enriched GO terms for molecular functions were nucleotide binding (GO:0000166, P=8.24−19) and nucleic acid binding (GO:0003676, P=1.94−07), for biological processes were signal transduction (GO:0007165, P=5.26−08) and regulation of transcription, DNA-dependent (GO:0006355, P=9.19−06), and for cellular component were cytoplasm (GO:0005737, P=8.98−25) and nucleus (GO:0005634, P=1.91−23; Fig. 1).
Significant pathways
KEGG pathway enrichment analysis was performed to further evaluate the biological significance of the DEGs. The most significant pathway in our KEGG analysis was Pathways in cancer (P=0.000336681). Furthermore, Melanoma (P=0.000336681) and Regulation of actin cytoskeleton (P=0.00165229) were revealed to be highly enriched. The top 15 enriched KEGG pathways of the DEGs are listed in Table IV.
PPI network construction and visualization
By analyzing the identified 1,329 DEGs using Cytoscape and the reactome plugin, 330 genes (node) and 796 gene-gene interactions (edge) were identified. The result was visualized in Cytoscape and the majority of the nodes were located within one network. To modify the PPI network, the sizes of the nodes were set according to their interaction density with the other nodes. The node color of the upregulated DEGs were made red and those downregulated were made blue (Fig. 2). The more that one gene interacts with the other genes, the larger the node was and the more central this gene occurs within the network. The genes, GNG4, MAPK8 and PIK3R2 were the three predominantly upregulated genes, while EP300, CREB1 and PIK3CB were the predominantly downregulated genes in the PPI network. The details of the nodes are listed in Table V.
Table VNetwork features of differentially expressed genes included in protein-protein interaction network sorted by degree. |
Discussion
Cardiovascular events are important in the prevention and treatment of cardiovascular diseases. When it occurs in patients with risk factors, the heart function must be re-evaluated, and the prevention and treatment strategy must be adjusted. For patients who had experienced cardiovascular events, the prevention and treatment strategies are not uniform between different regions and hospitals. There are divergences between different area and different grades of hospitals (17), conservative and aggressive strategies are being used, not to mention the circumstances vary among individuals, efficient and effective personalized evaluation and treatment are urged (18). Previous research has revealed that whole-genome sequencing can be used in cardiovascular disease risk-prediction algorithms, to more accurately forecast whether patients will develop disease (19). However, there remains a lack of research about microarray profiling in recurrent cardiovascular events. The present study performed a microarray profiling of peripheral blood samples from patients with AMI, downloaded from the GEO database, to focus on the DEGs of those with or without recurrent cardiovascular events 18 months following AMI.
R is an integrated suite of software facilities for data manipulation, calculation and graphical display. Using R software and certain packages, the present study identified the DEGs between patients with AMI, with or without recurrent cardiovascular disease. A total of 1,329 genes were identified and 1,023 were upregulated in recurrent group compared with the no recurrent group, while 306 of them were downregulated. The genes with the most significant P-value and fold change >1.5 in the up and downregulated DEGs are listed in Tables II and III. Among them, TUBB1 (tubulin β1, class VI; P=0.00544; fold change=1.56) encodes a member of the β tubulin protein family, and this protein is specifically expressed in platelets and megakaryocytes, and may be involved in proplatelet production and platelet release. Previous research revealed that the prevalence of TUBB1 was higher among healthy individuals compared with patients with cardiovascular disease (20). This may be associated with the TUBB1 function of suppressing microtubule dynamics, fragmenting microtubules and inhibiting cell division (21). Although there is little previous research about other significant genes involved in cardiovascular diseases, the method in the present study may be the initial and alternative way to explore the pathological mechanism of recurrent cardiovascular events.
To further investigate the roles of the DEGs identified in the pathological mechanism of recurrent cardiovascular events, GO enrichment analysis and KEGG pathway analysis was used. GO is widely used as the tool for the organization and functional annotation of molecular aspect (22). It was revealed that the significantly enriched GO terms for molecular functions were nucleotide binding and nucleic acid binding, for biological processes were signal transduction and regulation of transcription (DNA-dependent), and for cellular component were cytoplasm and nucleus. The GO terms mentioned above are basic and vital to the biological and pathological process. Fibroblast growth factor receptor signaling pathway (GO:0008543; P=0.00151494), blood coagulation (GO:0007596; P=0.00166723) and cell adhesion (GO:0007155; P=00170222) were also significantly enriched in GO biological process. Ronca et al (23) reported that fibroblast growth factor receptor-1 gene knockout impairs cardiac and haematopoietic development in murine embryonic stem cells, and the fibroblast growth factor receptor is required for cardiomyocyte differentiation. Yukawa et al (24) demonstrated that impaired fibroblast growth factor receptor gene would suppress the growth of vascular smooth muscle. As for blood coagulation and cell adhesion, which are associated with the formation and breaking off of thrombosis, they are important in both primary and recurrent cardiovascular events.
In KEGG pathway analysis, regulation of actin cytoskeleton is significantly enriched. Actin cytoskeleton is involved in the inward remodeling process associated with cytoskeletal modifications. It is also involved in reducing the passive diameter of resistance vessels, which are the vascular components of the circulatory system, and exert a preponderant role in the regulation of blood flow and the modulation of blood pressure (25). Therefore, the regulation of actin cytoskeleton may have profound consequences on the incidence of cardiovascular events.
The results from PPI network analysis of the top 10 up and downregulated DEGs revealed the significant nodes, including GNG4, MAPK8, PIK3R2, EP300, CREB1 and PIK3CB. MAPK8 is one member of the MAPK family, which has vast implications in signaling and crosstalk with other signaling networks. The MAPK signal pathway is highly associated with mitochondria, the power houses of the cell, which provide >80% of ATP for normal cardiomyocyte function and have a crucial role in cell death (26). EP300 is the node with the most interactions with other nodes in the PPI network, and previous research revealed that it is associated with arterial stiffness prior to hypertension, increased pulse pressure, and structural vessel wall changes (27). CREB1, also termed CREB, phosphorylation induced by the prostacyclin/IP pathway may suppress cardiac fibrosis, which is a consequence of numerous cardiovascular diseases, and contributes to impaired ventricular function (28). The PPI results suggested that MAPK8, EP300 and CREB1 may be important in the development of recurrent cardiovascular events.
The results from the present study suggested that DEGs exist between patients with AMI, with and without recurrent cardiovascular events. These genes are involved in different GO enrichment terms and signaling pathways, from which insights into the pathological processes of recurrent events can be obtained. Several genes, including TUBB1, GNG4, MAPK8, PIK3R2, EP300 and CREB1, with or without previous research, may provide potential candidates for distinguishing the susceptibility to recurrent cardiovascular events in the future. Therefore, the present research may provide important references for the prevention and treatment strategies in patients with primary cardiovascular events. Nevertheless, the genes and the associated GO enrichment terms and pathways identified here require further investigation and confirmation.
In conclusion, the present study revealed the underlying molecular differences between patients with AMI, with and without recurrent cardiovascular events, including DEGs, their biological function, signaling pathways and key genes in the PPI network, which may contribute to the prevention of recurrent events and personalized treatment for primary cardiovascular events. Further functional studies may provide additional insights into the role of the DEGs in the pathological process of recurrent cardiovascular events.
Acknowledgments
The present research was supported by a grant from the National Natural Science Foundation of China (no. 81173166).
References
Go AS, Mozaffarian D, Roger VL, Benjamin EJ, Berry JD, Blaha MJ, Dai S, Ford ES, Fox CS, Franco S, et al: Heart disease and stroke statistics-2014 update: A report from the American heart association. Circulation. 129:e28–e292. 2014. View Article : Google Scholar | |
Goff DC Jr, Lloyd-Jones DM, Bennett G, Coady S, D'Agostino RB, Gibbons R, Greenland P, Lackland DT, Levy D, O'Donnell CJ, et al: 2013 ACC/AHA guideline on the assessment of cardiovascular risk: A report of the American college of cardiology/American heart association task force on practice guidelines. Circulation. 129(25 Suppl 2): S49–S73. 2014. View Article : Google Scholar | |
Perk J, De Backer G, Gohlke H, Graham I, Reiner Z, Verschuren M, Albus C, Benlian P, Boysen G, Cifkova R, et al: European Guidelines on cardiovascular disease prevention in clinical practice (version 2012). The fifth joint task force of the european society of cardiology and other societies on cardiovascular disease prevention in clinical practice (constituted by representatives of nine societies and by invited experts). Eur Heart J. 33:1635–1701. 2012. View Article : Google Scholar : PubMed/NCBI | |
Ceška R and Štulc T: Implementation of cardiovascular disease prevention guidelines into clinical practice: An unmet challenge? Curr Pharm Des. 21:1180–1184. 2015. View Article : Google Scholar | |
van Staa TP, Gulliford M, Ng ES, Goldacre B and Smeeth L: Prediction of cardiovascular risk using framingham, ASSIGN and QRISK2: How well do they predict individual rather than population risk? PLoS One. 9:e1064552014. View Article : Google Scholar : PubMed/NCBI | |
Taljaard M, Tuna M, Bennett C, Perez R, Rosella L, Tu JV, Sanmartin C, Hennessy D, Tanuseputro P, Lebenbaum M and Manuel DG: Cardiovascular disease population risk tool (CVDPoRT): Predictive algorithm for assessing CVD risk in the community setting. A study protocol. BMJ Open. 10:e0067012014. View Article : Google Scholar | |
Cui J, Forbes A, Kirby A, Marschner I, Simes J, Hunt D, West M and Tonkin A: Semi-parametric risk prediction models for recurrent cardiovascular events in the LIPID study. BMC Med Res Methodol. 10:272010. View Article : Google Scholar : PubMed/NCBI | |
Xu F, Teng X, Yuan X, Sun J, Wu H, Zheng Z, Tang Y and Hu S: LCK: A new biomarker candidate for the early diagnosis of acute myocardial infarction. Mol Biol Rep. 41:8047–8053. 2014. View Article : Google Scholar : PubMed/NCBI | |
Duan L, Xiong X, Liu Y and Wang J: miRNA-1: Functional roles and dysregulation in heart disease. Mol Biosyst. 10:2775–2782. 2014. View Article : Google Scholar : PubMed/NCBI | |
Tikkanen E, Havulinna AS, Palotie A, Salomaa V and Ripatti S: Genetic risk prediction and a 2-stage risk screening strategy for coronary heart disease. Arterioscler Thromb Vasc Biol. 33:2261–2266. 2013. View Article : Google Scholar : PubMed/NCBI | |
Suresh R, Li X, Chiriac A, Goel K, Terzic A, Perez-Terzic C and Nelson TJ: Transcriptome from circulating cells suggests dysregulated pathways associated with long-term recurrent events following first-time myocardial infarction. J Mol Cell Cardiol. 74:13–21. 2014. View Article : Google Scholar : PubMed/NCBI | |
Gautier L, Cope L, Bolstad BM and Irizarry RA: Affy-analysis of Affymetrix GeneChip data at the probe level. Bioinformatics. 20:307–315. 2004. View Article : Google Scholar : PubMed/NCBI | |
Smyth GK, Michaud J and Scott HS: Use of within-array replicate spots for assessing differential expression in microarray experiments. Bioinformatics. 21:2067–2075. 2005. View Article : Google Scholar : PubMed/NCBI | |
Nogales-Cadenas R, Carmona-Saez P, Vazquez M, Vicente C, Yang X, Tirado F, Carazo JM and Pascual-Montano A: GeneCodis: Interpreting gene lists through enrichment analysis and integration of diverse biological information. Nucleic Acids Res. 37:W317–W322. 2009. View Article : Google Scholar : PubMed/NCBI | |
Carmona-Saez P, Chagoyen M, Tirado F, Carazo JM and Pascual-Montano A: GENECODIS: A web-based tool for finding significant concurrent annotations in gene lists. Genome Biol. 8:R32007. View Article : Google Scholar : PubMed/NCBI | |
Saito R, Smoot ME, Ono K, Ruscheinski J, Wang PL, Lotia S, Pico AR, Bader GD and Ideker T: A travel guide to cytoscape plugins. Nat Methods. 9:1069–1076. 2012. View Article : Google Scholar : PubMed/NCBI | |
Mortensen MB and Falk E: Real-life evaluation of European and American high-risk strategies for primary prevention of cardiovascular disease in patients with first myocardial infarction. BMJ Open. 4:e0059912014. View Article : Google Scholar : PubMed/NCBI | |
Montecucco F, Carbone F, Dini FL, Fiuza M, Pinto FJ, Martelli A, Palombo D, Sambuceti G, Mach F and De Caterina R: Implementation strategies of systems medicine in clinical research and home care for cardiovascular disease patients. Eur J Intern Med. 25:785–794. 2014. View Article : Google Scholar : PubMed/NCBI | |
Musunuru K: Personalized genomes and cardiovascular disease. Cold Spring Harb Perspect Med. 5:a0140682014. View Article : Google Scholar : PubMed/NCBI | |
Freson K, De Vos R, Wittevrongel C, Thys C, Defoor J, Vanhees L, Vermylen J, Peerlinck K and Van Geet C: The TUBB1 Q43P functional polymorphism reduces the risk of cardiovascular disease in men by modulating platelet function and structure. Blood. 106:2356–2362. 2005. View Article : Google Scholar : PubMed/NCBI | |
Yang H, Ganguly A, Yin S and Cabral F: Megakaryocyte lineage-specific class VI β-tubulin suppresses microtubule dynamics, fragments microtubules, and blocks cell division. Cytoskeleton (Hoboken). 68:175–187. 2011. View Article : Google Scholar | |
Lovering RC, Camon EB, Blake JA and Diehl AD: Access to immunology through the gene ontology. Immunology. 125:154–160. 2008. View Article : Google Scholar : PubMed/NCBI | |
Ronca R, Gualandi L, Crescini E, Calza S, Presta M and Dell'Era P: Fibroblast growth factor receptor-1 phosphorylation requirement for cardiomyocyte differentiation in murine embryonic stem cells. J Cell Mol Med. 13:1489–1498. 2009. View Article : Google Scholar : PubMed/NCBI | |
Yukawa H, Miyatake SI, Saiki M, Takahashi JC, Mima T, Ueno H, Nagata I, Kikuchi H and Hashimoto N: In vitro growth suppression of vascular smooth muscle cells using adenovirus-mediated gene transfer of a truncated form of fibroblast growth factor receptor. Atherosclerosis. 141:125–132. 1998. View Article : Google Scholar : PubMed/NCBI | |
Castorena-Gonzalez JA, Staiculescu MC, Foote C and Martinez-Lemus LA: Mechanisms of the inward remodeling process in resistance vessels: Is the actin cytoskeleton involved? Microcirculation. 21:219–229. 2014. View Article : Google Scholar : PubMed/NCBI | |
Javadov S, Jang S and Agostini B: Crosstalk between mitogen-activated protein kinases and mitochondria in cardiac diseases: Therapeutic perspectives. Pharmacol Ther. 144:202–225. 2014. View Article : Google Scholar : PubMed/NCBI | |
Herrera VL, Decano JL, Giordano N, Moran AM and Ruiz-Opazo N: Aortic and carotid arterial stiffness and epigenetic regulator gene expression changes precede blood pressure rise in stroke-prone Dahl salt-sensitive hypertensive rats. PLoS One. 9:e1078882014. View Article : Google Scholar : PubMed/NCBI | |
Chan EC, Dusting GJ, Guo N, Peshavariya HM, Taylor CJ, Dilley R, Narumiya S and Jiang F: Prostacyclin receptor suppresses cardiac fibrosis: Role of CREB phosphorylation. J Mol Cell Cardiol. 49:176–185. 2010. View Article : Google Scholar : PubMed/NCBI |