Expression Profiles of Six Atherosclerosis-Associated microRNAs That Cluster in Patients with Hyperhomocysteinemia: A Clinical Study

Abstract

The aim of this study is to discuss the hypothesis that expression of plasma atherosclerosis-associated microRNAs (miRNAs) in hyperhomocysteinemia (Hhcy) patients could predict the presence of atherosclerosis from different channels. Six plasma miRNAs (miR-145, miR-155, miR-222, miR-133, miR-217, and miR-30) selected for our study have been confirmed as critical gene regulators involved in atherosclerosis and can be steadily determined in plasma. Expression of the above six plasma circulating miRNAs revealed significant upregulation of two miRNAs (miR-133 and miR-217) and downregulation of three miRNAs (miR-145, miR-155, and miR-222). Six candidate miRNAs showed a significant correlation with homocysteine (Hcy) or lipid parameters. The results of this study indicated that miR-217 was further significantly upregulated in Hhcy + ATH groups than in normal control, Hhcy-, and atherosclerosis-alone (ATH) groups and it showed a significant negative correlation with Hcy and triglycerides. More specifically, miR-217 showed the most specific expression patterns in all patients with atherosclerosis (ATH and Hhcy + ATH groups), which may have been a diagnostic value for Hhcy complicated with atherosclerosis, and predicted the progress of atherosclerosis in Hhcy patients effectively.

Introduction

Atherosclerosis is the leading cause of morbidity and mortality worldwide (Libby et al., 2011). The pathogenesis of atherosclerosis is complicated by involvement of interactions of multiple genetic and environmental factors. Atherogenesis refers to the development of atheromatous plaques in the inner lining of the arteries, which is characterized by arterial endothelial dysfunction, vascular inflammation, the buildup of lipids, cholesterol, cellular debris, and calcium, within the intima of the vessel wall (Kovacic and Bakran, 2012; Gao et al., 2016). These interaction processes, in the context of chronic inflammation, are responsible for plaque progression leading to plaque rupture and the clinical complications, such as myocardial infarction, ischemic gangrene, and stroke (Hosin et al., 2014).

The development of atherosclerosis is known to have determined other metabolic risk factors, such as hypertension, diabetes mellitus, and dyslipidemia. Hyperhomocysteinemia (Hhcy) has been regarded as an emerging risk factor for the development of atherosclerosis (Tribouilloy et al., 2000; Schaffer et al., 2014). Hhcy increases the risk of carotid artery plaque. The potential mechanism is related with homocysteine initiating an inflammatory response in vascular smooth muscle cells (VSMCs), migration of VSMCs, and triggering proliferation (Jeon et al., 2014; Meng et al., 2016). Although the role of Hhcy in atherosclerosis is documented, the regulatory mechanism is still nebulous. Recent studies have suggested that microRNAs (miRNAs) work as a key factor in atherosclerosis development and dyslipidemia (Urbich et al., 2008; Zhang et al., 2015; Baldan and Fernandez-Hernando, 2016; Gao et al., 2016). However, the exact correlation on the expression of atherosclerosis-associated miRNAs in Hhcy patients remains to be elucidated further.

miRNAs are ∼22 nucleotide-long noncoding RNAs and post-transcriptionally inhibit gene expression usually by binding the three prime untranslated regions (3′-UTR) of target messenger RNAs (mRNAs) in mammals (Deddens et al., 2013; Maegdefessel, 2014; Gao et al., 2016). miRNAs have proven major roles in almost all biological processes by influencing the stability and translation of their target mRNAs. In previous years, emerging evidences have demonstrated that circulating miRNAs could be considered novel and potential biomarkers for the diagnosis and prognosis of diseases (Lee et al., 2017; Liu et al., 2017; Matsuzaki et al., 2017). Therefore, confirming the expression patterns of atherosclerosis-associated plasma miRNAs in patients with Hhcy may help us reanalyze the association between atherosclerosis and Hhcy from different perspectives.

We aimed to address the hypothesis that expression of plasma atherosclerosis-associated miRNAs in Hhcy patients could predict the presence of atherosclerosis from different channels. Six plasma miRNAs confirmed as critical gene regulators involved in atherosclerosis and can be stably determined in plasma were selected for our study. miR-217 promotes endothelial senescence and it has been shown to express in human carotid plaques (Duan et al., 2016). miR-222 is an antiangiogenic gene regulating miRNAs involved in VSMC growth (Dentelli et al., 2010). miR-133 inhibited the proliferation of endometrial epithelial cells (Gao et al., 2016). miR-30 induces cell senescence (Huang et al., 2016). miR-145 promotes differentiation and prevents the proliferation of VSMCs (Zhao et al., 2015). miR-155 facilitated the inflammatory response of macrophages (Du et al., 2014; Tian et al., 2014).

Materials and Methods

Informed consent

The study followed the protocol for sample collection, which was approved by the Ethics Committee of First Affiliated Hospital of Shihezi University Medical College. All requirements were met according to the Declaration of Helsinki. Patients were fully informed of the study procedures before they provided their consent.

Research subjects

For this study, hyperhomocysteinemia (Hhcy) is taken as plasma homocysteine (Hcy) level above 15 μm/L (Kesherwani et al., 2015). Carotid artery atherosclerosis is characterized by carotid artery ultrasonography (USG) for plaque formation to categorize the subjects into plaque or no plaque groups. About 310 newly diagnosed Hhcy-alone subjects were screened in the Department of Cardiovascular Diseases in First Affiliated Hospital of Shihezi University Medical College from January 2012 to December 2015. Patients with history of hypertension, ischemic heart disease, diabetes mellitus, chronic liver diseases, chronic renal diseases, and those patients with excessive drinking and smoking were excluded. Finally, only 167 Hhcy patients were involved in the study. Out of these, 30 Hhcy and carotid artery atherosclerosis combined (Hhcy + ATH) subjects and 25 Hhcy stand-alone patients were picked. Both groups were matched for age, sex, body mass index (BMI), and blood pressure (BP). In addition, 25 carotid artery atherosclerosis-alone (ATH) subjects and 25 normal control (NC) subjects were picked and matched to the Hhcy + ATH group for age, sex, BMI, and BP. These subjects were drawn from 860 individuals who were already receiving health checkups at a physical examination center of the First Affiliated Hospital of Shihezi University Medical College during the same period.

Carotid artery USG for plaque formation

All patients underwent carotid artery USG to identify plaque formation. Individual patients were asked to lie down in a semidark room in a supine position, with their neck slightly extended and turned from the imaging sensor. Both the right and left carotid arteries and the bifurcation were visualized by an experienced radiologist via an USG device (Hitachi) using a 9 MHz linear ultrasound probe. According to study criteria, the absence of atherosclerotic plaque was considered “normal,” whereas positive results are indicative of atherosclerotic plaque (Arslan et al., 2016).

Biochemical assays

Whole blood samples were collected from each patient in ethylene diamine tetraacetic acid (EDTA) tubes, by fasting within 8 h, in the morning. Plasma Hcy, triglycerides (TG), total cholesterol (TC), high-density lipoprotein (HDL), low-density lipoprotein (LDL), apolipoprotein A, apolipoprotein B, apolipoprotein A/B, fasting blood glucose, alanine aminotransferase, aspartate aminotransferase, uric acid, and serum creatinine (Cre) were measured by using Hitachi 7600 automated biochemistry analyzer in the First Affiliated Hospital of Shihezi University Medical College.

In conclusion, whole blood samples were collected in tubes containing EDTA, by fasting within 8 h, in the morning. These samples were immediately centrifuged at 3000 rpm for 10 min at room temperature. After separation phase, plasma was collected and divided into two aliquots and it was frozen at −80°C later for RNA isolation.

RNA isolation and quantitative real-time PCR for miRNA expression

The miRNAs were extracted from a serum volume of 400 μL by using the miRcute miRNA Isolation Kit (Tiangen, Beijing, China) following the manufacturer's instructions and the guide from previous methods (Lv et al., 2015). Briefly, samples were supplemented with 3.5 μL miRNeasy Serum/Plasma Spike-In Control (1.6 × 10⁸ copies/μL working solution) (Qiagen, Germany). We demonstrated that the cel-miR-54 could be used for normalization of the RNA preparation (Baraniskin et al., 2013; Kuhlmann et al., 2014). Spike-in normalization with synthetic Caenorhabditis elegans-derived cel-miR-54 sequence was performed to allow relative comparison across the analyzed plasma samples (Mitchell et al., 2008). The cDNA was generated using the miRcute miRNA First-Strand cDNA Synthesis Kit (Tiangen), which was performed based on the protocols. Real-time PCR of miR-217, miR-222, miR-30, miR-133, miR-145, and miR-155 was performed to utilize the miRcute miRNA qPCR Detection Kit (SYBR Green) (Tiangen). The reaction volume was 20 μL along with a 3.0 μL cDNA used as a template. PCR was performed under the following conditions: initial denaturation at 94°C for 2 min followed by 45 cycles of denaturation at 95°C for 5 s and annealing and extension at 65°C for 40 s. The comparative Ct method (ΔCt) was exploited to calculate the relative expression level of miR. Relative quantification of miRNA expression was calculated by using the 2^−ΔΔCt method (Wang et al., 2010).

Statistical analysis

SPSS software for Windows version 20.0 (SPSS, Inc., Chicago) and GraphPad Prism 5.0 software (GraphPad Software, Inc., La Jolla, CA) were used in statistical analysis. For baseline characteristics of patients and controls, continuous variables were summarized as mean ± standard, and discrete variables as counts and proportions. For the normally distributed data, one-way analysis of variance (ANOVA) and multiple comparison least significant difference (LSD) tests were applied. The Kruskal–Wallis test was carried out to compare data that were not normally distributed. The significance of miRNA level differences between the carotid artery plaque group and no-plaque group was assessed by the independent sample Student's t-test. The χ ² test was used to compare gender distributions. Pearson's correlations were used to examine relationships between miRNAs and Hcy, lipid parameters, as well as carotid atherosclerotic plaque value. HemI Heatmap illustrator version 1.0 (http://hemi.biocuckoo.org/) was used to perform an agglomerative hierarchical cluster analysis based on the miRNA expression patterns as a panel. Receiver operating characteristic (ROC) curve analysis was undertaken to evaluate the capacity of six candidate miRNAs for detection of all Hhcy patients or all atherosclerosis patients. p < 0.05 was considered as a statistically significant difference.

Results

Significant differences of clinical features among four groups

The basic characteristics and clinical features of the studied groups are provided in Table 1. Serum Hcy, TG, TC, and LDL-cholesterol (LDL-c) show differences among the groups (p < 0.05). No significant differences were found in other clinical factors, including gender distribution, BMI, age, and HDL-cholesterol. As demonstrated in Figure 1, serum TC level in the Hhcy + ATH group was significantly higher than that in the Hhcy, ATH, and NC groups. In addition, serum TG level in the Hhcy group was higher than in the Hhcy + ATH and NC groups. Serum LDL-c levels in Hhcy, Hhcy + ATH, and ATH groups were all significantly higher than the NC group, while no differences in serum TG and LDL-c levels were found between the Hhcy + ATH and ATH subjects.

FIG. 1.

Multiple comparisons of baseline characteristics (Hcy [A], TC [B], TG [C], and LDL-c [D]) among NC, Hhcy, Hhcy + ATH, and ATH subjects. Data are shown as mean ± SEM. p Values were generated by one-way ANOVA test followed by the LSD post hoc multiple comparisons test. p < 0.05 or p < 0.001 was considered significant. ANOVA, analysis of variance; ATH, atherosclerosis; Hcy, homocysteine; Hhcy, hyperhomocysteinemia; LDL-c, low-density lipoprotein-cholesterol; LSD, least significant difference; NC, normal control; SEM, standard error of the mean; TC, total cholesterol; TG, triglycerides. Color images available online at www.liebertpub.com/dna

Table 1.

Significant Differences in Clinical Features Among the Four Groups

	NC	Hhcy	Hhcy + ATH	ATH	p
Males/females	13/12	14/11	20/10	15/10	0.720
Age (years)	48.32 ± 4.81	47.16 ± 5.50	49.00 ± 5.47	46.92 ± 5.20	0.423
BMI (kg/m²)	23.89 ± 2.62	23.47 ± 2.30	25.06 ± 3.56	23.13 ± 2.68	0.073
TC (mmol/L)	4.04 ± 0.78	4.60 ± 0.99	5.56 ± 1.26	4.34 ± 1.04	<0.001^**
TG (mmol/L)	1.12 ± 0.07	1.37 ± 0.04	1.45 ± 0.09	1.41 ± 0.11	0.004^**
HDL-c (mmol/L)	1.46 ± 0.48	1.58 ± 0.63	1.59 ± 0.57	1.82 ± 0.66	0.188
LDL-c (mmol/L)	2.64 ± 0.84	3.46 ± 0.88	3.36 ± 0.93	3.18 ± 0.81	0.006^*
ApoA (g/L)	1.40 ± 0.27	1.39 ± 0.27	1.39 ± 0.41	1.47 ± 0.31	0.775
ApoA/B	1.68 ± 0.64	1.57 ± 0.50	1.42 ± 0.45	1.53 ± 0.43	0.314
ApoB (g/L)	0.93 ± 0.27	0.93 ± 0.19	1.01 ± 0.28	1.01 ± 0.31	0.458
FBG (mmol/L)	5.06 ± 0.71	4.93 ± 0.55	5.07 ± 0.50	4.93 ± 0.59	0.702
ALT (U/L)	21.80 ± 12.05	19.8 ± 10.17	27.00 ± 12.80	20.48 ± 10.86	0.087
AST (U/L)	20.64 ± 9.84	20.84 ± 10.02	20.57 ± 12.36	21.20 ± 10.73	0.997
UA (μmol/L)	271.88 ± 49.30	278.64 ± 61.74	282.40 ± 66.96	265.40 ± 77.28	0.782
Cre (μmol/L)	71.30 ± 20.81	75.72 ± 24.12	71.30 ± 16.16	69.88 ± 19.84	0.757
Hcy (μmol/L)	8.60 ± 2.60	18.31 ± 4.85	21.76 ± 3.48	11.40 ± 2.62	<0.001^**

Continuous and categorical variable data are expressed as mean ± SEM. The statistical p-value was generated by the one-way ANOVA test or Kruskal–Wallis test. χ ² test was used to compare gender distribution. ^* p < 0.05 or ^** p < 0.001 was considered significant.

ALT, alanine aminotransferase; ANOVA, analysis of variance; ApoA, apolipoprotein A; ApoB, apolipoprotein B; Apo(A/B), apolipoprotein A/B; AST, aspartate aminotransferase; ATH, atherosclerosis; BMI, body mass index; Cre, creatinine; FBG, fasting blood glucose; Hcy, homocysteine; Hhcy, hyperhomocysteinemia; HDL-c, high-density lipoprotein-cholesterol; LDL-c, low-density lipoprotein-cholesterol; NC, normal control; SEM, standard error of the mean; TC, total cholesterol; TG, triglycerides; UA, uric acid.

All six miRNAs can be stably detected in plasma samples from all subjects

Six miRNAs were stably detected in serum samples of all subjects. No significant differences were discovered in the total RNA concentration among the four groups (Fig 2A: NC 11.10 ± 10 ng/μL, Hhcy 10.78 ± 1.41 ng/μL, Hhcy + ATH 11.18 ± 1.41 ng/μL, and ATH 10.88 ± 1.88 ng/μL). The primers for six candidate miRNAs and cel-miR-54 are listed in Supplementary Table S1 (Supplementary Data are available online at www.liebertpub.com/dna).

FIG. 2.

(A) Concentration of total RNA samples has no significant difference among four groups. (B) The six miRNAs and cel-miR-54 served as the spike-in control, all showed reliable Ct values in all subjects. (C) The relative expression levels of six candidate plasma miRNAs (shown in log₂ scale) in the NC, Hhcy, Hhcy + ATH, and ATH groups. miR-54 is a synthetic Caenorhabditis elegans miR-54 miRNA mimic (cel-miR-54) that served as the spike-in control to monitor the effectiveness of RT-PCR and qPCR. The horizontal lines indicate the mean. p-Values were generated by one-way ANOVA test followed by the LSD post hoc multiple comparisons test. p < 0.05 or p < 0.001 was considered significant. Ct, threshold cycle; RT-PCR, real-time polymerase chain reaction; qPCR, quantitative polymerase chain reaction. Color images available online at www.liebertpub.com/dna

In the quantitative real-time PCR (qRT-PCR) analysis (Fig. 2B), all six miRNAs and cel-miR-54 showed reliable Ct values in all subjects. The spike-in control miRNA mimic (cel-miR-54) demonstrated both stable and moderate abundance in all subjects and no differences among the four groups.

Five of six candidate miRNAs have significantly different expression levels among the four groups

As shown in Figure 2C, the relative expression of miR-145 showed significant downregulation in NC (4.13 ± 0.85), Hhcy (3.07 ± 0.77), Hhcy + ATH (0.96 ± 0.98), and ATH (1.23 ± 0.38) groups, respectively, especially in the Hhcy + ATH group. No significant difference was observed between the Hhcy + ATH and ATH subjects. miR-155 showed a significant gradual decrease trend from NC (4.55 ± 0.67) to Hhcy (4.05 ± 0.43) and then to Hhcy + ATH (2.93 ± 0.54) and ATH (1.73 ± 0.51). miR-222 showed a significant stepwise decrease trend from NC (1.11 ± 0.45) to Hhcy (0.99 ± 0.52) and then to Hhcy + ATH (0.57 ± 0.42) and ATH (0.53 ± 0.45), no significant difference was found between the Hhcy + ATH and ATH subjects, as well as that between NC and Hhcy groups. Both miR-133 and miR-217 showed a significant stepwise increase trend from NC (0.68 ± 0.39 and 1.43 ± 0.63, respectively) to Hhcy (0.98 ± 0.46 and 2.29 ± 0.57, respectively) and then to Hhcy + ATH (1.84 ± 0.56 and 2.72 ± 0.70, respectively) or ATH (1.37 ± 0.64 and 2.51 ± 0.99, respectively) group. Finally, miR-30 showed no significant difference in all subjects.

Correlations between miRNAs and Hcy as well as lipid parameters

Detailed data on correlations between six candidate miRNAs and Hcy as well as lipid parameters of all subjects are presented in Table 2. As showed in Supplementary Figure S1, Pearson's correlation analysis demonstrated that Hcy was positively correlated with TC and LDL-c, respectively. miR-145 expression level showed a significant negative correlation with Hcy, TC, and LDL-c (Supplementary Fig. S2). miR-155 expression level also showed a significant negative correlation with Hcy (Supplementary Fig. S3), whereas no associations were found between miR-155 and lipid parameters. Similarly, expression level of miR-222 displayed a significant negative correlation with TC only (Supplementary Fig. S4). Upregulated miRNAs (miR-133 and miR-217) were positively correlated with Hcy and blood lipids (TC or LDL-c) (Supplementary Figs. S5 and S6).

Table 2.

Correlations Between microRNAs and Hcy and Lipid Profiles in All Groups

	Hcy	TC	LDL-c	TG
miR-145
r	−0.445^*	−0.400^*	−0.212^*	−0.144
p	<0.001	<0.001	0.030	0.144
miR-155
r	−0.228^*	−0.176	−0.139	−0.053
p	0.019	0.072	0.156	0.591
miR-222
r	−0.083	−0.237^*	−0.115	−0.101
p	0.402	0.015	0.243	0.304
miR-133
r	0.418^*	0.197^*	0.221^*	0.032
p	<0.001	0.044	0.024	0.747
miR-217
r	0.433^*	0.291^*	0.163	−0.128
p	<0.001	0.003	0.096	0.195
miR-30
r	−0.133	0.020	0.069	−0.183
p	0.175	0.836	0.483	0.061
Hcy
r		0.374^*	0.263^*	0.025
p		<0.001	0.007	0.796
TC
r			0.339^*	0.078
p			<0.001	0.429
LDL-c
r				0.057
p				0.560

Pearson's correlations were presented as correlation coefficients (r) and significance (p), p < 0.05 or p < 0.001 was considered significant.

Hierarchical clustering based on the five differentially expressed plasma miRNAs among four groups could separate a majority of subjects of each group

Hierarchical clustering analysis can find linear combinations of miRNAs, that is, maximizing the probability of correctly assigning subjects (Sturn et al., 2002). The five miRNA expression levels are different among the four groups included to generate a heatmap diagram shown in Figure 3. Hierarchical clustering analysis based on five differentially expressed plasma miRNAs could separate a majority of subjects from each group. NC subjects (n = 25) were clustered together with only four Hhcy patients mixed in it. Twenty-one of 25 (21/25) Hhcy-alone subjects were clustered together without any other subjects mixed in it. All of Hhcy + ATH (n = 30) subjects with 10 ATH subjects mixed in it. Fifteen of 25 (15/25) ATH subjects were clustered together without any other subjects mixed in it. With regard to variable clustering, miR-145 was clustered with miR-155, and miR-133 was clustered with miR-217, respectively (Fig. 3).

FIG. 3.

A majority of subjects of each group clustered together in the hierarchical clustering analysis based on five differentially expressed plasma miRNAs. Each column represents an individual plasma miRNA, and each row represents an individual subject. Red, black, and green colors indicate upregulation, unchanged expression, and downregulation compared to spike-in control, respectively. The color bar displays different colors that correspond to the relative expression levels of miRNAs. Color images available online at www.liebertpub.com/dna

ROC curves of six candidate plasma miRNAs for detection of all Hhcy patients or all ATH patients

For all of those subjects who were divided into Hhcy patients (Hhcy + ATH and Hhcy) and ATH patients (Hhcy + ATH and ATH), ROC curves were constructed to estimate the sensitivity and specificity of six candidate plasma miRNAs for the detection of all Hhcy and ATH patients. The areas under the curve (AUC) of the six miRNAs are displayed in Figure 11. miR-217 has the largest AUC among the six candidate miRNAs for the detection of entire Hhcy patients (AUC = 0.731, 95% confidence interval [CI], 0.634–0.827, p < 0.001) (Fig. 4A). While miR-217 has the largest AUC in all ATH patients (AUC = 0.851, 95% CI, 0.781–0.922, p < 0.001) (Fig. 4B) from all of the subjects.

FIG. 4.

(A) ROC curves for the ability of the six candidate plasma miRNAs were plotted to differentiate all the Hhcy patients (Hhcy + ATH and Hhcy) from the NC and ATH cases. (B) ROC curves for the ability of the six candidate plasma miRNAs to differentiate all the atherosclerosis patients (Hhcy + ATH and ATH) from the NC and Hhcy cases. The AUC was then estimated with 95% confidence interval. AUC, area under the curve; ROC, receiver operating characteristic. Color images available online at www.liebertpub.com/dna

Discussion

In this study, we investigated the association of six candidate plasma circulating miRNAs with Hhcy and atherosclerosis patients. We characterize the circulating miRNA profile in patients having Hhcy complicated with ATH and identify circulating miRNAs that can best reflect the level of blood lipids in Hhcy patients at the risk of atherosclerosis and further tested whether those miRNAs assist in detecting the presence of atherosclerosis in Hhcy patients. Expression of six specific plasma circulating miRNAs revealed significant upregulation of two miRNAs (miR-217 and miR-133) and downregulation of three miRNAs (miR-145, miR-155, and miR-222). All of the studied plasma circulating miRNAs can be stably detected in plasma samples from subjects under study using qRT-PCR. Importantly, we were especially interested in whether the patterns of all examined miRNAs treated together as a panel could provide more information than individual miRNAs. Hierarchical clustering analysis based on five differentially expressed miRNAs could separate a majority of subjects of each group indicating a better recognition of subjects through miRNAs treated as a panel. The results pointed out that the five candidate plasma circulating miRNAs treated as a panel featured specific expression pattern and could correctly assign the subjects of each group.

This study can be considered as the only one that is able to show the link between atherosclerosis-associated miRNAs and Hhcy in humans. In this study, all the subjects were newly diagnosed as hyperhomocysteinemia patients and had never undergone any therapeutic interventions. Our data showed that patients in Hhcy, Hhcy + ATH, and ATH groups differed from NC individuals in the mean values of TG, TC, and LDL-c. Moreover, TC and LDL-c were positively correlated with Hhcy, indicating that Hcy may contribute to an increase in the circulating lipid levels. This result is consistent with the results of other studies that have assessed these associations (Liao, 2006; Chernyavskiy et al., 2016). Circulating LDL-c levels are a major risk factor for atherosclerosis, and cholesterol metabolism is currently an important pharmacologic target to promote prevention of atherosclerotic cardiovascular disease (Giral et al., 2016).

Our qPCR analysis identified that five of the six atherosclerosis-specific miRNAs had significantly different expressions among the four groups, and their levels were correlated with Hcy or blood lipids. The results suggest that candidate plasma miRNAs may be involved in the process of atherosclerosis.

Initially, miR-145 was significantly downregulated in the Hhcy + ATH group than in NC, Hhcy, and ATH groups. The level of miR-145 expression showed a significantly negative correlation with Hcy, TC, and LDL-c. miR-145 was downregulated in injured or atherosclerotic vessels containing proliferating and less differentiated smooth muscle cells (Navickas et al., 2016). Climent et al. (2015) found that miR-145 acts as communication molecules between endothelial cells and smooth muscle to enhance the angiogenic and vessel stabilization properties of endothelial cells. Furthermore, miR-145 represses Kruppel-like factor-4 (Klf4), which proved to be involved in multiple events, including neointimal proliferation. Klf4 plays a key role in regulating the VSMC phenotype because it facilitates VSMC migration and downregulates VSMC differentiation marker genes (Garvey et al., 2010). These findings imply that miR-145 can stabilize plaques through a communication between vascular smooth muscle and endothelial cells, regarded as crucial processes in the pathogenesis of cardiovascular disease. Another study indicated that miR-145 is regarded as a potential biomarker of acute coronary syndrome with higher sensitivity for ST-segment elevation myocardial infarction and worse outcome for acute myocardial infarction (Dong et al., 2015).

Furthermore, miR-155 was significantly downregulated in the ATH group than in NC, Hhcy, and Hhcy + ATH groups. The level of miR-155 expression showed a significantly negative correlation with Hcy. A single study reported that miR-155 was inversely associated with complicated proatherogenic metabolic risk factors (Zhu et al., 2014). Several researchers have revealed that miR-155 is positively related to proinflammatory cytokine expression in macrophages during atherosclerosis development. In addition, Nazari-Jahantigh et al. (2012) showed that miR-155 promoted atherosclerosis in a partial carotid ligation ApoE^−/− mice by repressing B cell lymphoma 6 protein in macrophages, thus enhancing vascular inflammation. However, further studies used … mice on a high-fat Western diet to suggest dual behavior of miR-155. The above study showed an atheroprotective effect in earlier stages, which is attributed to inhibition of macrophage proliferation. Also, the study plays an opposite atherogenic role in later stages, which is described in previous studies (Wei et al., 2015). Our study has underlined an anti-inflammatory role of miR-155 in accordance with the above studies. Previous study has suggested that increased miR-155 expression is mediated by an anti-inflammatory effect presumably through targeting the calcium-regulated heat stable protein (CARHSP1), which regulates the stability of TNF-amRNA (Li et al., 2016). It suggested that miR-155 acted in an antiatherogenic role, which is much similar to the fact that was shown in our data.

Moreover, miR-222 was significantly downregulated in atherosclerosis groups (AHT + Hhcy and ATH) than in nonatherosclerosis groups (NC and Hhcy). Also, the level of miR-222 expression showed a significantly negative correlation with TG. Zhu et al. (2011) revealed that miR-222 was highly expressed in human umbilical vein endothelial cells. miR-222 regulated Ang II-induced endothelial inflammation and migration and exerted antiangiogenic effects. It has been reported that miR-222 inhibits angiogenesis via targeting c-Kit, signal transducer, activator of transcription (STAT) 5A, and zinc finger E-box-binding homeobox (ZEB) 2 (Poliseno et al., 2006; Dentelli et al., 2010). It promoted VSMC proliferation and neointimal hyperplasia by targeting p57(Kip2) (Liu et al., 2009), and indirectly reduced endothelial nitric oxide synthase expression. Thus, miR-222 may regulate inflammation, neointimal formation, and neovascularization of the plaque, leading to coronary artery disease right from the initiation of atherosclerosis to acute coronary syndrome.

Next, miR-133 was significantly upregulated in Hhcy + ATH groups than in NC, Hhcy, and ATH groups. The level of miR-133 expression showed a significantly negative correlation with Hcy, TG, and LDL-c. miR-133 suppressed angiogenesis properties of endothelial cells such as proliferation rate, cell viability, and migration activity by targeting vascular endothelial growth factor receptor 2 and fibroblast growth factor receptor 1 (Soufi-Zomorrod et al., 2016). Similar to our results, miR-133 is considered as protective and antifibrotic in smooth muscle and is elevated in patients with coronary atherosclerosis (Torella et al., 2011; Hulsmans and Holvoet, 2013).

Finally, miR-217 was significantly upregulated in Hhcy + ATH groups than in NC, Hhcy, and ATH groups. The level of miR-217 expression showed a significantly negative correlation with Hcy and TG. It is more important that miR-217 showed the most specific expression patterns in all patients with atherosclerosis (ATH and Hhcy + ATH groups). Recently, miR-217 was pinpointed to expression in human carotid plaques, which promotes endothelial senescence via the regulation of silent information regulator (SIRT) 1 expression (Menghini et al., 2009; Duan et al., 2016). SIRT1 has been reported to serve as a key regulator of vascular endothelial homeostasis controlling angiogenesis and endothelial dysfunction (Menghini et al., 2009). Therefore, miR-217 participates as an early step in the pathogenesis of atherosclerosis. According to our study, miR-217 has a diagnostic value in Hhcy patients and ATH patients, which may be useful in predicting the progress of atherosclerosis in Hhcy patients.

By all means, some limitations of our study should be necessarily acknowledged. First, miR-30 showed no significant difference for all subjects because of smaller sample size. This study was a single-center study and consequently multicenter studies with larger sample sizes are needed. Second, carotid intima-media thickness cannot apply as an evaluation index for atherosclerosis due to the lack of equipment in the hospital. Finally, the ultrasonic resolution is based on the operator's experience and level of technical skills and so may bring great influence on diagnosis and lead to a greater error.

In summary, this study provided a substantial clinical significance. To our knowledge, this study can be regarded as the first to evaluate atherosclerosis-specific circulating miRNA expression profiles in patients with Hhcy. We have shown that five of the six atherosclerosis-related circulating miRNAs had significantly different expressions among NC, Hhcy, Hhcy + ATH, and ATH subjects. Our study identified that miR-217 may be helpful in predicting the progress of atherosclerosis in Hhcy patients. We are providing a novel way of identifying the severity of atherosclerosis, which can also work as potential biomarkers for Hhcy. Moreover, the results of the current studies highlight a further insight into the function of circulating miRNAs in the process and progression of atherosclerosis. However, prospective large-scale studies are required to determine the potential value of circulating miRNAs for atherosclerosis in patients with Hhcy.

Footnotes

Acknowledgments

This work was supported by the National Natural Science Foundation of China (grant no. 81160001). We thank Dr. Azeem from Pakistan for assistance in reviewing the language.

Disclosure Statement

No competing financial interests exist.

References

Arslan

, Tüzün

F.A.

, Arslan

, Demir

, Tamer

, et al. (2016). The relationship between serum paraoxonase levels and carotid atherosclerotic plaque formation in Alzheimer's patients. Neurol Neurochirurgia Polska, 50, 403–409.

Baldan

, and Fernandez-Hernando

(2016). Truths and controversies concerning the role of miRNAs in atherosclerosis and lipid metabolism. Curr Opin Lipidol, 27, 623–629.

Baraniskin

, Nopel-Dunnebacke

, Ahrens

, Jensen

S.G.

, Zollner

, et al. (2013). Circulating U2 small nuclear RNA fragments as a novel diagnostic biomarker for pancreatic and colorectal adenocarcinoma. Int J Cancer, 132, E48–E57.

Chernyavskiy

, Veeranki

, Sen

, and Tyagi

S.C.

(2016). Atherogenesis: hyperhomocysteinemia interactions with LDL, macrophage function, paraoxonase 1, and exercise. Ann N Y Acad Sci, 1363, 138–154.

Climent

, Quintavalle

, Miragoli

, Chen

, Condorelli

, et al. (2015). TGFbeta triggers miR-143/145 transfer from smooth muscle cells to endothelial cells, thereby modulating vessel stabilization. Circ Res, 116, 1753–1764.

Deddens

J.C.

, Colijn

J.M.

, Oerlemans

MIFJ

, Pasterkamp

, Chamuleau

S.A.

, et al. (2013). Circulating microRNAs as novel biomarkers for the early diagnosis of acute coronary syndrome. J Cardiovasc Transl Res, 6, 884–898.

Dentelli

, Rosso

, Orso

, Olgasi

, Taverna

, et al. (2010). MicroRNA-222 controls neovascularization by regulating signal transducer and activator of transcription 5A expression. Arterioscler Thromb Vasc Biol, 30, 1562–1568.

Dong

Y.M.

, Liu

X.X.

, Wei

G.Q.

, Da

Y.N.

, Cha

, et al. (2015). Prediction of long-term outcome after acute myocardial infarction using circulating miR-145. Scand J Clin Lab Invest, 75, 85–91.

, Yu

, Wang

, Hui

, Carnevale

, et al. (2014). MicroRNA-155 deficiency results in decreased macrophage inflammation and attenuated atherogenesis in apolipoprotein E-deficient mice. Arterioscler Thromb Vasc Biol, 34, 759–767.

10.

Duan

, Li

, Yan

, Yang

, Wu

, et al. (2016). MicroRNA-217 suppresses homocysteine-induced proliferation and migration of vascular smooth muscle cells via N-methyl-D-aspartic acid receptor inhibition. Clin Exp Pharmacol Physiol, 43, 967–975.

11.

Gao

, Peng

, Ren

, He

N.-y.

, Li

, et al. (2016). Functional regulatory roles of microRNAs in atherosclerosis. Clin Chim Acta, 460, 164–171.

12.

Garvey

S.M.

, Sinden

D.S.

, Schoppee Bortz

P.D.

, and Wamhoff

B.R.

(2010). Cyclosporine up-regulates Kruppel-like factor-4 (KLF4) in vascular smooth muscle cells and drives phenotypic modulation in vivo. J Pharmacol Exp Ther, 333, 34–42.

13.

Giral

, Kratzer

, and Landmesser

(2016). MicroRNAs in lipid metabolism and atherosclerosis. Best Pract Res Clin Endocrinol Metab, 30, 665–676.

14.

Hosin

A.A.

, Prasad

, Viiri

L.E.

, Davies

A.H.

, and Shalhoub

(2014). MicroRNAs in atherosclerosis. J Vasc Res, 51, 338–349.

15.

Huang

, Chen

, Zhou

, Yu

, Huang

, et al. (2016). Circulating miR-30 is related to carotid artery atherosclerosis. Clin Exp Hypertens, 38, 489–494.

16.

Hulsmans

, and Holvoet

(2013). MicroRNA-containing microvesicles regulating inflammation in association with atherosclerotic disease. Cardiovasc Res, 100, 7–18.

17.

Jeon

S.B.

, Kang

D.W.

, Kim

J.S.

, and Kwon

S.U.

(2014). Homocysteine, small-vessel disease, and atherosclerosis: an MRI study of 825 stroke patients. Neurology, 83, 695–701.

18.

Kesherwani

, Nandi

S.S.

, Sharawat

S.K.

, Shahshahan

H.R.

, and Mishra

P.K.

(2015). Hydrogen sulfide mitigates homocysteine-mediated pathological remodeling by inducing miR-133a in cardiomyocytes. Mol Cell Biochem, 404, 241–250.

19.

Kovacic

, and Bakran

(2012). Genetic susceptibility to atherosclerosis. Stroke Res Treat 2012, 362941.

20.

Kuhlmann

J.D.

, Baraniskin

, Hahn

S.A.

, Mosel

, Bredemeier

, et al. (2014). Circulating U2 small nuclear RNA fragments as a novel diagnostic tool for patients with epithelial ovarian cancer. Clin Chem, 60, 206–213.

21.

Lee

M.S.

, Lee

F.Y.

, Chen

Y.L.

, Sung

P.H.

, Chiang

H.J.

, et al. (2017). Investigated the safety of intra-renal arterial transfusion of autologous CD34+ cells and time courses of creatinine levels, endothelial dysfunction biomarkers and micro-RNAs in chronic kidney disease patients-phase I clinical trial. Oncotarget, 8, 17750–17762.

22.

, Kong

, Chen

, Liu

, Hu

, et al. (2016). MiR-155 acts as an anti-inflammatory factor in atherosclerosis-associated foam cell formation by repressing calcium-regulated heat stable protein 1. Sci Rep 6, 21789.

23.

Liao

(2006). Hyperhomocysteinemia decreases circulating high-density lipoprotein by inhibiting apolipoprotein A-I protein synthesis and enhancing HDL cholesterol clearance. Circ Res, 99, 598–606.

24.

Libby

, Ridker

P.M.

, and Hansson

G.K.

(2011). Progress and challenges in translating the biology of atherosclerosis. Nature, 473, 317–325.

25.

Liu

, Cheng

, Zhang

, Lin

, Yang

, et al. (2009). A necessary role of miR-221 and miR-222 in vascular smooth muscle cell proliferation and neointimal hyperplasia. Circ Res, 104, 476–487.

26.

Liu

, Yuan

, Chen

, Zhang

, Chen

, et al. (2017). Circulating miR-208b: a potentially sensitive and reliable biomarker for the diagnosis and prognosis of acute myocardial infarction. Clin Lab, 63, 101–109.

27.

, Zhou

Y.H.

, Wu

, Shao

, Lu

C.L.

, et al. (2015). The changes in miR-130b levels in human serum and the correlation with the severity of diabetic nephropathy. Diabetes Metab Res Rev, 31, 717–724.

28.

Maegdefessel

(2014). The emerging role of microRNAs in cardiovascular disease. J Intern Med, 276, 633–644.

29.

Matsuzaki

, Fujita

, Jingushi

, Kawashima

, Ujike

, et al. (2017). MiR-21-5p in urinary extracellular vesicles is a novel biomarker of urothelial carcinoma. Oncotarget, 8, 24668–24678.

30.

Meng

, Liu

, Zhou

, Pan

, Zhai

, et al. (2016). Polyphenols and polypeptides in Chinese rice wine inhibit homocysteine-induced proliferation and migration of vascular smooth muscle cells. J Cardiovasc Pharmacol, 67, 482–490.

31.

Menghini

, Casagrande

, Cardellini

, Martelli

, Terrinoni

, et al. (2009). MicroRNA 217 modulates endothelial cell senescence via silent information regulator 1. Circulation, 120, 1524–1532.

32.

Mitchell

P.S.

, Parkin

R.K.

, Kroh

E.M.

, Fritz

B.R.

, Wyman

S.K.

, et al. (2008). Circulating microRNAs as stable blood-based markers for cancer detection. Proc Natl Acad Sci U S A, 105, 10513–10518.

33.

Navickas

, Gal

, Laucevičius

, Taparauskaitė

, Zdanytė

, et al. (2016). Identifying circulating microRNAs as biomarkers of cardiovascular disease: a systematic review. Cardiovasc Res, 111, 322–337.

34.

Nazari-Jahantigh

, Wei

, Noels

, Akhtar

, Zhou

, et al. (2012). MicroRNA-155 promotes atherosclerosis by repressing Bcl6 in macrophages. J Clin Invest, 122, 4190–4202.

35.

Poliseno

, Tuccoli

, Mariani

, Evangelista

, Citti

, et al. (2006). MicroRNAs modulate the angiogenic properties of HUVECs. Blood, 108, 3068–3071.

36.

Schaffer

, Verdoia

, Cassetti

, Marino

, Suryapranata

, et al. (2014). Relationship between homocysteine and coronary artery disease. Results from a large prospective cohort study. Thromb Res, 134, 288–293.

37.

Soufi-Zomorrod

, Hajifathali

, Kouhkan

, Mehdizadeh

, Rad

S.M.

, et al. (2016). MicroRNAs modulating angiogenesis: miR-129-1 and miR-133 act as angio-miR in HUVECs. Tumour Biol, 37, 9527–9534.

38.

Sturn

, Quackenbush

, and Trajanoski

(2002). Genesis: cluster analysis of microarray data. Bioinformatics, 18, 207–208.

39.

Tian

F.J.

, An

L.N.

, Wang

G.K.

, Zhu

J.Q.

, Li

, et al. (2014). Elevated microRNA-155 promotes foam cell formation by targeting HBP1 in atherogenesis. Cardiovasc Res, 103, 100–110.

40.

Torella

, Iaconetti

, Catalucci

, Ellison

G.M.

, Leone

, et al. (2011). MicroRNA-133 controls vascular smooth muscle cell phenotypic switch in vitro and vascular remodeling in vivo. Circ Res, 109, 880–893.

41.

Tribouilloy

C.M.

, Peltier

, Iannetta Peltier

M.C.

, Trojette

, Andrejak

, et al. (2000). Plasma homocysteine and severity of thoracic aortic atherosclerosis. Chest, 118, 1685–1689.

42.

Urbich

, Kuehbacher

, and Dimmeler

(2008). Role of microRNAs in vascular diseases, inflammation, and angiogenesis. Cardiovasc Res, 79, 581–588.

43.

Wang

G.K.

, Zhu

J.Q.

, Zhang

J.T.

, Li

, et al. (2010). Circulating microRNA: a novel potential biomarker for early diagnosis of acute myocardial infarction in humans. Eur Heart J, 31, 659–666.

44.

Wei

, Zhu

, Corbalan-Campos

, Heyll

, Weber

, et al. (2015). Regulation of Csf1r and Bcl6 in macrophages mediates the stage-specific effects of microRNA-155 on atherosclerosis. Arterioscler Thromb Vasc Biol, 35, 796–803.

45.

Zhang

H.P.

, Wang

Y.H.

, Cao

C.J.

, Yang

X.M.

, Ma

S.C.

, et al. (2016). A regulatory circuit involving miR-143 and DNMT3a mediates vascular smooth muscle cell proliferation induced by homocysteine. Mol Med Rep, 13, 483–490.

46.

Zhao

, Zhao

S.P.

, and Zhao

Y.H.

(2015). MicroRNA-143/-145 in cardiovascular diseases. Biomed Res Int 2015, 531740.

47.

Zhu

G.F.

, Yang

L.X.

, Guo

R.W.

, Liu

, Shi

Y.K.

, et al. (2014). MicroRNA-155 is inversely associated with severity of coronary stenotic lesions calculated by the Gensini score. Coron Artery Dis, 25, 304–310.

48.

Zhu

, Zhang

, Chen

, Liu

, Lin

, et al. (2011). Endothelial enriched microRNAs regulate angiotensin II-induced endothelial inflammation and migration. Atherosclerosis, 215, 286–293.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.07 MB

0.09 MB

0.04 MB

0.09 MB

0.07 MB

0.02 MB