The extracellular miRNAs circulate in the bloodstream and may serve as novel diagnostic and therapeutic biomarkers. The aim of the present study was to investigate circulating Toll-like receptor 4 (TLR4)-responsive miRNA expression in patients with coronary artery disease (CAD) and to examine the effects of renin–angiotensin system (RAS) blockade and statins on miRNA levels. This study included 41 patients with CAD and 20 subjects without CAD (non-CAD). Plasma TLR4-responsive miRNA samples were analysed using a microarray assay for 1700 human miRNA. The candidate miRNAs were verified with real-time reverse transcription (RT)-PCR. Patients with CAD were randomized to 12 months of combined treatment with either telmisartan and atorvastatin [angiotensin II receptor blocker (ARB)] or enalapril and atorvastatin [angiotensin-converting enzyme inhibitor (ACEI)]. Plasma samples were obtained from peripheral blood at baseline and after 12 months. The microarray assay showed significant differences in seven TLR4-responsive miRNAs between the CAD and non-CAD groups (P<0.05). Real-time PCR verified that miR-31, miR-181a, miR-16 and miR-145 were significantly lower in the CAD group than in the non-CAD group (P<0.01). Levels of TLR4 protein were higher in the CAD group than in the non-CAD group (P<0.01) and were negatively correlated with levels of TLR4-responsive miRNAs. Receiver operating characteristic (ROC) curve analysis revealed that a panel of these four miRNAs was sensitive and specific enough to distinguish CAD from non-CAD [area under the curve (AUC)=0.93, 95% CI (confidence interval)=0.99–0.87]. Both ARB and ACEI groups showed increased TLR4-responsive miRNAs and diminished levels of TLR4 protein (P<0.05). Changes in miRNAs and TLR4 levels were greater in the ARB group than in the ACEI group (P<0.05). Circulating TLR4-responsive miRNAs including miR-31, miR-181a, miR-16 and miR-145 were significantly lower in patients with CAD compared with controls and these miRNAs may be involved in the pathogenesis of CAD.

CLINICAL PERSPECTIVES

  • Our previous reports have demonstrated that activation of the TLR4 signal is involved in the downstream release of inflammatory cytokines in circulating monocytes obtained from patients with CAD.

  • The present study shows a novel finding that circulating TLR4-responsive miRNAs including miR-31, miR-181a, miR-16 and miR-145 are associated with activation of TLR4 signal in patients with CAD. We have also carried out a randomized and prospective study using RAS blockade and statins. After 12 months of treatment, both ARB and ACEI groups showed increased TLR4-responsive miRNAs. ROC curve analysis revealed that a panel of these four miRNAs was sensitive and specific enough to distinguish CAD from non-CAD.

  • Circulating TLR4-responsive miRNAs including miR-31, miR-181a, miR-16 and miR-145 were significantly lower in patients with CAD compared with controls, these miRNAs may be involved in the pathogenesis of CAD.

INTRODUCTION

Coronary artery disease (CAD) is still a leading cause of death around the world. Chronic inflammation of the arterial wall is a key element in the pathogenesis of atherosclerosis [1]. Sites of atherosclerotic plaque development in the arterial wall are characterized by monocyte infiltration [2]. It has been reported that miRNAs are the most abundant family of small non-coding RNAs and regulate mRNA translation of target genes through the RNA interference pathway [3,4]. Our previous reports have demonstrated that dysregulation of some intracellular miRNAs including toll-like receptor 4 (TLR4)-responsive miRNAs in circulating monocytes and vascular endothelial cell-enriched miRNAs in circulating endothelial progenitor cells might contribute to the pathogenesis of coronary atherosclerosis [5,6]. Studies have shown that miRNAs are abundantly present in body fluid and can be used as biomarkers for some diseases [79]. It has also been reported that circulating levels of vascular- and inflammation-associated miRNAs, such as miR-126, miR-17, miR-92a and miR-155, are down-regulated in patients with CAD [10]. The mechanism by which miRNAs are released into the circulation is unclear. However, increasing evidence suggests that miRNAs are actively secreted in microvesicles including apoptotic bodies and lipoproteins [1113]. On the basis of these observations, circulating miRNAs were given special attention, since they may have a potential role to play in diagnosis and prognosis of CAD.

Clinical studies have reported that renin–angiotensin system (RAS) blockade, including angiotensin II receptor blockers (ARBs) and angiotensin-converting enzyme inhibitors (ACEIs), have an anti-atherosclerotic effect in patients with cardiovascular risk factors or CAD [14,15]. It has also been reported that 3-hydroxy-3-methylglutaryl-CoA reductase inhibitors (statins) have anti-inflammatory properties by virtue of their pleiotropic effects [16]. Moreover, an experimental study has shown that RAS blockade and statins have a combined effect on inflammatory profiles, such as inflammatory cytokines and atherosclerosis [17]. However, it remains uncertain whether combined treatment with RAS blockade and statins would affect levels of circulating miRNAs in patients with CAD.

In the present study, our aim was to explore the regulation of circulating TLR4-responsive miRNAs in CAD and whether this could be modified by combined treatment with RAS blockade and a statins by comparing the effects of ARB with those of ACEI in patients with CAD.

MATERIALS AND METHODS

Study population

Forty one patients with stable CAD were admitted to our hospital for coronary angiography. None of these patients had previous exposure to statins or RAS blockade therapies. CAD was diagnosed on the basis of the presence of (i) a history of typical chest pain on effort, (ii) documented exercise-induced myocardial ischaemia, (iii) angiographically proven significant coronary stenosis and (iv) absence of acute coronary syndromes (ACS) for 6 months before blood sampling. Patients were excluded from the study if they had clinical signs of acute infection, severe renal failure (serum creatine levels > 3 mg/dl) or rheumatoid disease or if they were suspected of having a malignant or primary wasting disorder.

Twenty control subjects were identified from a list of outpatients attending our university hospital for treatment of hypertension, dyslipidaemia, arrhythmia or diabetes mellitus and showing no evidence of CAD on history or physical examination. Subjects were selected from the case loads of clinicians who were not related to the present study in order to reduce bias and were stratified according to sex and age to match each CAD case. Peripheral blood samples were obtained from all subjects. Approval for the study protocol was obtained from the ethics committee of the Iwate Medical University School of Medicine (H17–73) and written informed consent was obtained from all subjects.

Study design

The present study was designed as a prospective, randomized and open study. The randomization list, with an allocation ratio 1:1 and block size of four, was generated by computer. The research nurse controlled the randomization list and assigned participants to the present study at the first visit. First, patients with CAD were stratified according to sex and age in years (<40 years, 40–50 years, 51–60 years, 61–70 years and 71 years and above). After stratification, patients with CAD were randomized to one of the two treatment groups. One treatment group received telmisartan (40 mg/day) and atorvastatin (10 mg/day) (ARB group), whereas the other received enalapril (5 mg/day) and atorvastatin (10 mg/day) (ACEI group) for a period of 12 months.

Blood sampling

Fasting peripheral blood was collected from patients with CAD in the morning after an overnight fast before treatment with RAS blockade and statin for baseline data and again after 12 months of treatment with RAS blockade and statin. Fasting peripheral blood was also collected from controls in the morning after an overnight fast. EDTA–plasma was separated after centrifugation at 2500 g for 15 min and the supernatant was stored at–80°C until use.

Cell preparation

Peripheral blood mononuclear cells (PBMCs) were isolated from peripheral blood samples obtained from all subjects by Ficoll-Paque density gradient centrifugation. To exclude platelet contamination, isolated PBMCs were washed and centrifuged twice with PBS containing EDTA for 15 min at 400 g at 20°C [18]. After washing three times with PBS, the PBMCs were resuspended at a final concentration of 1×106 cells/ml in RPMI1640 (Sigma–Aldrich).

RNA extraction and TLR4-responsive miRNA screening

For the screening of 146 differentially regulated TLR4-responsive miRNAs (miR-31, miR-7/7ab, miR-216b/216b-5p, miR-140/140–5p/876–3p/1244, miR-145, miR-103a/107/107ab, miR-217, miR-203, miR-33a-3p/365/365–3p, miR-21/590–5p, miR-503, miR-153, miR-15abc/16/16abc/195/322/424/497/1907, let-7/98/4458/4500, miR-181abcd/4262, miR-34ac/34bc-5p/449abc/449c-5p, miR-146ac/146b-5p, miR-130ac/301ab/301b/301b-3p/454/721/4295/3666, miR-122/122a/1352, miR-144, miR-23abc/23b-3p, miR-25/32/92abc/363/363–3p/367, miR-216a, miR-455–5p, miR-93/93a/105/106a/291a-3p/294/295/302abcde/372/373/428/519ad/520be/520acd-3p/1378/1420ac, miR-26ab/1297/4465, miR-200bc/429/548a, miR-190/190ab, miR-125a-5p/125b-5p/351/670/4319, miR-1ab/206/613, miR-425/425–5p/489, miR-101/101ab, miR-219–5p/508/508–3p/4782–3p, miR-18ab/4735–3p, miR-338/338–3p) based on miRBase (the miRNA database) [19], 10 RNA samples obtained from five subjects with CAD (0 day) and five subjects with non-CAD were pooled and miRNA expression was measured by microarray screening. RNA was extracted from the plasma using 3D-Gene RNA extraction reagent from a liquid sample according to the manufacturer's instructions (Toray). Extracted total RNA was labelled with a 3D-Gene miRNA labelling kit (Toray). Labelled RNAs were hybridized on to 3D-Gene Human miRNA Oligo chips (Toray). The annotation and oligonucleotide sequences of the probes conformed to the miRBase miRNA database. After stringent washing, fluorescent signals were scanned with the 3D-Gene Scanner (Toray Industries) and analysed using 3D-Gene Extraction software (Toray Industries). The raw data of each spot was normalized by substitution with a mean intensity of the background signal determined by all blank spots’ signal intensities with 95% confidence intervals (CIs). Measurements of spots with signal intensities greater than two S.D. from the background signal intensity were considered to be valid. A relative expression level of a given miRNA was calculated by comparing the signal intensities of the valid spots throughout the microarray experiments. The normalized data were globally normalized per array such that the median of the signal intensity was adjusted to 25.

RNA isolation for real-time RT-PCR

RNA was extracted using mirVana Paris miRNA isolation kit (Ambion) in accordance with the manufacturer's instructions. Because there is no suitable endogenous control for plasma RNA, miR-39, a miRNA from Caenorhabditis elegans not present in mammalian tissues, was added as a spiked-in control. Plasma (200 μl) was added to denaturating solution and mixed thoroughly. Synthetic miR-39 was then added to the mixture and mixed thoroughly. After chloroform addition and phase separation, the aqueous layer was mixed with absolute ethanol. The solution was then loaded on to the cartridge provided with the mirVana Paris miRNA isolation kit. The columns were washed in accordance with the manufacturer's instructions and the RNA was eluted in 50 nuclease-free H2O (95°C). To minimize DNA contamination, the eluted RNA was treated with DNase I (Ambion AM1906).

Real-time RT-PCR

According to array results, candidate TLR4-responsive miRNAs were chosen for further validation with real-time reverse transcription (RT)-PCR. Complementary DNA was synthesized using the Reverse Transcription TaqMan MicroRNA Reverse Transcription Kit (Applied Biosystems) according to the manufacturer's instructions. Quantitative RT-PCR was performed with specific TaqMan miRNA assays (Applied Biosystems) for quantification of TLR4-responsive microRNAs (e.g. miR-31, miR-181a, miR-16 and miR-143). The amplification steps consisted of denaturation at 95°C, followed by 40 cycles of denaturation at 95°C for 15 s and then annealing at 60°C for 1 min. Relative quantification was carried out using the ΔΔ threshold cycle (Ct) method for recurrent compared with primary miR-39 which was spiked-in as external control [20]. Replicates with a Ct > 40 were excluded. The assay was run in triplicate for each case to allow for assessment of technical variability. To account for PCR amplification of contaminating genomic DNA, a control without RT was included.

Levels of circulating miRNA were normalized to miR-39 which was spiked-in as an external control. To improve the accuracy of real-time RT-PCR for quantification, amplifications were performed in triplicate for each RNA sample. To account for PCR amplification of contaminating genomic DNA, a control without RT was included.

Flow cytometric analysis

The amount of TLR4 and CD14 on PBMC cell surfaces was measured by FACS. Isolated PBMCs were incubated with FITC-conjugated mouse anti-human TLR4 antibodies (Santa Cruz Biotechnology) and PerCP-conjugated CD14 antibody (Becton Dickinson). Isotype-matched irrelevant control IgG was used as a control (Becton Dickinson). TLR4 levels in CD14-positive cells were measured by a FACScan flow cytometer (Becton Dickinson) and shown as mean fluorescence intensity (MFI).

Laboratory data

Laboratory data were measured by standard biochemical methods in our hospital laboratory. All analyses were performed in the same run from samples that were frozen and stored at −80°C immediately after centrifugation. Fasting serum high-density lipoprotein (HDL)-cholesterol and triacylglycerol (triglyceride) levels were calculated by routine enzymatic methods. Levels of low-density lipoprotein (LDL)-cholesterol were calculated using a direct Friedewald's formula [21]. High sensitivity C-reactive protein (hsCRP) was quantified by a latex-enhanced immunonephelometric assay (detection range: 0.035–2.200 mg/dl; CardioPhase hs-CRP, Dade Behring Marburg GmbH).

Statistical analysis

All values are presented as means±S.D. Kolmogorov–Smirnov analysis was performed to assess data distribution. Unpaired Student's t test was performed for normally distributed data and non-parametric Mann–Whitney test was performed where this was not appropriate. Statistical analysis of categorical variables was also carried out using χ2 analysis and Fisher exact analysis where the number of subjects in the category was less than 10. After 12 months of treatment with statins, comparisons between the ARB and ACEI groups were analysed by two-way ANOVA repeated measure for normally distributed variables and by the Kruskal-Wallis test for non-normally distributed variables. When applicable, significant differences were further analysed with Dunnett post-hoc tests. Pearson's correlation coefficients were used to examine the relationship between miRNAs and TLR4 levels. The receiver operating characteristic (ROC) curve was plotted to determine the cut-off levels for predicting CAD. The level was set to obtain the maximum sensitivity plus specificity. A value of P<0.05 was considered statistically significant.

RESULTS

Baseline and clinical characteristics

Baseline characteristics of the two study groups are shown in Table 1. There were no significant differences in age, percent male, body mass index (BMI) and blood pressure between the CAD and non-CAD groups. There were significant differences in other parameters between the two groups (P<0.05).

Table 1
Baseline and clinical characteristics of the study population

Values are means±S.D. or number of subjects (percentage). HbA1c, glycated haemoglobin; DBP, diastolic blood pressure; N/A, not available; SBP, systolic blood pressure. *P<0.05 compared with non-CAD group. †P<0.05 compared with baseline.

CAD group
ParameterTotal (n=41)ARB (n=21)ACEI (n=20)Non-CAD group (n=20)
Age (years) 66.4±13.2 65.8±13.8 67.0±12.8 61.6±12.5 
Male (n35 (85%) 19 (90%) 16 (80%) 17 (85%) 
BMI (kg/m225.5±3.3 26.2±3.0 24.8±3.6 25.4±3.3 
Hypertension (n30 (73%)* 17 (81%)* 13 (65%) 11 (55%) 
Diabetes mellitus (n23 (56%)* 13 (62%)* 10 (50%)* 4 (20%) 
Previous angina (n8 (20%)* 4 (19%)* 4 (20%)* 
Smoking (n14 (34%)* 6 (29%)* 8 (40%)* 3 (15%) 
SBP (mmHg) 128.1±16.0 129.8±17.4 126.3±16.8 128.3±16.9 
DBP (mmHg) 69.0±9.4 69.7±9.5 68.3±9.6 68.6±10.8 
HbA1c (%) 5.99±1.13* 5.95±0.90* 6.03±1.37* 5.53±0.83 
HDL-cholesterol (mg/dl)     
 Baseline 47.1±11.7 46.9±10.8 47.4±12.9 45.6±9.6 
 Follow-up 46.0±13.6 45.0±13.2 46.5±12.2 N/A 
LDL-cholesterol (mg/dl)     
 Baseline 117.6±27.8 117.4±31.1 117.5±24.7 112.1±14.9 
 Follow-up 80.9±19.1 79.8±19.7† 81.3±19.6† N/A 
hsCRP (mg/dl)     
 Baseline 0.14±0.35* 0.16±0.45* 0.12±0.20* 0.05±0.03 
 Follow-up 0.08±0.12* 0.09±0.09† 0.08±0.16† N/A 
Medication (n    
 Nitrates 27 (66%) 13 (62%) 14 (70%) N/A 
 Calcium antagonists 23 (56%) 12 (57%) 11(55%) N/A 
 β-Blockers 21(51%) 10 (48%) 11 (55%) N/A 
CAD group
ParameterTotal (n=41)ARB (n=21)ACEI (n=20)Non-CAD group (n=20)
Age (years) 66.4±13.2 65.8±13.8 67.0±12.8 61.6±12.5 
Male (n35 (85%) 19 (90%) 16 (80%) 17 (85%) 
BMI (kg/m225.5±3.3 26.2±3.0 24.8±3.6 25.4±3.3 
Hypertension (n30 (73%)* 17 (81%)* 13 (65%) 11 (55%) 
Diabetes mellitus (n23 (56%)* 13 (62%)* 10 (50%)* 4 (20%) 
Previous angina (n8 (20%)* 4 (19%)* 4 (20%)* 
Smoking (n14 (34%)* 6 (29%)* 8 (40%)* 3 (15%) 
SBP (mmHg) 128.1±16.0 129.8±17.4 126.3±16.8 128.3±16.9 
DBP (mmHg) 69.0±9.4 69.7±9.5 68.3±9.6 68.6±10.8 
HbA1c (%) 5.99±1.13* 5.95±0.90* 6.03±1.37* 5.53±0.83 
HDL-cholesterol (mg/dl)     
 Baseline 47.1±11.7 46.9±10.8 47.4±12.9 45.6±9.6 
 Follow-up 46.0±13.6 45.0±13.2 46.5±12.2 N/A 
LDL-cholesterol (mg/dl)     
 Baseline 117.6±27.8 117.4±31.1 117.5±24.7 112.1±14.9 
 Follow-up 80.9±19.1 79.8±19.7† 81.3±19.6† N/A 
hsCRP (mg/dl)     
 Baseline 0.14±0.35* 0.16±0.45* 0.12±0.20* 0.05±0.03 
 Follow-up 0.08±0.12* 0.09±0.09† 0.08±0.16† N/A 
Medication (n    
 Nitrates 27 (66%) 13 (62%) 14 (70%) N/A 
 Calcium antagonists 23 (56%) 12 (57%) 11(55%) N/A 
 β-Blockers 21(51%) 10 (48%) 11 (55%) N/A 

Clinical characteristics in ARB and ACEI groups

As shown in Table 1, there were no significant differences in baseline and clinical characteristics between the ARB and ACEI groups.

All patients with CAD were monitored once a month on an outpatient basis to check compliance and to record side effects or major vascular events. There were no dropout cases due to poor compliance or side effects during the 12 month follow-up period. In addition, no patients experienced major vascular events. Levels of LDL-cholesterol and hsCRP were decreased in both ARB and ACEI groups after 12 months of treatment (Table 1).

Plasma TLR4-responsive miRNA profiling

TLR4-responsive miRNA measurement using microarray screening showed seven miRNAs (miR-200c, miR-31, miR-181a, miR-519d, miR-16, miR-145, let-7) as the miRNAs most significantly lower in the CAD group compared with the non-CAD group. These seven TLR4-responsive miRNAs were considered for further validation using real-time RT-PCR.

Real-time RT-PCR for TLR4-responsive miRNA

The present study used cel-miR-39 as a normalization control. Plasma levels of miR-31, miR-181a, miR-16 and miR-145 were lower in the CAD group than the non-CAD group (Figure 1). There were no significant differences in the levels of miR-200c, miR-519d or let-7 between the CAD and non-CAD groups (miR-200c; 2.12±1.85 compared with 2.54±1.83: miR-519d; 1.63±1.22 compared with 1.59±1.32: let-7; 0.56±0.19 compared with 0.61±0.15; CAD group compared with non-CAD group, not significant).

Plasma levels of miR-31, miR-181a, miR-16 and miR-145 in CAD and non-CAD groups

Figure 1
Plasma levels of miR-31, miR-181a, miR-16 and miR-145 in CAD and non-CAD groups
Figure 1
Plasma levels of miR-31, miR-181a, miR-16 and miR-145 in CAD and non-CAD groups

Levels of TLR4 MFI in PBMCs

Levels of TLR4 MFI were higher in the CAD group than in the non-CAD group (P<0.01) (Figure 2). Levels of TLR4 MFI were weakly negatively correlated with plasma levels of miR-31, miR-181a, miR-16 and miR-145 in all subjects (TLR4 MFI compared with miR-31; R=–0.47, P<0.01: TLR4 MFI compared with miR-181a; R=–0.26, P<0.05: TLR4 MFI compared with miR-16; R=–0.26, P<0.05: TLR4 MFI compared with miR-145; R=–0.28, P<0.05; Figure 3).

Levels of TLR4 MFI in CAD and non-CAD groups

Figure 2
Levels of TLR4 MFI in CAD and non-CAD groups
Figure 2
Levels of TLR4 MFI in CAD and non-CAD groups

Correlation between miR-31 (A), miR-181a (B), miR-16 (C), miR-145 (D) and TLR4 MFI in all subjects

Figure 3
Correlation between miR-31 (A), miR-181a (B), miR-16 (C), miR-145 (D) and TLR4 MFI in all subjects
Figure 3
Correlation between miR-31 (A), miR-181a (B), miR-16 (C), miR-145 (D) and TLR4 MFI in all subjects

miRNA signature for identification of CAD

A ROC curve analysis was used to analyse the diagnostic accuracy of plasma miR-31, miR-181a, miR-16 and miR-145. ROC curve analyses revealed that plasma miR-31, miR-181a, miR-16 and miR-145 could serve as valuable biomarkers for differentiating CAD from controls with AUCs (areas under the ROC curve) of 0.78 (95% CI: 0.91–0.65), 0.75 (95% CI: 0.88–0.62) and 0.78 (95% CI: 0.93–0.64) and 0.75 (95% CI: 0.93–0.57) respectively (Figure 4). The combination of all four circulating miRNAs displayed the highest discriminatory power with AUC=0.93 (95% CI=0.99–0.87) and 79% sensitivity and 93% specificity for the identification of CAD (Figure 4A).

ROC curve analysis for miRNAs to identify CAD subjects

Figure 4
ROC curve analysis for miRNAs to identify CAD subjects

(A) Combination of miR-31, miR-181a, miR-16 and miR-145; (B) miR-31; (C) miR-181a; (D) miR-16; (E) miR-145.

Figure 4
ROC curve analysis for miRNAs to identify CAD subjects

(A) Combination of miR-31, miR-181a, miR-16 and miR-145; (B) miR-31; (C) miR-181a; (D) miR-16; (E) miR-145.

Effect of RAS blockade and statin on miRNAs and TLR4 levels

Figures 5 and 6 show changes in miR-31, miR-181a, miR-16, miR-145 and TLR4 levels after combined treatment with RAS blockade and statin. There was no significant difference in baseline levels of miR-31, miR-181a, miR-16, miR-145 and TLR4 MFI between the ARB and ACEI groups (miR-31: 1.48±1.09 compared with 1.55±1.04; miR-181a: 1.75±0.67 compared with 1.82±0.40; miR-16: 1.67±0.21 compared with 1.71±0.35; miR-145: 1.55±0.69 compared with 1.55±0.65; TLR4 MFI: 11.58±5.61 compared with 12.27±5.99; ARB group compared with ACEI group, not significant).

Effect of ARB (n=21) and ACEI (n=20) on miR-31, miR-181a, miR-16 and miR-145 levels in patients with CAD.

Figure 5
Effect of ARB (n=21) and ACEI (n=20) on miR-31, miR-181a, miR-16 and miR-145 levels in patients with CAD.

Treatment with ARB significantly increased miR-31 (A), miR-181a (B), miR-16 (C) and miR-145 (D) levels compared with ACEI.

Figure 5
Effect of ARB (n=21) and ACEI (n=20) on miR-31, miR-181a, miR-16 and miR-145 levels in patients with CAD.

Treatment with ARB significantly increased miR-31 (A), miR-181a (B), miR-16 (C) and miR-145 (D) levels compared with ACEI.

For both ARB and ACEI, combined treatment resulted in increased miR-31, miR-181a, miR-16 and miR-145 levels after 12 months (miR-31: 1.51±1.05 compared with 1.67±0.26; miR-181a: 1.78±0.55 compared with 2.83±1.20; miR-16: 1.69±0.29 compared with 2.45±0.69; miR-145: 1.55±0.74 compared with 2.09±1.09; baseline compared with 12 months, all P<0.01). On the other hand, levels of TLR4 MFI were decreased after 12 months (11.92±5.73 compared with 6.09±3.89; baseline compared with 12 months, P<0.01). Increases in miR-31, miR-181a, miR-16 and miR-145 levels after treatment were greater in the ARB group than in the ACEI group (Figure 5). In addition, decreases in TLR4 MFI after treatment were greater in the ARB group than in the ACEI group (Figure 6).

Effect of ARB (n=21) and ACEI (n=20) on TLR4 MFI in patients with CAD

Figure 6
Effect of ARB (n=21) and ACEI (n=20) on TLR4 MFI in patients with CAD

Treatment with ARB significantly decreased TLR4 MFI compared with ACEI.

Figure 6
Effect of ARB (n=21) and ACEI (n=20) on TLR4 MFI in patients with CAD

Treatment with ARB significantly decreased TLR4 MFI compared with ACEI.

DISCUSSION

The main findings of the present study are: (i) plasma levels of TLR4-responsive miRNAs were lower in the CAD group than in the non-CAD group; (ii) levels of TLR4 in PBMC were higher in the CAD group than in the non-CAD group and were negatively correlated with TLR4-responsive miRNAs; (iii) a randomized clinical study using RAS blockade (ARB or ACEI) and a statin showed both ARB and ACEI groups to have a marked increase in plasma levels of TLR4-responsive miRNAs and a decrease in TLR4 levels. In particular, changes in TLR4-responsive miRNAs and TLR4 levels after treatment were greater in the ARB group than in the ACEI group.

The existence of circulating miRNAs notwithstanding, high plasma RNase activity suggests that circulating miRNAs are shielded from enzymatic degeneration and then undergo immediate degradation to free RNA, such as during repeated freezing-thawing [7]. It has been reported that microvesicles, lipid vesicles, lipoprotein complexes, apoptotic bodies and exosomes protect and carry circulating miRNAs [13,22,23]. Previous reports have demonstrated that levels of distinct miRNAs in exosomes were higher than in their donor cells suggesting an active mechanism of miRNAs loading microvesicular bodies and exosomes [24,25]. Circulating miRNAs, including heart-, vascular- and muscle-specific miRNAs, have been reported to be biomarkers for acute myocardial infarction [26,27]. HDL has been implicated in shielding circulating miRNAs and scavenger receptor class B type I mediated miRNAs during delivery to recipient cells [12]. From these observations, the presence of miRNAs in microvesicles suggests that circulating miRNAs may act as an intercellular communication system potentially not only serving as biomarkers but also contributing to disease progression [28,29]. A novel finding of the present study is that miRNA profiling shows significantly lower levels of TLR4-responsive miRNAs (miR-31, miR-181a, miR-16 and miR-145) in the CAD group compared with the non-CAD group and the validation study using real-time RT-PCR confirmed the results from miRNA profiling of microarray screening in CAD. In addition, levels of TLR4 protein in PBMCs were higher in the CAD group than in the non-CAD group and were weakly negatively correlated with levels of circulating TLR4-responsive miRNAs. Our previous study has already shown an activation of TLR4 signalling pathway and dysregulation of intracellular TLR4-related miRNAs, such as miR-146a/b, in PBMCs obtained from patients with CAD [5]. A number of identified miRNAs, such as miR-146, miR-181a, miR-151, miR-31, miR-223, miR-145 and miR-155, are known to regulate various biological pathways including the immune system through the development and function of immune components [30]. It has also been reported that some miRNAs bound as ligands to receptors of the TLR family and thus controlled tumour immune response [31]. These observations permit speculation that repression of circulating TLR4-responsive miRNAs may attenuate binding to the TLR4 receptor and activate the TLR4 signalling pathway in CAD. Finally, activation of the TLR4 signalling pathway may be involved in maturation of atherosclerosis-related chemokines and the progression of coronary atherosclerosis in patients with CAD. Indeed, the ROC curve of TLR4-responsive miRNAs including miR-31, miR-181a, miR-16 and miR-145 showed that these TLR-4-responsive miRNAs appeared to be sensitive predictors of CAD.

In addition, our prospective randomized study has shown some important findings. Combined treatment with RAS blockade and statins increased circulating levels of TLR4-responsive miRNAs and reduced TLR4 levels in patients with CAD. RAS blockade including ARB and ACEI exert potent anti-atherosclerotic effects, which are mediated not only by their anti-hypertensive properties but also by their anti-inflammatory properties [32]. It has been reported that angiotensin II up-regulated TLR4-dependent signal and then induced an inflammatory response [33]. Our previous report has already demonstrated that combined treatment with RAS blockade and atorvastatin attenuated monocyte levels of TLR4 in patients with CAD [5]. RAS blockade may, therefore, decrease monocytic activation of TLR4 through inhibition of angiotensin II. Although the present study has not confirmed whether combined treatment with RAS blockade and atorvastatin directly or indirectly increased circulating levels of TLR4-responsive miRNAs, increases in circulating TLR4-responsive miRNAs after treatment may reflect down-regulation of the TLR4 signalling pathway in patients with CAD. In addition, changes in these levels were greater in the ARB group than in the ACEI group. An apolipoprotein E (apoE)-knockout mouse model treated with RAS blockade and statin showed that combined treatment with ARB and statins decreased inflammatory mediators including inflammatory cytokines compared with combined treatment with ACEI and statins or treatment with ARB, ACEI or statins individually [17]. An in vitro study has reported that telmisartan attenuated inflammatory cytokines, such as interleukin 6 and tumour necrosis factor-α, via suppression of cytokine-promoter genes [34]. These observations suggest that combined treatment with ARB and statin may have anti-inflammatory properties compared with ACEI. We, therefore, believe that changes in circulating levels of TLR4-responsive miRNA after treatment with RAS blockade and statin reflect a compensatory effect on inflammation. Further experimental studies will be necessary to explore the mechanisms by which CAD and the various therapies affect tissue compared with circulating miRNA levels.

Study limitation

A limitation of the present study is the small number of CAD patients receiving each type of therapy. The present study used relatively low doses of statins (atorvastatin=10 mg/day) and RAS blockade (telmisartan=40 mg/day and enalapril=5 mg/day) compared with other reports from western countries, because these are the most commonly prescribed statins and RAS blockade doses in Japan (maximum approved doses are: atorvastatin=20 mg daily; telmisartan=80 mg daily; enalapril=10 mg daily). However, there was no difference between the ARB and ACEI groups in blood pressure after 12 months of therapy (result not shown), which suggests that enalapril and telmisartan have the same anti-hypertensive effects. The present study has not identified whether combined treatment with RAS blockade and statin is really effective in terms of reducing blood pressure and preventing end-organ damage. In addition, the present study could not elucidate the mechanism by which combined treatment with RAS blockade and statin affects circulating TLR4-responsive miRNAs in CAD patients. Further studies will, therefore, be needed to determine a causal relationship between these therapeutic agents in CAD.

Conclusions

The present study has suggested that repression of TLR4-responsive miRNAs including miR-31, miR-181a, miR-16 and miR-145 may contribute to chronic inflammation via activation of TLR4 signalling pathway and may thereby play an important role in the progression of coronary atherosclerosis. In addition, combined treatment with RAS blockade and statin down-regulates TLR4 activity via expression of circulating TLR4-responsive miRNAs in CAD, possibly contributing to the beneficial effects of RAS blockade and statins on chronic inflammation in this disorder.

Abbreviations

     
  • ACEI

    angiotensin-converting enzyme inhibitor

  •  
  • ARB

    angiotensin II receptor blocker

  •  
  • AUC

    area under the curve

  •  
  • BMI

    body mass index

  •  
  • CAD

    coronary artery disease

  •  
  • CI

    confidence interval

  •  
  • HDL

    high-density lipoprotein

  •  
  • hsCRP

    high sensitivity C-reactive protein

  •  
  • LDL

    low-density lipoprotein

  •  
  • MFI

    mean fluorescence intensity

  •  
  • PBMC

    peripheral blood mononuclear cell

  •  
  • RAS

    renin–angiotensin system

  •  
  • ROC

    receiver operating characteristic

  •  
  • RT

    reverse transcription

  •  
  • TLR4

    Toll-like receptor 4

AUTHOR CONTRIBUTION

The present study was designed and conducted by Mamoru Satoh, Yuji Takahashi, Yoshihiro Morino and Motoyuki Nakamura. Data analysis was performed by Mamoru Satoh, Makiko Tamada, Yuh Ishikawa, Kan Takahashi, Tomonori Itoh and Tsuyoshi Tabuchi. The paper was written by Mamoru Satoh.

FUNDING

This study was supported by the General Scientific Research from the Japanese Ministry of Education, Science, Sports and Culture [grant number 25461138].

References

References
1
Libby
 
P.
 
Inflammation in atherosclerosis
Nature
2002
, vol. 
420
 (pg. 
868
-
874
)
[PubMed]
2
Weber
 
C.
Soehnlein
 
O.
 
ApoE controls the interface linking lipids and inflammation in atherosclerosis
J. Clin. Invest.
2011
, vol. 
121
 (pg. 
3825
-
3827
)
[PubMed]
3
Bartel
 
D.P.
 
MicroRNAs: genomics, biogenesis, mechanism, and function
Cell
2004
, vol. 
116
 (pg. 
281
-
297
)
[PubMed]
4
Kim
 
V.N.
 
Small RNAs: classification, biogenesis, and function
Mol. Cells
2005
, vol. 
19
 (pg. 
1
-
15
)
[PubMed]
5
Takahashi
 
Y.
Satoh
 
M.
Minami
 
Y.
Tabuchi
 
T.
Itoh
 
T.
Nakamura
 
M.
 
Expression of miR-146a/b is associated with the Toll-like receptor 4 signal in coronary artery disease: effect of renin-angiotensin system blockade and statins on miRNA-146a/b and Toll-like receptor 4 levels
Clin. Sci.
2011
, vol. 
119
 (pg. 
395
-
405
)
[PubMed]
6
Minami
 
Y.
Satoh
 
M.
Maesawa
 
C.
Takahashi
 
Y.
Tabuchi
 
T.
Itoh
 
T.
Nakamura
 
M.
 
Effect of atorvastatin on microRNA 221/222 expression in endothelial progenitor cells obtained from patients with coronary artery disease
Eur. J. Clin. Invest.
2009
, vol. 
39
 (pg. 
359
-
367
)
[PubMed]
7
Mitchell
 
P.S.
Parkin
 
R.K.
Kroh
 
E.M.
Fritz
 
B.R.
Wyman
 
S.K.
Pogosova-Agadjanyan
 
E.L.
Peterson
 
A.
Noteboom
 
J.
O’Briant
 
K.C.
Allen
 
A.
, et al 
Circulating microRNAs as stable blood-based markers for cancer detection
Proc. Natl. Acad. Sci. USA.
2008
, vol. 
29
 (pg. 
10513
-
10518
)
8
Skog
 
J.
Würdinger
 
T.
van Rijn
 
S.
Meijer
 
D.H.
Gainche
 
L.
Sena-Esteves
 
M.
Curry
 
W. T.
Carter
 
B.S.
Krichevsky
 
A.M.
Breakefield
 
X.O.
 
Glioblastoma microvesicles transport RNA and proteins that promote tumour growth and provide diagnostic biomarkers
Nat. Cell Biol.
2008
, vol. 
10
 (pg. 
1470
-
1476
)
[PubMed]
9
Chim
 
S.S.
Shing
 
T.K.
Hung
 
E.C.
Leung
 
T.Y.
Lau
 
T.K.
Chiu
 
R.W.
Lo
 
Y.M.
 
Detection and characterization of placental microRNAs in maternal plasma
Clin. Chem.
2008
, vol. 
54
 (pg. 
482
-
490
)
[PubMed]
10
Fichtlscherer
 
S.
De Rosa
 
S.
Fox
 
H.
Schwietz
 
T.
Fischer
 
A.
Liebetrau
 
C.
Weber
 
M.
Hamm
 
C.W.
Röxe
 
T.
Müller-Ardogan
 
M.
, et al 
Circulating microRNAs in patients with coronary artery disease
Circ. Res.
2010
, vol. 
107
 (pg. 
677
-
684
)
[PubMed]
11
Zhu
 
H.
Fan
 
G.C.
 
Extracellular/circulating microRNAs and their potential role in cardiovascular disease
Am. J. Cardiovasc. Dis.
2011
, vol. 
30
 (pg. 
138
-
149
)
[PubMed]
12
Vickers
 
K.C.
Palmisano
 
B.T.
Shoucri
 
B.M.
Shamburek
 
R.D.
Remaley
 
AT.
 
MicroRNAs are transported in plasma and delivered to recipient cells by high-density lipoproteins
Nat. Cell Biol.
2011
, vol. 
13
 (pg. 
423
-
433
)
[PubMed]
13
Mause
 
S.F.
Weber
 
C.
 
Microparticles: protagonists of a novel communication network for intercellular information exchange
Circ. Res.
2010
, vol. 
107
 (pg. 
1047
-
1057
)
[PubMed]
14
Hornig
 
B.
Landmesser
 
U.
Kohler
 
C.
Ahlersmann
 
D.
Spiekermann
 
S.
Christoph
 
A.
Tatge
 
H.
Drexler
 
H.
 
Comparative effect of ACE inhibition and angiotensin II type 1 receptor antagonism on bioavailability of nitric oxide in patients with coronary artery disease: role of superoxide dismutase
Circulation
2001
, vol. 
103
 (pg. 
799
-
805
)
[PubMed]
15
O’Driscoll
 
G.
Green
 
D.
Rankin
 
J.
Stanton
 
K.
Taylor
 
R.
 
Improvement in endothelial function by angiotensin converting enzyme inhibition in insulin-dependent diabetes mellitus
J. Clin. Invest.
1997
, vol. 
100
 (pg. 
678
-
684
)
[PubMed]
16
Strandberg
 
T.E.
Vanhanen
 
H.
Tikkanen
 
M.J.
 
Effect of statins on C-reactive protein in patients with coronary artery disease
Lancet
1999
, vol. 
353
 (pg. 
118
-
119
)
[PubMed]
17
Grothusen
 
C.
Bley
 
S.
Selle
 
T.
Luchtefeld
 
M.
Grote
 
K.
Tietge
 
U.J.
Drexler
 
H.
Schiffer
 
B.
 
Combined effects of HMG-CoA-reductase inhibition and renin-angiotensin system blockade on experimental atherosclerosis
Atherosclerosis
2005
, vol. 
182
 (pg. 
57
-
69
)
[PubMed]
18
Pawlowski
 
N.
Kaplan
 
G.
Hamill
 
A.L.
Cohn
 
Z.A.
Scott
 
W.A.
 
Arachidonic acid metabolism by human monocytes. Studies with platelet-depleted cultures
J. Exp. Med.
1983
, vol. 
158
 (pg. 
393
-
412
)
[PubMed]
19
Griffiths-Jones
 
S.
Saini
 
H.K.
van Dongen
 
S.
Enright
 
A.J.
 
miRBase: tools for microRNA genomics
Nucleic Acids Res.
2008
, vol. 
36
 
(Database issue)
(pg. 
D154
-
D158
)
[PubMed]
20
Livak
 
K.J.
Schmittgen
 
T.D.
 
Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method
Methods
2001
, vol. 
25
 (pg. 
402
-
408
)
[PubMed]
21
Friedewald
 
W.T.
Levy
 
R.I.
Fredrickson
 
D.S.
 
Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge
Clin. Chem.
1972
, vol. 
18
 (pg. 
499
-
502
)
[PubMed]
22
Arroyo
 
J.D.
Chevillet
 
J.R.
Kroh
 
E.M.
Ruf
 
I.K.
Pritchard
 
C.C.
Gibson
 
D.F.
Mitchell
 
P.S.
Bennett
 
C.F.
Pogosova-Agadjanyan
 
E.L.
Stirewalt
 
D.L.
, et al 
Argonaute2 complexes carry a population of circulating microRNAs independent of vesicles in human plasma
Proc. Natl. Acad. Sci. USA.
2011
, vol. 
108
 (pg. 
5003
-
5008
)
[PubMed]
23
Février
 
B.
Raposo
 
G.
 
Exosomes: endosomal-derived vesicles shipping extracellular messages
Curr. Opin. Cell Biol.
2004
, vol. 
16
 (pg. 
415
-
421
)
[PubMed]
24
Valadi
 
H.
Ekström
 
K.
Bossios
 
A.
Sjöstrand
 
M.
Lee
 
J.J.
Lötvall
 
J.O.
 
Exosome-mediated transfer of mRNAs and microRNAs is a novel mechanism of genetic exchange between cells
Nat. Cell Biol.
2007
, vol. 
9
 (pg. 
654
-
659
)
[PubMed]
25
Collino
 
F.
Deregibus
 
M.C.
Bruno
 
S.
Sterpone
 
L.
Aghemo
 
G.
Viltono
 
L.
Tetta
 
C.
Camussi
 
G.
 
Microvesicles derived from adult human bone marrow and tissue specific mesenchymal stem cells shuttle selected pattern of miRNAs
PLoS ONE
2010
, vol. 
5
 pg. 
e11803
 
[PubMed]
26
Adachi
 
T.
Nakanishi
 
M.
Otsuka
 
Y.
Nishimura
 
K.
Hirokawa
 
G.
Goto
 
Y.
Nonogi
 
H.
Iwai
 
N.
 
Plasma microRNA 499 as a biomarker of acute myocardial infarction
Clin. Chem.
2010
, vol. 
56
 (pg. 
1183
-
1185
)
[PubMed]
27
Ai
 
J.
Zhang
 
R.
Li
 
Y.
Pu
 
J.
Lu
 
Y.
Jiao
 
J.
Li
 
K.
Yu
 
B.
Li
 
Z.
Wang
 
R.
, et al 
Circulating microRNA-1 as a potential novel biomarker for acute myocardial infarction
Biochem. Biophys. Res. Commun.
2010
, vol. 
391
 (pg. 
73
-
77
)
[PubMed]
28
Zernecke
 
A.
Bidzhekov
 
K.
Noels
 
H.
Shagdarsuren
 
E.
Gan
 
L.
Denecke
 
B.
Hristov
 
M.
Köppel
 
T.
Jahantigh
 
M.N.
Lutgens
 
E.
, et al 
Delivery of microRNA-126 by apoptotic bodies induces CXCL12-dependent vascular protection
Sci. Signal.
2009
, vol. 
2
 pg. 
ra81
 
[PubMed]
29
Ohshima
 
K.
Inoue
 
K.
Fujiwara
 
A.
Hatakeyama
 
K.
Kanto
 
K.
Watanabe
 
Y.
Muramatsu
 
K.
Fukuda
 
Y.
Ogura
 
S.
Yamaguchi
 
K.
Mochizuki
 
T.
 
Let-7 microRNA family is selectively secreted into the extracellular environment via exosomes in a metastatic gastric cancer cell line
PLoS ONE
2010
, vol. 
5
 pg. 
e13247
 
[PubMed]
30
Naeem
 
A.
Zhong
 
K.
Moisá
 
S.J.
Drackley
 
J.K.
Moyes
 
K.M.
Loor
 
J.J.
 
Bioinformatics analysis of microRNA and putative target genes in bovine mammary tissue infected with Streptococcus uberis
J. Dairy Sci.
2010
, vol. 
95
 (pg. 
6397
-
6408
)
31
Fabbri
 
M.
Paone
 
A.
Calore
 
F.
Galli
 
R.
Gaudio
 
E.
Santhanam
 
R.
Lovat
 
F.
Fadda
 
P.
Mao
 
C.
Nuovo
 
G.J.
, et al 
MicroRNAs bind to toll-like receptors to induce prometastatic inflammatory response
Proc. Natl. Acad. Sci. USA.
2012
, vol. 
109
 (pg. 
E2110
-
E2116
)
[PubMed]
32
Schmieder
 
R.E.
Hilgers
 
K.F.
Schlaich
 
M.P.
Schmidt
 
B.M.
 
Renin-angiotensin system and cardiovascular risk
Lancet
2007
, vol. 
369
 (pg. 
1208
-
1219
)
[PubMed]
33
Ji
 
Y.
Liu
 
J.
Wang
 
Z.
Liu
 
N.
 
Angiotensin II induces inflammatory response partly via toll-like receptor 4-dependent signaling pathway in vascular smooth muscle cells
Cell Physiol. Biochem.
2009
, vol. 
23
 (pg. 
265
-
276
)
[PubMed]
34
Tian
 
Q.
Miyazaki
 
R.
Ichiki
 
T.
Imayama
 
I.
Inanaga
 
K.
Ohtsubo
 
H.
Yano
 
K.
Takeda
 
K.
Sunagawa
 
K.
 
Inhibition of tumor necrosis factor-alpha-induced interleukin-6 expression by telmisartan through cross-talk of peroxisome proliferator-activated receptor-gamma with nuclear factor kappa B and CCAAT/enhancer-binding protein-beta
Hypertension
2009
, vol. 
53
 (pg. 
798
-
804
)
[PubMed]