OUP user menu

Circulating microRNAs: novel biomarkers for cardiovascular diseases?

Stefanie Dimmeler, Andreas M. Zeiher
DOI: http://dx.doi.org/10.1093/eurheartj/ehq221 2705-2707 First published online: 6 July 2010

This editorial refers to ‘Circulating microRNAs are new and sensitive biomarkers of myocardial infarction?, by Y. D'Alessandra et al., on page 2765

MicroRNAs (miRNAs) comprise short, non-coding RNAs, which are important for many aspects of homeostasis and disease.1,2 MiRNAs are generally considered to act as intracellular endogenous RNAs to control gene expression on a post-translational level. However, recent studies demonstrated that miRNA can be detected in circulating blood and that these circulating miRNAs might be useful disease biomarkers, e.g. for certain forms of cancer.3,4

D'Alessandra and colleagues have now demonstrated that circulating muscle-derived miRNAs might be useful as biomarkers of acute myocardial infarction (AMI)5 (Figure 1A). Specifically, the authors showed that muscle-enriched miRNAs, such as miR-1, miR-133a, miR-133b, and miR-499-5p, are elevated in plasma obtained from mice after coronary artery ligation as well as in humans with AMI.5 In line with these findings, other groups recently reported similar results.6,7 In experimental AMI models and in patients, Wang et al. showed that the muscle-enriched miR-1, miR-133a and miR-499 as well as the cardiac-specific miR-208 are elevated in plasma.6 Likewise, miR-499,7 miR-1,8,9 miR-133a/b, and miR-2089 were shown to be increased in small cohorts of humans after AMI. All of these studies suggest that circulating myocardial-derived miRNAs might be useful as potential biomarkers for infarction.

Figure 1

(A) Release of cardiac- and muscle-enriched miRNAs after AMI. (B) Possible mechanisms of miRNA release.

miRNAs as biomarkers for AMI: specificity, sensitivity, and practicability?

Although these recent findings are highly interesting, establishing circulating miRNAs as biomarkers for AMI is challenging and requires that several criteria are fulfilled. First of all, the detected miRNA should rather specifically identify myocardial injury. It is well known that some miRNAs are expressed in a cell type- and tissue-specific manner, and, indeed, one of the miRNAs, namely miR-208a, is encoded by the α-myosin heavy chain gene and therefore is exclusively expressed in cardiac myocytes.10 Consistently, the level of miR-208 in blood from healthy volunteers or patients without an acute ischaemic event is very low, whereas circulating levels of miR-208 were significantly elevated after induction of AMI.6 However, D'Alessandra and co-workers were unable to detect elevated levels of circulating miR-208 in AMI patients, probably because the levels are below the detection limit of the protocol used to detect circulating miRNAs in this study. The other miRNAs, miR-1, miR-133a/b, and miR-499, that were proposed as potential biomarker for AMI are not exclusively expressed in cardiac myocytes, but are also detected in skeletal muscle cells.1 Therefore, one may expect that injury to the skeletal muscle can lead to an increase in the levels of these miRNAs as well. However, in a model of hind limb ischaemia, that induces significant muscle injury, D'Alessandra et al. did not see any increase of miR-1, miR-133a/b, and miR-499. It is unclear how these results can be interpreted, since other studies indeed detected elevated circulating levels of miR-1 and miR-133a/b, but not miR-208, in sham-operated mice, indicating that tissue injury itself can lead to increased levels of muscle-enriched miRNAs. Consistently, in a side-by-side comparison of different miRNAs, the cardiac-specific miRNA miR-208 showed a superior receiver operating characteristic curve, similar to the well-established marker cardiac troponin I, indicating a high sensitivity and specificity.6 These data need to be confirmed in a larger scale prospective study.

In addition to a very high specificity and sensitivity, a clinically useful biomarker for AMI should be measurable early after onset of myocardial injury, and the measurements should be rapid and robust. Two of the studies addressed the kinetics of circulating miRNAs compared with established markers of AMI such as troponin. Muscle-derived miRNAs were shown to be elevated within 1 h after induction of coronary artery ligation in the mouse model. In humans, miR-1 and miR-133a/b were elevated as early as 156 min after onset of symptoms and declined thereafter, whereas miR-499 further increased, achieving maximal levels ∼9 h after symptom onset. More importantly, RNA isolation from plasma and subsequent quantification by real-time PCR as done in the published studies is time consuming, thereby challenging the measurement of circulating miRNAs for use as a rapid bedside test.

How are miRNAs released?

The mechanism by which circulating miRNAs are released into the circulation is unclear. However, the findings that miRNAs are detected in serum or plasma in a remarkably stable form, that can withstand repetitive freezing and thawing cycles,3,11 indicates that circulating miRNAs are protected from RNase-dependent degradation. Possible mechanisms mediating this protection include the storage of miRNAs in microvesicles or the formation of protein–miRNA complexes (Figure 1B). Indeed, increasing evidence suggests that miRNAs are actively secreted in microvesicles or exosomes,4,12 and a recent study proposes that a ceramide-dependent pathway controls for intercellular transfer of microRNAs via exosomes.13 Others showed that apopotic bodies devoid of endothelial cells contain endothelial miRNAs.14 Which of the proposed mechanisms accounts for the release of miRNAs during AMI remains to be elucidated. It might be interesting to investigate whether levels of circulating miRNAs correlate with circulating microparticles, which are well known to be increased after AMI.

Do circulating miRNAs have a biological function?

Besides using circulating miRNAs as biomarkers, one may speculate that circulating miRNAs control gene expression in an inter cellular manner. In principle, extracellular RNA can be taken up by cells, as shown for RNA which had been incorporated into microvesicles.4 Moreover, injected miRNAs were shown to be transported by endothelial apoptotic bodies into atherosclerotic lesions, where they control downstream targets.14 However, the concentration of circulating miRNAs detected in human plasma is rather low (in general <10 pM, S. Dimmeler, unpublished results) and an active uptake mechanism needs to be considered allowing these low concentrations to produce a biological response. Therefore, it remains to be determined whether a modulation of circulating miRNA translates into alteration of systemic gene expression.

In conclusion, the discovery of miRNAs circulating with the blood opens up intriguing possibilities for the use of the circulating miRNA patterns as a biomarker for cardiovascular diseases. This is not only of potential relevance for the identification of patients with AMI, where these markers need to compete with established and highly sensitive markers such as high sensitivity troponins, but might also be very useful for risk stratification, e.g. of patients with heart failure. In such a patient cohort, miR-423 was recently identified as a potentially promising marker.15

Acknowledgement

S.D. is supported by an ERC advanced grant (Angiomirs) and the DFG (Exc 147-1 and SFB834).

Conflict of interest: none declared.

Footnotes

  • The opinions expressed in this article are not necessarily those of the Editors of the European Heart Journal or of the European Society of Cardiology.

  • doi:10.1093/eurheartj/ehq167

References