Comparison of different miR-21 inhibitor chemistries in a cardiac disease model.

We would like to comment on a recent study that analyzed the role of microRNA-21 (miR-21) in a mouse model of cardiac disease (1). Using miR-21–deficient mice and novel, very short, 8-nucleotide anti–miR21 oligonucleotides, the authors failed to detect any modulation of pressure overload-induced myocardial hypertrophy and fibrosis and concluded that miR-21 plays no role in cardiac disease. In contrast, we and others reported that inhibition of miR-21 with highly specific, 22and 15nucleotide-long anti–miR-21 oligonucleotides effectively inhibits myocardial and pulmonary fibrosis (2, 3). While genetic deletion of a target may lead to compensation during development and is often different from pharmacological inhibition of this target in the adult organism, the discrepancy between the therapeutic trials using long versus short 8-mer oligonucleotides is striking. We therefore carried out a direct head-to-head comparison of three different oligonucleotide chemistries (Figure 1A) in the same model of pressure overload-induced cardiac hypertrophy (transaortic constriction [TAC]). The two 22-mer oligonucleotides were complementary to the full-length miR-21, while the 8-mer was complementary to nucleotides 2 to 9 of miR-21, locked nucleic acid modified (LNA modified), and identical to the oligonucleotide used in the report by Patrick et al. (Figure 1A). Therapeutic efficacy of different anti–miR-21s in a mouse model of cardiac disease. Consistent with the results by Patrick et al., 8-mer anti– miR-21 led to repression of cardiac miR21 on day 2 after the last dose (Figure 1B). However, treatment with 22-mer anti–miR21s resulted in a more efficacious repression of miR-21 (by 80%). In addition, the repression was maintained only in samples treated with 22-mers but not 8-mer throughout the course of the experiment (on day 19 after the last dose, Figure 1B). It is well established that the number of phosphothioate bonds is inversely correlated to the excretion rate of oligonucleotides (4), which could partially explain the lack of efficacy of 8-mer anti–miR-21 in the cardiac fibrosis model. In addition to comparing the ability of three different anti–miR-21 oligonucleotides to repress miR-21, we also tested their ability to modify disease phenotypes of TAC. As shown in Figure 1C, interstitial fibrosis and cardiac mass were significantly increased three weeks after TAC in control mice but were strongly attenuated by treatment with both cholesteroland F/MOE-modified long 22-mer oligonucleotides. In addition, 22mers prevented the decline in cardiac function, as determined by echocardiography. In