Open Access

Bioinformatics analysis of the regulatory lncRNA‑miRNA‑mRNA network and drug prediction in patients with hypertrophic cardiomyopathy

  • Authors:
    • Jiajianghui Li
    • Zining Wu
    • Deqiang Zheng
    • Yue Sun
    • Sisi Wang
    • Yuxiang Yan
  • View Affiliations

  • Published online on: May 23, 2019     https://doi.org/10.3892/mmr.2019.10289
  • Pages: 549-558
  • Copyright : © Li et al. This is an open access article distributed under the terms of Creative Commons Attribution License [CC BY 4.0].

Metrics: Total Views: 0 (Spandidos Publications: | PMC Statistics: )
Total PDF Downloads: 0 (Spandidos Publications: | PMC Statistics: )


Abstract

Hypertrophic cardiomyopathy (HCM) is a complex inherited cardiovascular disease. The present study investigated the long noncoding (lnc)RNA/microRNA (mi)RNA/mRNA expression pattern of patients with HCM and aimed to identify key molecules involved in the development of this condition. An integrated strategy was conducted to identify differentially expressed miRNAs (DEmiRs), differentially expressed lncRNAs (DElncs) and differentially expressed genes (DEGs) based on the GSE36961 (mRNA), GSE36946 (miRNA), GSE68316 (lncRNA/mRNA) and GSE32453 (mRNA) expression profiles downloaded from the Gene Expression Omnibus datasets. Bioinformatics tools were employed to perform function and pathway enrichment analysis, protein‑protein interaction, lncRNA‑miRNA‑mRNA and hub gene networks. Subsequently, DEGs were used as targets to predict drugs. The results indicated that a total of 2,234 DElncs (1,120 upregulated and 1,114 downregulated), 5 DEmiRs (2 upregulated and 3 downregulated) and 42 DEGs (35 upregulated and 7 downregulated) were identified in 4 microarray profiles. Gene ontology analysis revealed that DEGs were mainly involved in actin filament and stress fiber formation and in calcium ion binding, whereas Kyoto Encyclopedia of Genes and Genomes pathway analysis identified the hypoxia inducible factor‑1, transforming growth factor‑β and tumor necrosis factor signaling pathways as the main pathways involved in these processes. The hub genes were screened using cytoHubba. A total of 1,086 lncRNA‑miRNA‑mRNA interactions including 67 lncRNAs, 5 miRNAs and 25 mRNAs were mined in the present study based on prediction websites. Drug prediction indicated that the targeted drugs mainly included angiotensin converting enzyme inhibitors or β‑blockers. A comprehensive bioinformatics analysis of the molecular regulatory lncRNA‑miRNA‑mRNA network was performed and potential therapeutic applications of drugs were predicted in HCM patients. The data may unravel the future molecular mechanism of HCM.

Introduction

Hypertrophic cardiomyopathy (HCM) is a heterogeneous monogenic myocardial disorder that is characterized by myocardial hypertrophy, asymmetric hypertrophy of the ventricular septum, decreased ventricular cavity and abnormal hypertrophy of cardiac muscle cells. The prevalence of HCM is >1 in 500 (0.2%) (1). Previous studies carried out in different regions and ethnic groups indicated that this disease exhibited a high degree of familial aggregation and heritability (24). It has been proposed that gene therapy is an effective way to eradicate HCM.

With the development of high-throughput gene expression profiling technology, microarray analysis has been widely used in the early diagnosis, treatment and prognosis of several diseases. MicroRNAs (miRNAs) are a novel class of small non-coding RNAs, which can negatively regulate gene expression at the post-transcriptional level by directly binding to mRNAs (5). Long non-coding RNAs (lncRNAs) and circular RNAs contain miRNA response elements, acting as competitors of endogenous RNAs (6). These molecules have emerged as essential regulatory molecules in various biological processes. In 2014, Song et al (7) performed a miRNA microarray analysis on heart tissues from 7 HCM patients and 5 healthy subjects and found that miR-451 was the main miRNA that was downregulated in the HCM subjects. The target gene of miR-451 was the tuberous sclerosis complex 1 gene (TSC1) that was significantly upregulated in the myocardial tissue of HCM patients (7). TSC1 is a known positive regulator of autophagy (8), which contributes to the development of HCM. However, the molecular mechanism of HCM with regard to non-coding RNAs is still unknown. Therefore, the present study aimed to perform in-depth data mining based on former microarray studies.

At present, big data mining and precision medicine have gained considerable attention. Digitization and informatics technologies facilitate the flow of information between various fields, laying the foundation for the conduct of interdisciplinary research. With the achievements of transcriptomics and genomics, the correlation between disease phenotypes and potential genotypes has attracted great scientific interest. The present study combined bioinformatics, transcriptomics and epidemiology in order to examine differentially expressed lncRNAs (DElncs), differentially expressed miRNAs (DEmiRs) and differentially expressed genes (DEGs). In addition to their expression, their associated regulatory network was investigated. The data can elucidate the underlying etiology of HCM and further provide reliable molecular targets for drug therapy.

Materials and methods

Raw data collection

LncRNA, miRNA and gene microarray expression profiles between HCM and healthy controls were investigated by the Gene Expression Omnibus database (GEO; http://www.ncbi.nlm.nih.gov/geo/) using the keywords ‘hypertrophic cardiomyopathy’ and ‘lncRNA’ or ‘miRNA’ or ‘mRNA’. Eligible datasets had to meet the following inclusion criteria: i) Microarray profiling studies on human patients with HCM; ii) HCM and non-HCM control/healthy samples for comparison; iii) reports of sample sizes; iv) a group label for each sample size; v) a corresponding annotation (gene symbol) or GeneBank ID in the platform file for each probe of the microarray and vi) availability of raw data. The exclusion criteria included: i) Animal samples or cell lines; ii) non-HCM samples; iii) non-microarray profiles; iv) samples without controls and v) probes without gene symbol or GeneBank ID annotation. Finally, 4 profiles of GSE36961, GSE36946 and GSE68316 were downloaded. The flow chart of the screening datasets is presented in Fig. 1. The characteristics of these microarray expression profiles are shown in Table I.

Table I.

Characteristics for GEO microarray in HCM patients.

Table I.

Characteristics for GEO microarray in HCM patients.

No. of GEO profileTypeSourceCaseControlPlatformAnnotation of platform
GSE36961mRNACardiac tissue10639GPL15389Illumina HumanHT-12 V3.0 expression beadchip
GSE36946miRNACardiac tissue10720GPL8179Illumina human v2 microRNA expression beadchip
GSE68316lncRNACardiac tissue  7  5GPL20113CapitalBio Human LncRNA Microarray v2.0
GSE68316mRNACardiac tissue  7  5GPL20113CapitalBio Human LncRNA Microarray v2.0
GSE32453mRNACardiac tissue  8  5GPL6104Illumina humanRef-8 v2.0 expression bead chip

[i] GEO, Gene Expression Omnibus; HCM, hypertrophic cardiomyopathy; lnc, long noncoding; miRNA, microRNA.

Raw data pretreatment and screening for DElncs, DEmiRs and DEGs

The GEO dataset is one of the mainstream public functional genomics data repositories (9). Nonetheless, the uploaded data are usually rough, incomplete or containing noise. Therefore, prior to data mining, data pretreatment must be carried out including data washing, data filtering and data normalization. With regard to the Illumina expression array, the MicroArray Quality Control method is conventional for background correction and quantile normalization (10). This method can eliminate non-experimental differences caused by technical discrepancies and ensure appropriate data comparisons among different samples.

The raw signal intensity data of GSE36961, GSE36946 and GSE32453 were converted by the Illumina package using the R software (version 3.5.1; http://www.r-project.org/). The Limma package was employed to identify DElncs, DEmiRs and DEGs, respectively using a linear model and the empirical Bayes model (11). GSE68316 was analyzed by the R-based GEO online tool GEO2R (http://www.ncbi.nlm.nih.gov/geo/geo2r/), which is useful for comparing 2 or more groups of samples, and can characterize differentially expressed RNA molecules based on t-test or analysis of variance (9). The gene symbol was annotated to the corresponding probes and for the genes mapped with more than one probe, the average expression values were calculated. Volcano plots of DElncs, DEmiRs and DEGs were provided using Gplot package in R.

A total of 3 mRNA datasets (GSE36961, GSE68316 and GSE32453) were selected to identify novel DEGs; however, the DEGs in each dataset were not combined due to the variation of distinct platforms, manufacturers and temperature (12). The data obtained from multiple sources could be compared or directly integrated. If not properly addressed, this integration would adversely affect all subsequent analysis. The ‘SVA’ package in the R software is an effective tool to reduce the data heterogeneity, remove batch effects and retain the variation due to sample types retained for further DEG analysis (12). The hierarchical clustering plot for the DEGs in the GSE36961, GSE68316 and GSE32453 datasets were provided using the ‘pheatmap’ package of the R software (https://cran.r-project.org/web/packages/pheatmap/index.html) following batch normalization. Ultimately, the mixed model of variance analysis with a false discovery rate (Benjamini-Hochberg test, FDR) (13) was used. An adjusted P<0.05 and a |log2FC| value >1 were applied in screening significantly different expression levels of RNA molecules.

Predicting potential lncRNAs and genes

Initially, the identified DEmiRs were used to predict lncRNAs. The lncRNA-miRNA interactions were based on the following 3 lncRNA target prediction websites including DIANA-LncBase (version 2; www.microrna.gr/LncBase) (14), LNCipedia (version 5.0; http://www.lncipedia.org) (15) and starBase (version 3.0; http://starbase.sysu.edu.cn/) (16). At least 2 of these websites included eligible predicted lncRNAs binding sites with the DEmiRs. The predicted lncRNAs obtained from the 3 websites were intersected with the identified DElncs (17), whereas the overlapping lncRNAs were retained for further analysis. Similarly, the miRNA-mRNA pairs were acquired from TargetScan version 7.1 (http://www.targetscan.org/), miRTarBase 7.0 (http://mirtarbase.mbc.nctu.edu.tw/php/index.php) (18) and miRBase 21 (http://www.mirbase.org) (19). The putative DElncs and DEGs that shared miRNA binding loci by target prediction were considered as one potential lncRNA-miRNA-mRNA interaction (17,20,21).

Functional enrichment and pathway analysis

To explore the main functions and pathways of DEGs, the Database for Annotation, Visualization and Integrated Discovery (DAVID; http://david.ncifcrf.gov/) was used for the Gene Ontology (GO) (22,23) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway (24,25) enrichment analyses. DAVID is an integrated biological database which contains gene functional classification, pathway-mining and clustering tools. These functions aim to systematically extract biological information from the uploaded gene lists (26).

The human disease database Malacards (http://www.malacards.org/) was selected for the HCM-related GO and KEGG pathway items in order to ensure the accuracy of the results. Malacards is an integrated compendium of human maladies and their annotations are mined from several well-known data sources. Malacards contain 14 annotation topics, including Summaries, Symptoms, Related diseases, Drugs and therapeutics, Gene and Expression, GO terms and Pathways (27). A P-value was adjusted using the Benjamini method or the FDR in multiple testing calibrations. A P<0.05 was selected as the threshold. The enrichment score was defined as the transformed-log10 (P-value).

Protein-protein interaction (PPI) network of DEGs, screening for hub genes and lncRNA-miRNA-mRNA network

Search Tool for the Retrieval of Interacting Genes (STRING; version 10.5; http://string-db.org/) is an online functional protein association network. The associations in STRING include direct (physical) and indirect (functional) interactions, as long as they are specific and biologically meaningful (28). The identified DEGs were input into STRING to unravel a potential PPI network. Hub genes are key genes playing a crucial role in biological processes. The regulation of other genes in the relevant pathway is often affected by hub genes. Therefore, hub genes are often considered an important target or a research hot spot (29). Hub genes can be screened according to the network topology. CytoHubba is an effective app of the Cytoscape version 3.6.1 (http://www.cytoscape.org/) plug-in used to identify hub genes more accurately by 12 topological analysis methods (30). Cytoscape is an open source software platform for visualizing molecular interaction networks and biological pathways. These networks are integrated with annotations of gene expression profiles and other data. Maximal Clique Centrality (MCC) was used to identify the top 20 hub genes. The paired lncRNA-miRNA-mRNA data were also input into Cytoscape software in order to generate a regulatory network image.

Drug prediction of hub genes

DrugBank (http://www.drugbank.ca) contains more than 7,000 drug entries and 4,000 non-redundant proteins (31). It is a comprehensive cheminformatic database including abundant biochemical and pharmacological details regarding drugs. DrugBank aids the identification of novel drug targets and the comparison of drug structures with potential mechanisms of action (32). The DEGs identified in the 3 mRNA datasets were input into DrugBank to examine their association with potential targeted drugs. The purpose of the present study was to explain the rationale of present drug therapy used for HCM and to explore additional potential target genes for future drug development.

Results

Detection of DElncs, DEmiRs and DEGs

The raw data of each dataset were normalized following data pretreatment. Fig. 2 demonstrates the boxplot of each sample prior to and following data normalization. A total of 2,234 DElncs (1,120 upregulated and 1,114 downregulated) were identified in GSE68316, whereas 5 DEmiRs were identified in GSE36946, of which 2 were upregulated (miR-373 and miR-514) and 3 were downregulated (miR-10a, miR-144 and miR-30c-5p). A total of 154 DEGs were identified in GSE36961, of which 47 were upregulated and 109 downregulated. A total of 1,402 DEGs (365 upregulated and 1,037 downregulated) were identified in GSE68316. Finally, 16 DEGs (8 upregulated and 8 downregulated) were identified in GSE32453. The volcano plots (Fig. 3) displayed the aberrantly expressed RNA molecules. A total of 42 DEGs (35 upregulated and 7 downregulated) were finally screened in the 3 mRNA microarray datasets (GSE36961, GSE68316 and GSE32453) following data merging and removing the batch effect (Fig. 4). Following prediction and intersection process, 112 non-redundant lncRNA-miRNA pairs and 49 miRNA-mRNA pairs containing 67 lncRNAs, 5 miRNAs and 25 mRNAs were mined in the present study. Ultimately, 1,086 pairs of lncRNA-miRNA-mRNA interactions were identified. The details of the predicted lncRNAs and mRNAs that were based on several prediction websites are listed in Tables SI and SII.

GO and KEGG analysis

GO and KEGG analyses were performed on the detected 42 DEGs to examine their biological functions in detail. GO analysis described the results from 3 categories: ‘Biological processes’ (BP), ‘cellular components’ (CC) and ‘molecular functions’ (MF) (33). GO/KEGG analysis is considered to be a powerful tool in revealing biological mechanisms or functional pathways underlying observed patterns in genomics or transcriptomics. A total of 36 GO terms (19 BP, 13 CC and 4 MF) and 30 KEGG terms were enriched. The significantly enriched HCM-related GO terms mainly included the following: Actin filament binding, stress fiber formation, calcium ion binding and transforming growth factor-β receptor binding (Fig. 5A). The top 10 KEGG pathways correlated highly with HCM. For example, the deficiency of tumor necrosis factor (TNF) receptor-associated factor 5 could substantially aggravate cardiac hypertrophy, cardiac dysfunction and fibrosis (34). TNF-α is an extremely important molecule used in cell proliferation, differentiation, growth and metabolism, which is closely related to the occurrence of cardiac hypertrophy (35). Hypoxia-inducible factor (HIF) can be induced during hemodynamical-mediated hypertrophic growth and its induction is accompanied by pathological stress (36,37). These molecules play important roles in the identified pathways. The bubble plot offered a visual representation of the aforementioned pathways (Fig. 5B).

Construction of PPI, hub gene and lncRNA-miRNA-mRNA networks

The PPI network analysis aimed to study the molecular mechanism of diseases and identify novel drug targets from a systematic perspective. The STRING database is designed to depict functional interactions between the expressed proteins by integrating known and predicted protein-protein association data among various species (28). The PPI network was conducted by STRING to explore the interactions between proteins encoded by 42 DEGs. A median confidence 0.4 was selected as a cut-off criterion. A total of 39 nodes (proteins encoded by genes) and 80 edges (connections between nodes) were screened following removal of the disconnected nodes in the network (Fig. 6A).

Furthermore, 1,086 pairs of lncRNA-miRNA-mRNA biomolecular interactions were integrated in a single network constructed by Cytoscape. The lncRNAs exhibited high tendency of aggregation with the miRNAs. miR-30c-5p and miR-541 exhibited the highest number of lncRNA binding sites, suggesting that these 2 miRNAs could bind to lncRNAs when competing with other miRNAs. It is worth mentioning that lncRNA ENSG00000269821 and ENSG00000224078 have binding sites for all the five miRNAs, which are thought to be of substantial significance and may provide clues to future researchers (Fig. 6B).

In the PPI network, some hub genes/proteins are highly connected with other proteins, suggesting a central regulatory role. CytoHubba provides a user-friendly interface to examine the interactions among hub nodes in the biological network by topological analysis (30), which can aid the identification of the essential networks involved. The top 20 hub genes were mined by the MCC method (Fig. 6C).

Drug Prediction of DEGs

HCM can be divided into hypertrophic obstructive cardiomyopathy (HOCM) and hypertrophic nonobstructive cardiomyopathy according to the presence of obstruction in the left ventricular outflow tract. Medical treatment should be the first choice for the majority of patients. Therefore, it is vital to select the most suitable drug candidate. At present, the main therapeutic drugs for HOCM are β-blockers, calcium channel blockers and amiodarone (38).

The DrugBank database is a unique, comprehensive drug repository covering detailed drug function, formulation, basic structure and drug target information (31). The DEGs were inserted into DrugBank and the targeted drugs of these genes were predicted. The approved drugs that were associated with HCM were selected from 536 results (Table II). The majority of these drugs were calcium channel blockers, angiotensin converting enzyme inhibitors (ACEIs) and/or β-blockers. The drug effects included the following: Controlling heart rate or arrhythmia, increasing ventricular filling and terminal diastolic volume, reducing the contractility of ventricle and improving the compliance of myocardium (39). These effects were consistent with the guidelines of HCM diagnosis and treatment. However, ACEIs, angiotensin-receptor blockers (ARBs) or diuretics can enhance myocardial contractility or reduce cardiac afterload, thereby increasing left ventricular outflow tract obstruction. Therefore, HOCM patients should be treated with caution.

Table II.

The predicted drugs of DEGs.

Table II.

The predicted drugs of DEGs.

NameAccession numberGroups Descriptiona
CaptoprilDB01197ApprovedCaptopril is a potent, competitive inhibitor of ACE and may be used in the treatment of hypertension.
RamiprilDB00178ApprovedRamipril is a prodrug belonging to the ACEI class of medications, may be used in the treatment of hypertension, myocardial infarction, stroke.
LabetalolDB00598ApprovedBlocker of both α- and β-adrenergic receptors that is used as an antihypertensive.
VerapamilDB00661ApprovedA calcium channel blocker that is a class IV anti-arrhythmia agent.
MexiletineDB00379Approved, InvestigationalAntiarrhythmic agent pharmacologically similar to lidocaine. It may have some anticonvulsant properties.
NicardipineDB00622Approved, InvestigationalA potent calcium channel blockader with marked vasodilator action.
PropranololDB00571Approved, InvestigationalA widely used non-cardioselective β-adrenergic antagonist. Used in acute myocardial infarction, arrhythmias, hypertension.
DiltiazemDB00343Approved, InvestigationalA benzothiazepine derivative with vasodilating action due to its antagonism of the actions of the calcium ion in membrane functions.
Amyl NitriteDB01612ApprovedAn antihypertensive medicine. Amyl nitrite is bioactive in mammals, being a vasodilator which is the basis of its use as a prescription medicine.
PropafenoneDB01182ApprovedAn antiarrhythmia agent that is particularly effective in ventricular arrhythmias. It also has weak β-blocking activity.
NifedipineDB01115ApprovedBoth a long and short-acting 1,4-dihydropyridine calcium channel blocker, preventing calcium-dependent myocyte contraction and vasoconstriction.
LabetalolDB00598ApprovedBlocker of both α- and β-adrenergic receptors that is used as an antihypertensive.
RamiprilDB00178ApprovedRamipril is a prodrug belonging to the ACEI class of medications, may be used in the treatment of hypertension, myocardial infarction, stroke
NimodipineDB00393Approved, InvestigationalA calcium channel blocker. It acts primarily on vascular smooth muscle cells by stabilizing voltage-gated L-type calcium channels in their inactive conformation.
BenazepriDB00542Approved, InvestigationalCan be converted into benazeprilat, a non-sulfhydryl ACEI. A medi cation used to treat hypertension, congestive heart failure and chronic renal failure.
ValsartanDB00177Approved, InvestigationalValsartan is an ARB that may be used to treat hypertension, isolated systolic hypertension, left ventricular hypertrophy.
EphedrineDB01364Approved, InvestigationalAffect the rate or intensity of cardiac contraction, blood vessel diameter, or blood volume.
VerapamilDB00661ApprovedA calcium channel blocker that is a class IV anti-arrhythmia agent.
DigoxinDB00390ApprovedControl ventricular rate in atrial fibrillation and in the management of congestive heart failure with atrial fibrillation.
DiltiazemDB00343Approved, InvestigationalA benzothiazepine derivative with vasodilating action due to its antagonism of the actions of the calcium ion in membrane functions.
TimololDB00373ApprovedA β-adrenergic antagonist similar in action to propranolol. Timolol has been proposed as an antihypertensive, antiarrhythmic and anti angina agent.
LisinoprilDB00722Approved, InvestigationalLisinopril is a potent, competitive inhibitor of ACE.
BenazeprilDB00542Approved, InvestigationalCan be converted into benazeprilat, a non-sulfhydryl ACEI. A medi cation used to treat hypertension, congestive heart failure and chronic renal failure.
PenbutololDB01359Approved, InvestigationalPenbutolol is a drug in the β-blocker class used to treat hypertension.

a The description of drug was extracted from DrugBank. ACE, angiotensin-converting enzyme; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin-receptor blocker; DEG, differentially expressed genes.

Discussion

HCM is a primary cardiovascular disease, which has been regarded as the most common risk factor of sudden death among young people; it is now well accepted that multiple mutations in gene encoding regions are responsible for the development of this disease (40). Therefore, investigating the etiology of HCM and identifying effective therapeutic indicators is of utmost urgency.

In recent years, several studies (41,42) have investigated the functions and clinical implications of non-coding RNAs in HCM. However, the impact of RNA crosstalk on HCM has not been previously addressed. The present study integrated the lncRNA, miRNA and mRNA expression profiles and produced an integrated lncRNA-miRNA-mRNA regulatory network, which provided insight at the post-transcriptional level of gene regulation. Moreover, key molecules involved in multiple physiopathological processes of HCM were explored by the application of bioinformatics technology and big data mining. Targeted drugs were predicted by DrugBank using identified hub genes, indicating that they may be optimal candidate drugs for future therapy.

The identified DEmiRs and DEGs exhibited consistency with other microarray analysis on cardiovascular disease-associated targets. For example, the underexpressed miR-373 and overexpressed miR-10 were consistently expressed in the plasma using microarrays and further validated by quantitative PCR analysis in 55 HCM patients with a moderate diagnostic value (43). The high throughput sequencing identified miR-30c-5p and specific mRNA targets that could affect heart cells by causing nuclear factor of activated T-cells hypertrophy and activating cardiac hypertrophy signaling (44). The plasma levels of miR-144-3p were elevated in patients with ventricular arrhythmia (45). ALOX5AP is a crucial regulatory factor used in the biosynthesis of inflammatory leukotrienes. Previous studies have reported that genetic variations in ALOX5AP exhibit significant associations with ischemic stroke and myocardial infarction (46,47). NFKBIA polymorphisms modulated the risk of coronary artery disease in the Chinese Uygur population by regulating various cytokines including interleukin-6, which is a key mediator of the inflammatory process and atherosclerotic plaque formation (48). These studies were consistent with the present findings.

With regard to the analysis of lncRNAs, the results of the present study were not fully investigated in previous studies. The Myosin Heavy Chain Associated RNA Transcripts (Mhrt) lncRNA was reported as cardiac-specific and abundant in adult hearts (49). Mhrt protected the heart from hypertrophy and failure by interacting with chromatin (49). Yang et al (50) performed lncRNA and mRNA microarray analysis on myocardial tissues from 7 HCM patients and 5 controls, and identified 965 upregulated and 461 downregulated lncRNAs. Bioinformatics analysis indicated that lncRNA-co-expressed mRNAs were mainly enriched in ribosome and oxidative phosphorylation modules. Overexpression of genes in complex I of the oxidative phosphorylation pathway may contribute to physiological hypertrophy of the heart.

At present, β-blockers (propranolol and penbutolol) and calcium antagonists (nifedipine, verapamil and diltiazem) are mainly prescribed for pharmacological treatment of HCM in routine clinical practice (51,52). Propranolol was reported to reduce the A-wave ratio in the apex cardiogram and could be indispensable in relieving emergency symptoms in patients with HCM. Calcium antagonists can control myocardial contractility, reduce the pressure gradient of left ventricular outflow tract and improve myocardial compliance and ventricular diastolic function. The majority of the drugs that were predicted in the present study are already in clinical use, suggesting that the screened genes were associated with the pathogenic mechanism of HCM. The remaining predicted drugs that are not currently used for clinical treatment can be the starting point for subsequent future investigations.

Data mining can extract the relevant biological information from high-dimension data. The integration of multiple histological techniques will become a new trend of disease diagnosis in the era of precision medicine. These molecules have the potential to be used as disease biomarkers in personalized medicine for patients and can improve the diagnosis, treatment and prevention of several diseases.

The present study contains certain limitations: i) The lncRNA-miRNA-mRNA large-scale crosstalk network did not indicate whether these RNAs were co-expressed in the same tissues. ii) The present study is a preliminary analysis, which requires validation in a large population by PCR or western blot analyses. In addition, the biological mechanism of specific non-coding RNAs requires verification by animal or cellular studies. iii) The specificity of the identified DElncs, DEmiRs and DEGs for HCM requires investigation in future studies.

In conclusion, the present study generated a holistic view of a candidate lncRNA-miRNA-mRNA network for HCM and proposed potential predicted drugs that could be used for this disease by integrating several microarray data. The authors hope that the present study will be beneficial for discovering new biomarkers or therapeutic drugs for the reduction of the risk of this life-threatening disease.

Supplementary Material

Supporting Data

Acknowledgements

Not applicable.

Funding

No funding was received.

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Authors' contributions

JL performed the data analysis and drafted the manuscript. ZW drafted the manuscript, prepared the figures and contributed towards the study design. DZ and YS were involved in data acquisition and performed data analysis. SW and YY revised the figures, and designed and supervised the study. DZ, YS, SW and YY also contributed to the manuscript revision. All authors have 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

1 

Maron BJ, Gardin JM, Flack JM, Gidding SS, Kurosaki TT and Bild DE: Prevalence of hypertrophic cardiomyopathy in a general population of young adults. Echocardiographic analysis of 4111 subjects in the CARDIA study. Coronary artery risk development in (Young) adults. Circulation. 92:785–789. 1995. View Article : Google Scholar : PubMed/NCBI

2 

Maron BJ, Maron MS and Semsarian C: Genetics of hypertrophic cardiomyopathy after 20 years: Clinical perspectives. J Am Coll Cardiol. 60:705–715. 2012. View Article : Google Scholar : PubMed/NCBI

3 

Semsarian C, Ingles J, Maron MS and Maron BJ: New perspectives on the prevalence of hypertrophic cardiomyopathy. J Am Coll Cardiol. 65:1249–1254. 2015. View Article : Google Scholar : PubMed/NCBI

4 

Helms AS and Day SM: Other side of the coin: The missing heritability in hypertrophic cardiomyopathy. Eur Heart J. 38:3469–3471. 2017. View Article : Google Scholar : PubMed/NCBI

5 

Bartel DP: MicroRNAs: Target recognition and regulatory functions. Cell. 136:215–233. 2009. View Article : Google Scholar : PubMed/NCBI

6 

Tay Y, Rinn J and Pandolfi PP: The multilayered complexity of ceRNA crosstalk and competition. Nature. 505:344–352. 2014. View Article : Google Scholar : PubMed/NCBI

7 

Song L, Su M, Wang SY, Zou Y, Wang X, Wang Y, Cui H, Zhao P, Hui R and Wang J: MiR-451 is decreased in hypertrophic cardiomyopathy and regulates autophagy by targeting TSC1. J Cell Mol Med. 18:2266–2274. 2014. View Article : Google Scholar : PubMed/NCBI

8 

Papadakis M, Hadley G, Xilouri M, Hoyte LC, Nagel S, McMenamin MM, Tsaknakis G, Watt SM, Drakesmith CW, Chen R, et al: Tsc1 (hamartin) confers neuroprotection against ischemia by inducing autophagy. Nat Med. 19:351–357. 2013. View Article : Google Scholar : PubMed/NCBI

9 

Barrett T, Wilhite SE, Ledoux P, Evangelista C, Kim IF, Tomashevsky M, Marshall KA, Phillippy KH, Sherman PM, Holko M, et al: NCBI GEO: Archive for functional genomics data sets-update. Nucleic Acids Res. 41:D991–D995. 2013. View Article : Google Scholar : PubMed/NCBI

10 

Li P, Piao Y, Shon HS and Ryu KH: Comparing the normalization methods for the differential analysis of Illumina high-throughput RNA-Seq data. BMC Bioinformatics. 16:3472015. View Article : Google Scholar : PubMed/NCBI

11 

Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W and Smyth GK: Limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 43:e472015. View Article : Google Scholar : PubMed/NCBI

12 

Parker HS, Bravo HC and Leek JT: Removing batch effects for prediction problems with frozen surrogate variable analysis. PeerJ. 2:e5612014. View Article : Google Scholar : PubMed/NCBI

13 

Benjamini Y and Hochberg Y: Controlling the False Discovery Rate: A practical and powerful approach to multiple testing. J R Statist Soc B. 57:289–300. 1995.

14 

Paraskevopoulou MD, Georgakilas G, Kostoulas N, Reczko M, Maragkakis M, Dalamagas TM and Hatzigeorgiou AG: DIANA-LncBase: Experimentally verified and computationally predicted microRNA targets on long non-coding RNAs. Nucleic Acids Res. 41:D239–D245. 2013. View Article : Google Scholar : PubMed/NCBI

15 

Volders PJ, Anckaert J, Verheggen K, Nuytens J, Martens L, Mestdagh P and Vandesompele J: LNCipedia 5: Towards a reference set of human long non-coding RNAs. Nucleic Acids Res. 47:D135–D139. 2019. View Article : Google Scholar : PubMed/NCBI

16 

Li JH, Liu S, Zhou H, Qu LH and Yang JH: starBase v2.0: Decoding miRNA-ceRNA, miRNA-ncRNA and protein-RNA interaction networks from large-scale CLIP-Seq data. Nucleic Acids Res. 42:D92–D97. 2014. View Article : Google Scholar : PubMed/NCBI

17 

Xing C, Cai ZZ, Gong J, Zhou J, Xu JJ and Guo F: Identification of potential biomarkers involved in gastric cancer through integrated analysis of non-coding RNA associated competing endogenous RNAs network. Clin Lab. 64:1661–1669. 2018. View Article : Google Scholar : PubMed/NCBI

18 

Chou CH, Shrestha S, Yang CD, Chang NW, Lin YL, Liao KW, Huang WC, Sun TH, Tu SJ, Lee WH, et al: miRTarBase update 2018: A resource for experimentally validated microRNA-target interactions. Nucleic Acids Res. 46:D296–D302. 2018. View Article : Google Scholar : PubMed/NCBI

19 

Kozomara A and Griffiths-Jones S: miRBase: Annotating high confidence microRNAs using deep sequencing data. Nucleic Acids Res. 42:D68–D73. 2014. View Article : Google Scholar : PubMed/NCBI

20 

Agarwal V, Bell GW, Nam JW and Bartel DP: Predicting effective microRNA target sites in mammalian mRNAs. Elife. Aug 12–2015.(Epub ahead of print). doi: 10.7554/eLife.05005. View Article : Google Scholar

21 

Xiong DD, Dang YW, Lin P, Wen DY, He RQ, Luo DZ, Feng ZB and Chen G: A circRNA-miRNA-mRNA network identification for exploring underlying pathogenesis and therapy strategy of hepatocellular carcinoma. J Transl Med. 16:2202018. View Article : Google Scholar : PubMed/NCBI

22 

Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, Davis AP, Dolinski K, Dwight SS, Eppig JT, et al: Gene ontology: Tool for the unification of biology. The Gene Ontology Consortium. Nat Genet. 25:25–29. 2000. View Article : Google Scholar : PubMed/NCBI

23 

The Gene Ontology Consortium: The gene ontology resource: 20 years and still GOing strong. Nucleic Acids Res. 47:D330–D338. 2019. View Article : Google Scholar : PubMed/NCBI

24 

Kanehisa M, Furumichi M, Tanabe M, Sato Y and Morishima K: KEGG: New perspectives on genomes, pathways, diseases and drugs. Nucleic Acids Res. 45:D353–D361. 2017. View Article : Google Scholar : PubMed/NCBI

25 

Kanehisa M, Sato Y, Furumichi M, Morishima K and Tanabe M: New approach for understanding genome variations in KEGG. Nucleic Acids Res. 47:D590–D595. 2019. View Article : Google Scholar : PubMed/NCBI

26 

Huang da W, Sherman BT and Lempicki RA: Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc. 4:44–57. 2009. View Article : Google Scholar : PubMed/NCBI

27 

Rappaport N, Twik M, Plaschkes I, Nudel R, Iny Stein T, Levitt J, Gershoni M, Morrey CP, Safran M and Lancet D: MalaCards: An amalgamated human disease compendium with diverse clinical and genetic annotation and structured search. Nucleic Acids Res. 45(D1): D877–D887. 2017. View Article : Google Scholar : PubMed/NCBI

28 

Szklarczyk D, Franceschini A, Wyder S, Forslund K, Heller D, Huerta-Cepas J, Simonovic M, Roth A, Santos A, Tsafou KP, et al: STRING v10: Protein-protein interaction networks, integrated over the tree of life. Nucleic Acids Res. 43:D447–D452. 2015. View Article : Google Scholar : PubMed/NCBI

29 

Liu GM, Wong L and Chua HN: Complex discovery from weighted PPI networks. Bioinformatics. 25:1891–1897. 2009. View Article : Google Scholar : PubMed/NCBI

30 

Chin CH, Chen SH, Wu HH, Ho CW, Ko MT and Lin CY: cytoHubba: Identifying hub objects and sub-networks from complex interactome. BMC Syst Biol. 8 (Suppl 4):S112014. View Article : Google Scholar : PubMed/NCBI

31 

Wishart DS, Knox C, Guo AC, Cheng D, Shrivastava S, Tzur D, Gautam B and Hassanali M: DrugBank: A knowledgebase for drugs, drug actions and drug targets. Nucleic Acids Res. 36:D901–D906. 2008. View Article : Google Scholar : PubMed/NCBI

32 

Law V, Knox C, Djoumbou Y, Jewison T, Guo AC, Liu Y, Maciejewski A, Arndt D, Wilson M, Neveu V, et al: DrugBank 4.0: Shedding new light on drug metabolism. Nucleic Acids Res. 42:D1091–D1097. 2014. View Article : Google Scholar : PubMed/NCBI

33 

Ebrahimie E, Fruzangohar M, Moussavi Nik SH and Newman M: Gene ontology-based analysis of zebrafish omics data using the web tool comparative gene ontology. Zebrafish. 14:492–494. 2017. View Article : Google Scholar : PubMed/NCBI

34 

Fang L, Ellims AH, Beale AL, Taylor AJ, Murphy A and Dart AM: Systemic inflammation is associated with myocardial fibrosis, diastolic dysfunction, and cardiac hypertrophy in patients with hypertrophic cardiomyopathy. Am J Transl Res. 9:5063–5073. 2017.PubMed/NCBI

35 

Zen K, Irie H, Doue T, Takamiya M, Yamano T, Sawada T, Azuma A and Matsubara H: Analysis of circulating apoptosis mediators and proinflammatory cytokines in patients with idiopathic hypertrophic cardiomyopathy: Comparison between nonobstructive and dilated-phase hypertrophic cardiomyopathy. Int Heart J. 46:231–244. 2005. View Article : Google Scholar : PubMed/NCBI

36 

Mirtschink P and Krek W: Hypoxia-driven glycolytic and fructolytic metabolic programs: Pivotal to hypertrophic heart disease. Biochim Biophys Acta. 1863:1822–1828. 2016. View Article : Google Scholar : PubMed/NCBI

37 

Semenza GL: Signal transduction to hypoxia-inducible factor 1. Biochem Pharmacol. 64:993–998. 2002. View Article : Google Scholar : PubMed/NCBI

38 

Sherrid MV: Drug therapy for hypertrophic cardiomypathy: Physiology and practice. Curr Cardiol Rev. 12:52–65. 2016. View Article : Google Scholar : PubMed/NCBI

39 

Ammirati E, Contri R, Coppini R, Cecchi F, Frigerio M and Olivotto I: Pharmacological treatment of hypertrophic cardiomyopathy: Current practice and novel perspectives. Eur J Heart Fail. 18:1106–1118. 2016. View Article : Google Scholar : PubMed/NCBI

40 

Burns C, Bagnall RD, Lam L, Semsarian C and Ingles J: Multiple gene variants in hypertrophic cardiomyopathy in the era of next-generation sequencing. Circ Cardiovasc Genet. 10(pii): e0016662017.PubMed/NCBI

41 

Ntelios D, Meditskou S, Efthimiadis G, Pitsis A, Nikolakaki E, Girtovitis F, Parcharidou D, Zegkos T, Kouidou S, Karvounis H and Tzimagiorgis G: Elevated plasma levels of miR-29a are associated with hemolysis in patients with hypertrophic cardiomyopathy. Clin Chim Acta. 471:321–326. 2017. View Article : Google Scholar : PubMed/NCBI

42 

Salman OF, El-Rayess HM, Abi Khalil C, Nemer G and Refaat MM: Inherited cardiomyopathies and the role of mutations in non-coding regions of the genome. Front Cardiovasc Med. 5:772018. View Article : Google Scholar : PubMed/NCBI

43 

Fang L, Ellims A, Moore XL, White DA, Taylor AJ, Chin-Dusting J and Dart AM: Circulating microRNAs as biomarkers for diffuse myocardial fibrosis in patients with hypertrophic cardiomyopathy. J Transl Med. 13:3142015. View Article : Google Scholar : PubMed/NCBI

44 

Xue J, Zhou D, Poulsen O, Hartley I, Imamura T, Xie EX and Haddad GG: Exploring miRNA-mRNA regulatory network in cardiac pathology in Na+/H+ exchanger isoform 1 transgenic mice. Physiol Genomics. 50:846–861. 2018. View Article : Google Scholar : PubMed/NCBI

45 

Yamada S, Hsiao YW, Chang SL, Lin YJ, Lo LW, Chung FP, Chiang SJ, Hu YF, Tuan TC, Chao TF, et al: Circulating microRNAs in arrhythmogenic right ventricular cardiomyopathy with ventricular arrhythmia. Europace 20 (FI1). F37–F45. 2018. View Article : Google Scholar

46 

Lee JD, Lee TH, Huang YC, Chang YJ, Chang CH, Hsu HL, Lin YH, Wu CY, Lee M, Huang YC, et al: ALOX5AP genetic variants and risk of atherothrombotic stroke in the Taiwanese population. J Clin Neurosci. 18:1634–1638. 2011. View Article : Google Scholar : PubMed/NCBI

47 

Helgadottir A, Manolescu A, Thorleifsson G, Gretarsdottir S, Jonsdottir H, Thorsteinsdottir U, Samani NJ, Gudmundsson G, Grant SF, Thorgeirsson G, et al: The gene encoding 5-lipoxygenase activating protein confers risk of myocardial infarction and stroke. Nat Genet. 36:233–239. 2004. View Article : Google Scholar : PubMed/NCBI

48 

Lai HM, Li XM, Yang YN, Ma YT, Xu R, Pan S, Zhai H, Liu F, Chen BD and Zhao Q: Genetic variation in NFKB1 and NFKBIA and susceptibility to coronary artery disease in a Chinese Uygur population. PLoS One. 10:e01291442015. View Article : Google Scholar : PubMed/NCBI

49 

Han P, Li W, Lin CH, Yang J, Shang C, Nuernberg ST, Jin KK, Xu W, Lin CY, Lin CJ, et al: A long noncoding RNA protects the heart from pathological hypertrophy. Nature. 514:102–106. 2014. View Article : Google Scholar : PubMed/NCBI

50 

Yang W, Li Y, He F and Wu H: Microarray profiling of long non-coding RNA (lncRNA) associated with hypertrophic cardiomyopathy. BMC Cardiovasc Disord. 15:622015. View Article : Google Scholar : PubMed/NCBI

51 

Sherrid MV, Shetty A, Winson G, Kim B, Musat D, Alviar CL, Homel P, Balaram SK and Swistel DG: Treatment of obstructive hypertrophic cardiomyopathy symptoms and gradient resistant to first-line therapy with β-blockade or verapamil. Circ Heart Fail. 6:694–702. 2013. View Article : Google Scholar : PubMed/NCBI

52 

Östman-Smith I: Beta-Blockers in pediatric hypertrophic cardiomyopathies. Rev Recent Clin Trials. 9:82–85. 2014. View Article : Google Scholar : PubMed/NCBI

Related Articles

Journal Cover

July-2019
Volume 20 Issue 1

Print ISSN: 1791-2997
Online ISSN:1791-3004

Sign up for eToc alerts

Recommend to Library

Copy and paste a formatted citation
x
Spandidos Publications style
Li J, Wu Z, Zheng D, Sun Y, Wang S and Yan Y: Bioinformatics analysis of the regulatory lncRNA‑miRNA‑mRNA network and drug prediction in patients with hypertrophic cardiomyopathy. Mol Med Rep 20: 549-558, 2019
APA
Li, J., Wu, Z., Zheng, D., Sun, Y., Wang, S., & Yan, Y. (2019). Bioinformatics analysis of the regulatory lncRNA‑miRNA‑mRNA network and drug prediction in patients with hypertrophic cardiomyopathy. Molecular Medicine Reports, 20, 549-558. https://doi.org/10.3892/mmr.2019.10289
MLA
Li, J., Wu, Z., Zheng, D., Sun, Y., Wang, S., Yan, Y."Bioinformatics analysis of the regulatory lncRNA‑miRNA‑mRNA network and drug prediction in patients with hypertrophic cardiomyopathy". Molecular Medicine Reports 20.1 (2019): 549-558.
Chicago
Li, J., Wu, Z., Zheng, D., Sun, Y., Wang, S., Yan, Y."Bioinformatics analysis of the regulatory lncRNA‑miRNA‑mRNA network and drug prediction in patients with hypertrophic cardiomyopathy". Molecular Medicine Reports 20, no. 1 (2019): 549-558. https://doi.org/10.3892/mmr.2019.10289