Introduction MicroRNA-21 (miR-21) has been implicated as a regulator of fibrosis. Inhibitors of miR-21 are currently undergoing clinical trials. Our aim was to explore mechanisms of how miR-21 inhibition attenuates fibrosis using a proteomics approach.

Methods and results Primary murine cardiac fibroblasts (CF) were transfected with miR-21 mimics and inhibitors. In line with previous reports, miR-21 overexpression increased CF proliferation (n=4 vs. 4; p<0.0001 for 2-way ANOVA). Unexpectedly, proteomic analysis revealed limited effects on extracellular matrix (ECM) secretion. Similarly, the ECM protein composition of hearts from miR-21 null mice was similar to wild-type controls. Thus, we searched for additional explanations as to how miR-21 might regulate fibrosis. In the community-based Bruneck study (n=671), we compared circulating miR-21 with a proteomic panel of 229 proteins associated with cardiovascular disease. Several platelet-derived pro-fibrotic factors significantly correlated with plasma miR-21 levels, including the latency-associated peptide of TGF-β1 (Pearson r 0.68; q<0.0001). Argonaute2 immunoprecipitation demonstrated enrichment of miR-21 in a human megakaryoblastic cell line. Furthermore, TGF-β1 was abundant in megakaryocytes as demonstrated by immunofluorescence staining of murine bone marrow. Platelet immunodepletion in mice (n=5 vs. 7 for control and anti-CD42b antibody injection, respectively) revealed a strong platelet dependency of plasma TGF-β1 levels (fold change: 2.37; p=0.003). Correlation of miR-21 with total TGF-β1 was confirmed in humans by ELISA measurements in platelet-poor plasma collected during the Bruneck 2015 follow-up (n=332; Pearson r: 0.63; q<0.0001). When mice were treated with antagomiR-21, no differences were observed in the platelet count or in their aggregation response to different agonists. However, a proteomics analysis of the platelet releasate revealed TGF-β1 as differentially regulated. An attenuated TGF-β1 release after antagomiR-21 treatment (n=5 vs. 5 for control and antagomiR-21) was validated in platelets activated with collagen and a thrombin receptor-activating peptide (p=0.004 and 0.002, respectively, for 2-way ANOVA). Mechanistically, Wiskott-Aldrich Syndrome protein, a negative regulator of platelet TGF-β1 secretion, was identified as a direct target of miR-21.

Conclusions This study reports a previously unrecognised effect of miR-21 inhibition on platelet TGF-β1 release. This novel systemic effect of miR-21 on platelets may contribute to the anti-fibrotic effects of miR-21 inhibitors.