Acute coronary syndrome (ACS) remains a major cause of death and disability worldwide and the number of people with cardiovascular risk factors is increasing, owing to obesity, diabetes, inactivity and ageing [1
]. Currently, no high-quality soluble biomarkers are available to accurately identify subjects who are at risk of developing acute manifestations of cardiovascular diseases. Hence, there is a crucial need to discover novel biomarkers that will improve risk prediction models of ACS and help with early identification of subjects eligible for preventive treatment, as well as better stratification of high-risk patients.
MicroRNA (miRNAs) are a class of small (~22 nucleotides) non-coding RNA, which are essential post-transcriptional modulators of gene expression by binding to the 3’ untranslated region of specific target genes, thereby leading to suppression or translational repression [2
]. Over the past decade, miRNAs have emerged as important regulators and fine-tuners of both physiological and pathological cellular processes, including those relevant to the cardiovascular system [3
]. The majority of miRNAs are intracellular; however, miRNAs can also be secreted as microvesicles or exosomes and apoptotic bodies into the blood circulation. MiRNAs remain stable in the blood or serum and membrane-derived vesicles or lipoproteins can carry and transport circulating miRNAs. Indeed, miRNAs isolated from plasma are highly stable in boiling water, prolonged room temperature incubation or repeated freeze-thawing [5
]. Circulating miRNAs are protected from plasma ribonucleases by their carriers, including lipid vesicles or protein conjugates, such as Argonaute 2 or other ribonucleoproteins [6
Several miRNAs were shown to take part in the pathogenesis of cardiac injury and atherosclerosis [7
] and we recently identified in stable coronary artery disease (CAD) patients a specific expression profile of circulating miRNAs as a suitable biomarker for accurate diagnosis of CAD [7
]. In considering miRNAs are reliable biomarkers of coronary disease, the aim of the present study was to investigate the expression of plasma-circulating miRNAs between high to very high-risk asymptomatic individuals according to SCORE (Systematic Coronary Risk Evaluation) model [9
] and patients who experienced an acute coronary event. We explored the association between circulating miRNAs and the risk of suffering an ACS, as well as the incremental value of a new miRNA-based signature to a clinical predictive risk model.
Circulating miRNAs are currently explored as companion biomarkers of cardiac injury, with potentially high diagnostic and prognostic powers. In the present study, we investigated the expression level of plasma-circulating miRNAs in a population of ACS patients compared to asymptomatic subjects at high to very high cardiovascular risk. We identified a new miRNA-based signature associated with the occurrence of ACS in a clinically relevant manner. The best combination of the selected miRNAs providing an additive value to the clinical model included: miR-122, miR-150, miR-195 and miR-16. This new model significantly improved the accuracy of ACS risk prediction with an increased AUC from 0.882 to 0.924, which is highly powerful when considering the clinical need of new biomarkers for identifying individuals at risk of ACS.
Based on a microarray profile of 758 miRNAs and the identification of 21 candidate miRNAs from the derivation cohort, we found significantly increased levels of 7 miRNAs in patients with ACS compared to non-ACS subjects. These data confirm numerous studies reporting deregulated levels of circulating miRNAs in patients with cardiovascular diseases [14
]. Interestingly, only five miRNAs remained associated with the risk of ACS, independently of other cardiovascular risk factors. Our observations corroborate previous data revealing that differential expressions of these five miRNAs are involved in pathophysiological mechanisms related to cardiovascular outcomes. Moreover, based on ROC analysis, we retained a combination of four miRNAs, including miR-122, miR-150, miR-195 and miR-16, which provided the most powerful incremental value to the clinical model. MiR-122 is the most abundant miRNA in liver and is closely related to lipid metabolism. In human studies, miR-122 was demonstrated to be remarkably up-regulated in hyperlipidemic patients and associated with the severity of coronary heart disease [17
]. MiR-122 increase was found in the cardiac tissue of patients who died from acute MI (AMI) [18
] and strongly correlated with long-term morbidity and mortality in ST-segment-elevation acute myocardial infarction (STEMI) patients treated with primary angioplasty [19
]. MiR-122 is likely involved in apoptotic and cell survival pathways since the overexpression of miR-122 in H9C2 cardiomyocytes resulted in increased caspase-3 activity and downregulation of the PTEN/PI3K/AKT pathway [21
MiR-150 was associated with inflammation and was reported to be implicated in the pathogenesis of various cardiovascular diseases; however, the expression profile of circulating miR-150 in cardiac injury could vary among different studies. Indeed, Devaux et al. observed that low circulating levels of miR-150 are associated with adverse left ventricular remodeling after first STEMI [23
], while circulating miR-150 was overexpressed in both STEMI and NSTEMI patients and in the early stage of AMI patients compared to control subjects [24
]. Interestingly, studies indicated that miR-150 exhibited a cardio-protective role by inhibiting monocyte accumulation and cell death in a mouse model of acute myocardial infarction [25
]. Indeed, the overexpression of miR-150 may directly repress the pro-apoptotic gene egr2 and p2x7r (pro-inflammatory ATP receptor) or the glucose-regulated protein-94 expression in human cardiomyocytes [26
MiR-16, a member of the miR-15 family, negatively regulates cellular growth and cell cycle progression [28
]. The expression of miR-16 is high in cardiomyocytes [29
] and in vivo its overexpression worsened cardiac injury in a rat model of AMI [30
]. On the contrary, the knockdown of miR-16 in AMI rats was cardio-protective, as evidenced by reduced cardiac apoptosis and infarct size, as well as improved cardiac function [30
]. Interestingly, the β2-adrenergic receptor was identified as a target of miR-16, which may partially explain the mechanisms of miR-16’s effects on cardiac injury. Circulating miR-16 was significantly up-regulated in Takotsubo cardiomyopathic patients compared with healthy controls [31
] and was positively associated with the development of left ventricular dysfunction after AMI [32
Increased expression of miR-195, described in cardiac hypertrophy and cardiac overexpression of miR-195, resulted in heart failure in transgenic mice [33
]. In addition, an inverse correlation of circulating miR-195 was observed in the presence of abdominal aortic aneurysms and increased aortic diameter, which is in line with the targets of miR-195, including collagens, proteoglycans and elastin [34
]. A small simple size study of 18 patients showed that circulating miR-195 expressions (evaluated in combination with miR-30 and let-7b) were up-regulated at 8 h and 12 h after the onset of AMI symptoms and correlated with the plasma concentrations of cardiac troponin I [35
]. Interestingly, in animal models of ischemia reperfusion injury [36
], as well as in cardiomyocytes subjected to hypoxia-reoxygenation [37
], the up-regulation of miR-195 enhanced cardiomyocyte apoptosis, suggesting that miR-195 might be a potential candidate marker and therapeutic target of AMI.
The strength and the originality of our study are based on the comparison of two well-characterized very high-risk populations that report on primary and secondary cardiovascular risk (CVR) prevention. Importantly, no interaction was found between the level of the selected miRNAs and medical treatments, thus supporting the significance of these miRNAs in ACS. Moreover, considering that miRNAs are active molecules, our study also suggests that miR-122, miR-150, miR-195 and miR-16 could represent new targets for therapeutic interventions. Additional studies are needed to determine whether differential expression of circulating miRNAs is a result of modulated intracellular production and/or different release mechanisms.
Some limitations in our study have to be taken into account. Although the sample size had enough statistical power to detect significant differences for all analyses, including a second independent cohort for validation would be useful. In addition, it would have been valuable to measure the baseline expression of the circulating miRNAs before the ACS event; however, such information required a longitudinal cohort study at high cardiovascular risk with at least 10-year follow-up.
In the perspective of our work, the monitoring of the miRNA signature both in ACS patients (secondary prevention) and in high-risk asymptomatic subjects (primary prevention) would help to determine whether the signature is a suitable companion biomarker for ACS prediction.
In conclusion, our study identified a powerful signature of circulating miRNAs providing an incremental value to traditional risk markers for ACS.