OUP user menu

Micromanaging restenosis by therapeutic inhibition of miR-92a

Mario Leonardo Squadrito, Michele De Palma
DOI: http://dx.doi.org/10.1093/cvr/cvu178 432-434 First published online: 31 July 2014

This editorial refers to ‘Inhibition of miR-92a improves re-endothelialization and prevents neointima formation following vascular injury’ by J.-M. Daniel et al., pp. 564–572, this issue.

The adverse consequences of procedure-associated endovascular injury may limit the long-term benefits of coronary interventions like angioplasty and stent insertion. Damage to the artery's endothelium and impaired re-endothelialization, also caused by the use of non-selective anti-proliferative drugs, may promote the development of restenosis—a pathological condition characterized by smooth muscle cell (SMC) hyperplasia, vessel thickening and lumen narrowing, and impaired blood flow—in a significant proportion of the patients. Reducing the occurrence of restenosis after coronary interventions thus remains an important medical need.1 In this issue of Cardiovascular Research, Daniel et al.2 report that both genetic and molecular (therapeutic) approaches to inhibit microRNA (miRNA)-92a facilitate arterial re-endothelialization and prevent restenosis in a mouse model of femoral artery injury.

miRNAs are a class of small non-coding RNAs that fine-tune gene expression at the post-transcriptional level. Several miRNAs, including miR-126, miR-132, miR-222, and miR-92a, have been implicated in the regulation of endothelial cell (EC) biology.3 miR-92a is a member of the miR-17-92 cluster, which encodes six distinct miRNAs broadly involved in physiological and pathological processes such as cell proliferation, development, immunity, and tumorigenesis.4 While being deregulated in several leukaemias and solid tumours, miR-92a also functions as a negative regulator of EC proliferation, angiogenesis, and vascular repair. Therapeutic modulation of miR-92a activity in ECs may, therefore, rescue the damaged endothelium after coronary interventions.58

Daniel et al.2 employed a mouse model of wire-induced injury of the femoral artery to analyse the temporal and cellular expression of miR-92a after vascular damage. They found that miR-92a levels increased post-injury and peaked at Day 10, a time-point when SMC hyperplasia was already evident in the injured artery. The analysis of cultured ECs and SMCs, as well as intact or endothelium-denudated arteries, suggested that ECs and not SMCs were the main source of miR-92a in the injured arteries. Moreover, transfection of miR-92a inhibited vascular endothelial growth factor-A (VEGFA)-induced EC proliferation and migration, but did not affect platelet-derived growth factor-BB (PDGFB)-induced SMC proliferation or migration, indicating that the functions of miR-92a are largely EC-autonomous. The authors then used two loss-of-function strategies to attenuate miR-92a activity in the damaged arteries. Both the systemic delivery of locked nucleic acid (LNA)-modified anti-miR-92a oligonucleotides9 and the conditional knockout of miR-92a in TIE2+ ECs stimulated re-endothelialization and decreased SMC hyperplasia and inflammatory–macrophage infiltration in the femoral artery after wire-induced injury (Figure 1). These data suggest that suppression of endothelial miR-92a activity promotes arterial re-endothelialization and limits SMC hyperplasia, at least in part, through direct pro-proliferative effects on ECs.2 Consistent with these findings, previous studies showed that inhibition of miR-92a enhances VEGFA-induced EC proliferation by activating mitogenic ERK and JNK signalling.6

Figure 1

Inhibition of restenosis by therapeutic or genetic miR-92a targeting. Endothelial damage consequent to coronary interventions (A) up-regulates the expression of miR-92a, which directly targets the pro-angiogenic and anti-inflammatory factors Sirt1, Itga5, and Klf2/4 (B), inhibits EC proliferation, and promotes the development of restenosis (C). Systemic treatment with anti-miR-92a oligonucleotides (D), or the genetic deletion of miR-92a in ECs (E), restores Sirt1, Itga5, and Klf2/4 levels in the healing endothelium, stimulates re-endothelialization, decreases SMC hyperplasia (via increased eNOS activity and NO production), and ultimately limits the development of restenosis (F). Please see text for references.

Besides direct pro-proliferative effects on ECs, inhibition of miR-92a may attenuate experimental restenosis through additional mechanisms. Among the validated targets of miR-92a are the deacetylase sirtuin-1 (Sirt1) and integrin-α5 (Itga5). SIRT1 is highly expressed in the angiogenic vasculature and promotes sprouting angiogenesis, whereas ITGA5 enables migration, pro-angiogenic signalling, and angiogenesis of ECs by modulating their interactions with the extra-cellular matrix.10 Daniel et al.2 observed increased expression of both SIRT1 and ITGA5 in the arterial ECs of anti-miR-92a–treated mice 2 weeks post-injury, suggesting that therapeutic inhibition of miR-92a stimulates re-endothelialization, at least in part, by de-repressing both pro-angiogenic factors. miR-92a also targets the transcription factors Krüppel-like factor-2 (KLF2) and 4,11 which confer anti-inflammatory and atheroprotective properties to the endothelium. De-repressed KLF2 and 4 may operate to down-regulate the expression of leucocyte adhesion molecules on ECs, hence limiting inflammatory cell infiltration, and to enhance endothelial nitric oxide (NO) synthetase (NOS3/eNOS) activity, which inhibits SMC proliferation through NO production.6,11,12 Thus, therapeutic inhibition of miR-92a may initiate an anti-atherosclerotic programme in ECs that limits inflammatory cell infiltration and SMC proliferation in the healing arteries.58

Daniel et al.2 and previous studies58 employed anti-miR-92a oligonucleotides delivered systemically in animal models of vascular injury. Because systemic anti-miR oligonucleotides target multiple organs and cell types,9,13 this approach may have altered miR-92a activity also in non-ECs. Furthermore, anti-miR-92a oligonucleotides may potentially target several mature miRNAs. Indeed, miR-92a, miR-92b, and miR-25 belong to the same miRNA family and thus share the seed-sequence used to design anti-miR oligonucleotides. Moreover, two mature miR-92a sequences exist that are expressed from two distinct genetic loci, miR-17-92 (encoding miR-92a-1) and miR-106-363 (encoding miR-92a-2). Based on the above, non-EC autonomous effects and potential co-targeting of distinct miRNA species could not be formally excluded in previous studies.58 To address these issues, Daniel et al.2 employed conditional knockout strategies either targeting the miR-92a-1 gene broadly in haematopoietic cells or specifically in TIE2-lineage cells, which comprise both ECs and haematopoietic cells.14 By this comparative analysis, the authors unequivocally showed that the miR-92a sequence encoded by the miR-17-92 locus is functionally important in the endothelial, but not in the haematopoietic, lineage.

Although not yet broadly tested in the clinic, anti-miR therapeutics are being successfully used in animal models to experimentally suppress miR activity.9,13 For example, systemic anti-miR-33 therapy holds promise for the treatment of dyslipidaemias and associated vascular/cardiac diseases, as shown by preclinical studies in animal models.13 Remarkably, a phase II clinical study recently demonstrated the efficacy of a LNA-modified anti-miR-122 for the inhibition of hepatitis C virus (HCV) replication in patients.15 It should be noted that both the pharmacokinetic properties and the intravenous route of administration of anti-miRs facilitate targeting of the liver, which may explain the reported success of anti-miR-122/33–based treatments in animal models and patients.13 On the other hand, it is currently unclear whether systemic administration of anti-miR-92a therapeutics would efficiently target the arterial endothelium of patients undergoing coronary interventions. Also, potential toxicities associated with the systemic down-regulation of miR-92a, particularly in non-ECs, should be considered. Although miR-92a knockout mice are viable and fertile, they show increased embryonic lethality as well as growth and skeletal defects.16 To alleviate these concerns, drug-eluting stents1 that deliver anti-miR-92a locally to the healing endothelium should be tested in large animal models to explore the advantages and disadvantages of this approach compared with systemic administration.

Conflict of interest: none declared.


Work in the authors' laboratory is, in part, funded by the European Research Council (ERC TIE2+ monocytes) and Fonds National Suisse de la Recherche Scientifique (SNSF grant 31003A-143978).


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