SLC35E3 identified as a target of novel‑m1061‑5p via microRNA profiling of patients with cardiovascular disease
- Authors:
- Published online on: January 25, 2018 https://doi.org/10.3892/mmr.2018.8498
- Pages: 5159-5167
-
Copyright: © Gao et al. This is an open access article distributed under the terms of Creative Commons Attribution License.
Abstract
Introduction
Cardiovascular disease (CVD) is the largest cause of mortality worldwide that involves the heart or blood vessels, and is associated with high blood pressure, diabetes, obesity, high blood cholesterol, poor diet and excessive alcohol consumption (1–6). microRNAs (miRNAs) are the most abundant class of regulatory noncoding RNA (ncRNA) involved in cell differentiation, expansion and apoptosis, and other biological processes by regulating over half of all human protein-coding genes. The dysfunction of miRNA may cause abnormal gene expression, affecting human health.
Previously, miRNAs have been regarded as a potential therapeutic target for a variety of CVDs, including atherosclerosis, myocardial infarction and hypertrophy (1–4). Additionally, miRNAs have been considered as key regulators in vascular biology (5–9). miRNA-21 (miR-21), miR-146a, miR-155, miR-221, miR-222 and miR-34a are reportedly associated with angiogenesis in patients with CVD (10–12); however, some miRNAs have been associated with the regulation of low-density lipoprotein and high-density lipoprotein (HDL) in atherosclerosis and other CVDs (13,14).
However, the etiology of CVD is complex and variations in miRNA expression patterns have been observed in patients with CVD. For example, miR-22 targets monocyte chemoattractant protein-1 and contributes to the pathogenesis of coronary artery disease (15). Therefore, a comprehensive understanding of miRNAs and target genes associated with various types of CVD is required. In the present study, miRNA profiles of blood samples from patients with CVD were investigated to improve understanding of the underlying mechanism of miRNA in the pathogenesis of CVD, and may therefore contribute to the effective treatment of CVDs.
Patients and methods
Patients
The present study included 6 patients diagnosed in the Shandong Provincial Hospital (Jinan, China) between May and September 2014. A total of 3 patients with CVD were diagnosed as atherosclerotic, 3 healthy volunteers were included as the control (CK). All patients provided written informed consent. The present study was approved by the ethics committee of Shandong Provincial Hospital.
Sample preparation and sequencing
Blood samples were prepared for isolating the RNA. Total RNA of all six samples were isolated and purified using TRIzol reagent (Invitrogen; Thermo Fischer Scientific, Inc., Waltham, MA) according to the manufacturer's protocol. RNA quality was assessed using a BioAnalyzer 2100 kit (Agilent Technologies, Inc. Santa Clara, CA) and a RNA 6000 Nano kit (Agilent Technologies, Inc.). Subsequently, RNA libraries were prepared using the Small RNA Sample Prep kit (Illumina, Inc., San Diego, CA) according to the manufacturer's protocol. Deep sequencing was performed via the HiSeq™ 2000 system (Illumina, Inc.).
Sequencing analysis
Removal of adaptor sequences was conducted using Cutadapt v. 1.9 software [http://cutadapt.readthedocs.org/1.9] (16). Low quality reads of >95% base length with Phred quality scores <20 were filtered using the FASTX-Toolkit [http://hannonlab.cshl.edu/fastx_toolkit/] (17). Additionally, reads with polyA and polyT were also removed, and reads of <15 nucleotides or >34 nucleotides in length were discarded via miRDeep [https://www.mdc-berlin.de/8551903/en/] (18). Clean sequencing reads from small RNA (sRNA) libraries were summarized for length distribution and sRNA annotation. The sRNAs were mapped to the ncRNAs deposited in the NCBI GenBank database (https://www.ncbi.nlm.nih.gov/genbank/) and Rfam database (http://rfam.janelia.org/) using the BLAST algorithm (https://blast.ncbi.nlm.nih.gov/Blast.cgi). Sequences that matched ncRNAs constituted rRNAs, tRNAs, small nuclear RNAs (snRNAs), and small nucleolar RNAs, were annotated. Furthermore, the unique sRNA sequences were analysed via BLAST against miRBase v.20 (ftp://mirbase.org/pub/mirbase/CURRENT/). Sequences in the libraries were filtered with the standard of ≤1 mismatch and ≥15 matches to miRNA database were considered as mature miRNAs of a known miRNA family. The identified mature miRNA sequences were aligned against a human genomic sequence using Bowtie v. 2.2.4 [http://bowtie-bio.sourceforge.net/2.2.4] (19). The novel miRNA prediction pipeline was performed with Perl scripts combined using miREAP [http://mireap.sourceforge.net/0.2] (20).
Differential expression analysis
Alterations in the expression levels of mature and novel miRNAs within CVD and CK groups were investigated in present study. Expression levels of all miRNAs were normalized to the transcript expression level per million reads. The fold change of the miRNA expression levels between CVD and CK groups was employed to collate the differentially expressed miRNA. The average abundance of miRNA expression within the CVD and CK groups were calculated. Log2 (CVD/CK) values were calculated to present the fold change. The differential expression analysis was adjusted with a q-value adjusted P-value. Fold change [log2 (CVD/CK)>1)] and P<0.05 were combined to identify the differentially expressed miRNAs associated with disease. The visual differential expression patterns of the 65 miRNAs were collected from the heatmap program in R (21).
Target gene prediction and analysis
Target gene prediction of miRNAs was performed using miRanda 3.3a [http://www.microrna.org/microrna/3.3a] (22); differentially expressed miRNAs were mapped to the human transcriptome. Sequences matching perfectly were identified as the predicted target genes. Predictions with less than five mismatches and the cleavage site from the 10th to 11th nucleotides perfectly matched were admitted and scored. Targets with P≤0.05 were retained. Gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment of target genes regulated by differentially expressed miRNAs were performed to predict miRNA function.
Reverse transcription-quantitative polymerase chain reaction (RT-qPCR)
Total RNA was extracted using the TRIzol reagent method. The cDNA first strand was synthesized using the miRcute miRNA First-Strand cDNA Synthesis kit according to the manufacture's protocol (Tiangen Biochem Co., Ltd., Beijing, China). PolyA was added to 3′-end of miRNA and RT was performed according to the manufacturer's protocol of the kit (Tiangen Biochem Co., Ltd.). qPCR was performed with the miRcute miRNA qPCR Detection kit (containing SYBR Green; Tiangen Biochem Co., Ltd.) using the LightCycler 96 Real-Time PCR system (Roche Diagnostics, Basel, Switzerland). A 20 µl PCR reaction volume contained 1 µl cDNA, 10 µl 2X miRcute miRNA Premix (with SYBR and ROX), 0.4 µl forward primer, 0.4 µl reverse primer, 8.2 µl ddH2O. Primers were designed using DNAMAN version 6.0 software (Lynnon Biosoft, San Ramon, CA, USA). The primer sequences were: hsa-novel-m1061-5p, 5′-TCAGTTGTTCCATGTCCTGCAG-3′ and solute carrier family 35 member E3 (SLC35E3), forward 5′-ACGACAGGTGATCCACCTGC-3′, reverse: 5′-TATGAACCAACAAATACACC-3′. All primers were synthesized by Sangon Biotech Co., Ltd. (Shanghai, China). The following thermocycling conditions were used for RT-qPCR: 95°C 5 min for pre-denature; 95°C 15 sec, 60°C 45 sec, 72°C 15 sec, 40 cycles for amplification; default dissociation condition. U6 served as the internal reference. The primer sequences of U6 were forward 5′-CTCGCTTCGGCAGCACA-3′ and reverse 5′-AACGCTTCACGAATTTGCGT-3′. The cycle threshold values were obtained; the relative interest/reference expression quantity of genes was calculated with the formula 2−ΔCq(miR-U6). Relative test/normal CK quantity was calculated with the formula 2−ΔΔCq[(test miR-U6)-(CK miR-U6)] (10–12).
Results and Discussion
Deep sequencing of miRNA libraries and identification of conserved and novel miRNAs
In the present study, six sRNA profiles of three CK and three CVD individuals were sequenced. A total of >64.6 million clean reads were generated, in the range of 2.1–21.2 million for individual samples. Small sequences were filtered as described, which were annotated to the NCBI GenBank and Rfam databases. The sRNAs annotated to rRNA, tRNA and snRNA were removed; the remaining small sequences of the six sRNA profiles were then aligned to the miRBase v.20 database. A total of 4,771 conserved miRNAs were identified, including 2,764, 1,319, 2,730, 2,191, 2,528, and 2,050 unique conserved miRNAs from CK1, CK2, CK3, CVD1, CVD2, and CVD3 samples, respectively (Table I). In addition, 1,520 miRNAs were predicted as novel miRNAs in all samples, the number in each sample ranged from 374–1,059.
Transcript expression levels of the 6,291 identified miRNAs, including previously reported and novel miRNAs, in the 6 samples were calculated; 5,035 miRNAs were expressed within the CK group and 4,521 miRNAs within the CVD group. As presented in Fig. 1A, 1,419 commonly expressed miRNAs were identified, with a proportion of 28.2%, among the CK group samples. Conversely, 1,909 commonly expressed miRNA were identified with a proportion of 42.2% among the CVD group samples (Fig. 1B). The results of the present study indicated that CVD may be associated with a higher proportion of commonly expressed miRNAs in a variety of individuals.
Differentially expressed miRNAs
In order to identify the miRNAs associated with CVD, the fold change of CVD group vs. the CK group was determined. A set of 65 abnormal miRNAs, that were included in the commonly expressed 1,909 miRNAs demonstrated ≥2-fold change and P<0.05 in the CVD group compared with the CK group (Fig. 1C). A >2-fold upregulation of 59 miRNAs and downregulation of 6 miRNAs was detected (Table II; Fig. 2). In addition, differential expression levels of six novel miRNAs were identified, in which five were upregulated and one downregulated.
In the present study, the sRNA libraries were used to identify abnormally expressed miRNAs. Consequently, a total of 65 miRNAs were identified, among which miR-33 with the function of HDL synthesis and cholesterol transport, was upregulated in the CVD group. A previous study reported that miR-33a/b is embedded within the introns of human sterol regulatory element-binding protein (hSREBP)-1 and −2, which encodes the transcriptional regulator of cholesterol synthesis (23). miR-33a/b binds the mRNA of ATP-binding cassette A1 (ABCA1), a key transporter of intracellular cholesterol efflux. Upregulated expression levels of miR-33a reduces cholesterol efflux activity of apolipoprotein A1 and HDL, raising intracellular cholesterol levels (24). Upregulated miR-144 has also been demonstrated to mediate the expression levels of ABCA1; the 3′-untranslated region of ABCA1 has been reported to be conservatively targeted by miR-144, thus reducing ABCA1 expression levels and cholesterol efflux of HDL (25). miR-126 was upregulated in the current study. miR-126 has been reported to be an endothelium-enriched miRNA that regulates the response of endothelial cells to vascular endothelial growth factor, and modulates vascular integrity and angiogenesis (26). A previous study reported that miR-126 regulates the expression levels of Sprouty-related protein and phosphoinositol-3 kinase regulatory subunit 2, which are responsible for the inhibition of angiogenic signalling (27).
Functional annotation for target genes
The biological functions of CVD-associated miRNAs were investigated using the 2,784 putative genes targeted by 65 differentially expressed miRNAs. The results of the present study revealed that the expression of 2,401 target genes were repressed by 59 upregulated miRNAs; expression of 383 target genes were reduced by the downreguation of the other 6 miRNAs. GO classification analysis for the 2,784 differently expressed miRNA target genes was performed (Table III). The P-value was combined with Bonferroni correction for multiple testing; 49 GO biological processes were enriched (Table III), including ‘regulation of axonogenesis’, ‘cell-cell adhesion’, ‘intracellular signal transduction’, ‘cellular localization’, ‘regulation of signal transduction’, ‘cellular protein modification process’, ‘positive regulation of cellular process’, ‘cellular component organization’, ‘regulation of biological quality’ and ‘regulation of transcription’.
The target genes of downregulated miRNAs were enriched in 12 KEGG pathways. Most of these pathways are responsible for lipid and glycan metabolism (Table IV). In particular, three downregulated miRNAs, hsa-miR-1268b, hsa-miR-1273d, hsa-miR-3187-5p, were associated with a-linolenic acid metabolism. hsa-miR-4492 was predicted to target phospholipase A2 group (ENSG00000100078; ENSG00000158786; ENSG00000168907; ENSG00000184381; ENSG00000187980) and fatty acid desaturase 2 (ENSG00000134824) which are key enzymes in a-linolenic acid metabolism. The target gene of hsa-miR-3187 was predicted to be phospholipase A2 group IVE gene (ENSG00000188089), and hsa-miR-1273d was predicted to target acyl-CoA oxidase 1 (ENSG00000161533).
Additionally, the target genes of upregulated miRNAs were enriched in 15 KEGG pathways, mainly in the ‘human diseases’ class (Table V). These KEGG pathways did not match the CVDs directly, but were mainly involved in ‘neurodegenerative diseases and cancers’ class.
Novel miRNA annotations
In the present study, six novel miRNAs were upregulated, including novel-m0499-5p, novel-m0970-5p, novel-m1042-5p, novel-m1061-5p and novel-m1953-5p, and novel-m1627-5p was downregulated. Target gene prediction of novel miRNAs, m0499-5p, m0970-5p, m1042-5p and m1953-5p was unsuccessful, which indicated that the functions of these miRNAs remain unidentified. Novel-m1627-5p was predicted to target 146 human genes; however, further investigations into these target genes are required.
With the analysis using miRanda 3.3a (http://www.microrna.org/microrna/3.3a), novel-m1061-5p was predicted to target four genes, including ENSG00000115042 [fumarylacetoacetate hydrolase domain containing 2A (FAHD2A)], ENSG00000116396 [potassium voltage-gated channel subfamily C member 4 (KCNC4)], ENSG00000205476 [coiled-coil domain containing 85C (CCDC85C)] and ENSG00000175782 (SLC35E3; GO term, GO:1901264 ‘carbohydrate derivative transport’; GO class, ‘biological process’), and these genes were observed on the website: http://asia.ensembl.org/index.html. FAHD2A, KCNC4 and CCDC85C were not enriched in the KEGG pathways or GO terms of differential genes. Differential ENSG00000175782 (SLC35E3) had the GO term of carbohydrate derivative transport involving in biological process. SLC35E3 is a member of the nucleoside sugar transporter subfamily E (28,29). The nucleoside sugar transporters are localized at the Golgi complex and the endoplasmic reticulum (ER). SLC35E3 transports cytosolic nucleotide sugars into the lumen of Golgi complex and ER, where nucleotide sugars are substrates for the glycosylation of proteins, lipids and proteoglycans (28). Deficiency in nucleotide sugar transporters has been associated with tumour metastasis, cellular immunity, organogenesis and morphogenesis (29). For instance, congenital disorder of glycosylation type IIc (also termed leukocyte adhesion deficiency-2) is caused by defective guanosine 5′-diphosphate transport (29). In the present study, upregulation of novel-m1061-5p in patients with CVD may reduce SLC35E3 expression levels, resulting in defects in glycol-conjugation. In addition, novel-m1061-5p may serve a marker or potential target in the prognosis or treatment of CVD; however, the underlying mechanism of this miRNA requires further investigation.
RT-qPCR analysis was performed to confirm the expression levels of novel-m1061-5p and SLC35E3. Expression levels of novel-m1061-5p within the three patients with CVD were significantly increased to be 3.71207-fold, 3.26909-fold and 2.40420-fold greater than in the CK group, respectively. Expression levels of SLC35E3 were significantly decreased by 0.33-fold, 0.28-fold, 0.41-fold within patients with CVD, respectively, compared with in the CK group. The results of the present study revealed that upregulation of novel-m1061-5p expression levels was associated with the repression of SLC35E3 expression levels within the 3 patients with CVD.
The sequencing data of the present study revealed the miRNA profiles and associated target genes in patients with CVD; however, more patients for large-scale data collection and further investigation to confirm gene function are required. Molecular detection may contribute to the prognosis and treatment of CVDs.
Acknowledgements
The present study was supported by Shandong Provincial Natural Science Foundation of China (grant no. BS2014YY056).
References
Bernardo BC, Ooi JY, Lin RC and McMullen JR: miRNA therapeutics: A new class of drugs with potential therapeutic applications in the heart. Future Med Chem. 7:1771–1792. 2015. View Article : Google Scholar : PubMed/NCBI | |
Kwekkeboom RF, Lei Z, Doevendans PA, Musters RJ and Sluijter JP: Targeted delivery of miRNA therapeutics for cardiovascular diseases: Opportunities and challenges. Clin Sci (Lond). 127:351–365. 2014. View Article : Google Scholar : PubMed/NCBI | |
Zhang Y, Wang S, Li Y, Zhang C, Xue J, Wu X and Wang C: Relationship of microRNA 616 gene polymorphism with prognosis of patients with premature coronary artery disease. Int J Clin Pharmacol Ther. 54:899–903. 2016. View Article : Google Scholar : PubMed/NCBI | |
Li HY, Zhao X, Liu YZ, Meng Z, Wang D, Yang F and Shi QW: Plasma MicroRNA-126-5p is associated with the complexity and severity of coronary artery disease in patients with stable angina pectoris. Cell Physiol Biochem. 39:837–846. 2016. View Article : Google Scholar : PubMed/NCBI | |
Cordes KR and Srivastava D: MicroRNA regulation of cardiovascular development. Circ Res. 104:724–732. 2009. View Article : Google Scholar : PubMed/NCBI | |
Cengiz M, Yavuzer S, Kılıçkıran Avcı B, Yürüyen M, Yavuzer H, Dikici SA, Karataş ÖF, Özen M, Uzun H, Öngen Z, et al: Circulating miR-21 and eNOS in subclinical atherosclerosis in patients with hypertension. Clin Exp Hypertens. 37:643–649. 2015. View Article : Google Scholar : PubMed/NCBI | |
Wang J, Yan Y, Song D and Liu B: Reduced plasma miR-146a is a predictor of poor coronary collateral circulation in patients with coronary artery disease. Biomed Res Int. 2016:42859422016. View Article : Google Scholar : PubMed/NCBI | |
Wang J, Yan Y, Song D, Liu L and Liu B: The association of plasma miR-155 and VCAM-1 levels with coronary collateral circulation. Biomark Med. 11:125–131. 2017. View Article : Google Scholar : PubMed/NCBI | |
Wang M, Li W, Chang GQ, Ye CS, Ou JS, Li XX, Liu Y, Cheang TY, Huang XL and Wang SM: MicroRNA-21 regulates vascular smooth muscle cell function via targeting tropomyosin 1 in arteriosclerosis obliterans of lower extremities. Arterioscler Thromb Vasc Biol. 31:2044–2053. 2011. View Article : Google Scholar : PubMed/NCBI | |
Hans FP, Moser M, Bode C and Grundmann S: MicroRNA regulation of angiogenesis and arteriogenesis. Trends Cardiovasc Med. 20:253–262. 2010. View Article : Google Scholar : PubMed/NCBI | |
Rubanyi GM: Mechanistic, technical, and clinical perspectives in therapeutic stimulation of coronary collateral development by angiogenic growth factors. Mol Ther. 21:725–738. 2013. View Article : Google Scholar : PubMed/NCBI | |
Liao LX, Zhao MB, Dong X, Jiang Y, Zeng KW and Tu PF: TDB protects vascular endothelial cells against oxygen-glucose deprivation/reperfusion-induced injury by targeting miR-34a to increase Bcl-2 expression. Sci Rep. 6:379592016. View Article : Google Scholar : PubMed/NCBI | |
Huang YQ, Cai AP, Chen JY, Huang C, Li J and Feng YQ: The relationship of plasma miR-29a and oxidized low density lipoprotein with atherosclerosis. Cell Physiol Biochem. 40:1521–1528. 2016. View Article : Google Scholar : PubMed/NCBI | |
Michell DL and Vickers KC: HDL and microRNA therapeutics in cardiovascular disease. Pharmacol Ther. 168:43–52. 2016. View Article : Google Scholar : PubMed/NCBI | |
Chen B, Luo L, Zhu W, Wei X, Li S, Huang Y, Liu M and Lin X: miR-22 contributes to the pathogenesis of patients with coronary artery disease by targeting MCP-1: An observational study. Medicine (Baltimore). 95:e44182016. View Article : Google Scholar : PubMed/NCBI | |
Martin M: Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet Journal. 17:10–12. 2011. View Article : Google Scholar | |
Gordon A and Hannon GJ: Fastx-toolkit. FASTQ/A short-reads pre-processing tools. 2010.http://hannonlab.cshl.edu/fastx_toolkit | |
Friedländer MR, Chen W, Adamidi C, Maaskola J, Einspanier R, Knespel S and Rajewsky N: Discovering microRNAs from deep sequencing data using miRDeep. Nat Biotechnol. 26:407–415. 2008. View Article : Google Scholar : PubMed/NCBI | |
Langmead B, Trapnell C, Pop M and Salzberg SL: Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol. 10:R252009. View Article : Google Scholar : PubMed/NCBI | |
Li QB and Wang J: MIREAP: microRNA discovery by deep sequencing. 2008.https://sourceforge.net/projects/mireap/ | |
Kolde R: Pheatmap: pretty heatmaps. R package version 061. 2012.https://www.r-project.org/ | |
Betel D, Wilson M, Gabow A, Marks DS and Sander C: The microRNA.org resource: targets and expression. Nucleic Acids Res. 36:D149–153. 2008. View Article : Google Scholar : PubMed/NCBI | |
Najafi-Shoushtari SH, Kristo F, Li Y, Shioda T, Cohen DE, Gerszten RE and Näär AM: MicroRNA-33 and the SREBP host genes cooperate to control cholesterol homeostasis. Science. 328:1566–1569. 2010. View Article : Google Scholar : PubMed/NCBI | |
Rayner KJ, Suárez Y, Dávalos A, Parathath S, Fitzgerald ML, Tamehiro N, Fisher EA, Moore KJ and Fernández-Hernando C: MiR-33 contributes to the regulation of cholesterol homeostasis. Science. 328:1570–1573. 2010. View Article : Google Scholar : PubMed/NCBI | |
Ramírez CM, Rotllan N, Vlassov AV, Dávalos A, Li M, Goedeke L, Aranda JF, Cirera-Salinas D, Araldi E, Salerno A, et al: Control of cholesterol metabolism and plasma high-density lipoprotein levels by microRNA-144. Circ Res. 112:1592–1601. 2013. View Article : Google Scholar : PubMed/NCBI | |
Fish JE, Santoro MM, Morton SU, Yu S, Yeh RF, Wythe JD, Ivey KN, Bruneau BG, Stainier DY and Srivastava D: miR-126 regulates angiogenic signaling and vascular integrity. Dev Cell. 15:272–284. 2008. View Article : Google Scholar : PubMed/NCBI | |
Wang S, Aurora AB, Johnson BA, Qi X, McAnally J, Hill JA, Richardson JA, Bassel-Duby R and Olson EN: The endothelial-specific microRNA miR-126 governs vascular integrity and angiogenesis. Dev Cell. 15:261–271. 2008. View Article : Google Scholar : PubMed/NCBI | |
Ishida N and Kawakita M: Molecular physiology and pathology of the nucleotide sugar transporter family (SLC35). Pflugers Arch. 447:768–775. 2004. View Article : Google Scholar : PubMed/NCBI | |
Song Z: Roles of the nucleotide sugar transporters (SLC35 family) in health and disease. Mol Aspects Med. 34:590–600. 2013. View Article : Google Scholar : PubMed/NCBI |