The exercise-induced myokine irisin acts as an endogenous anti-thrombotic agent by modulating multiple target cells

Abstract Background and Objective Exercise is known to reduce thrombotic risk and systemic inflammation, but the underlying mechanisms are not understood. One potential mechanism involves the exercise-induced myokine irisin, which is cleaved from fibronectin type III domain-containing protein 5 (FNDC5) during muscle contraction and has well-known metabolic benefits. Irisin's amino acid sequence is fully conserved across mammals, suggesting evolutionarily preserved function. The possible cardiovascular effects of irisin remain largely unexplored as irisin's receptors beyond the skeletal αVβ5 are unknown. In this study, we aimed to investigate the anti-thrombotic effects of irisin and identify its cellular targets and mechanisms of action. Methods Whole blood, platelet-rich plasma (PRP), washed platelets, and monocytes were obtained from healthy donors (n=25). Recombinant irisin was purified from transfected HEK293 cells and fluorescently labeled. Binding studies were performed on platelets, monocytes, and human umbilical vein endothelial cells (HUVECs) with flow cytometry under various conditions, including Mn²⁺ supplementation and temperature variations. Crosslinking was performed with DTSSP and proteomics with LC-MS/MS. Functional assays included measurement of platelet aggregation by light transmission aggregometry, platelet spreading by F-actin imaging, clot contraction by serial imaging, and expression of activation markers (CD62P and PAC-1 binding) after stimulation with collagen, convulxin, thrombin, arachidonic acid, or ADP, monocyte activation marker CD64, and PMA-induced von Willebrand factor (VWF) secretion in endothelialized microchannels. Thrombus formation in vivo was induced by laser-mediated injury of cremaster arterioles and monitored by intravital microscopy. Results Irisin demonstrated differential binding affinity across cell types, with endothelial cells showing the highest affinity, followed by monocytes (Kd ~30 nM, comparable to αVβ5 integrin), then platelets. Overall, irisin binding was enhanced by integrin activation (induced by Mn²⁺), cell activation, or warmer temperatures. In platelets, irisin binding was competitively inhibited by fibrinogen (p<0.0001) or PAC-1 (p<0.0001), while crosslinking and proteomics indicated αIIbβ3 integrin as the receptor, suggesting that irisin preferentially binds an active, open conformation of αIIbβ3. In platelets, irisin binding decreased CD62P exposure (p=0.04) especially at lower agonist concentrations, inhibited collagen- or convulxin-induced aggregation, and reduced thrombin-induced platelet spreading on fibrinogen-coated surface (p=0.03) without affecting overall clot contraction. For collagen, irisin inhibited aggregation dose-dependently: 250 nM irisin 54.7±19.3% vs control (p=0.12), 500 nM 35.9±24.2% (p=0.004), and 1µM 13.4±14.9% (p=0.0001). We obtained similar results for convulxin, but not with any other agonists tested. In whole blood, irisin inhibited shear-induced thrombus formation on collagen surfaces measured by T-TAS (50±16% vs control, p=0.32, for 1µM irisin and 6±6%, p=0.005, 2µM). Together, irisin inhibition of αIIbβ3seems to require a longer lag time of inside-out agonist activation as provided by the GPVI agonists. In monocytes, irisin binding was enhanced by PMA-induced activation, resulted in a slight decrease in activation marker CD64, and was abolished by pretreatment with CBRM1/5 (binds active αM), suggesting that αMβ2 is the receptor. In endothelial cells, irisin bound more robustly to thrombin- and PMA-activated HUVECs, and reduced PMA-induced VWF secretion by 48±11.8% (p=0.004, 0.5µM). Finally, our preliminary study in wild-type mice showed that irisin pretreatment decreased platelet recruitment and unstable thrombus formation after laser-induced arteriolar injury. Conclusions: Irisin functions as a multi-target anti-thrombotic myokine that concertedly modulates platelet, monocyte, and endothelial cell function through distinct integrin receptors. By binding to activated or partially active αIIbβ3 on platelets, putatively αMβ2 on monocytes, and likely an αV integrin on endothelial cells, irisin either blocks the intermediate conformations of the integrins or competes with physiological pro-thrombotic ligands resulting in both anti-thrombotic and anti-inflammatory effects. This coordinated multi-cellular modulation positions irisin as an endogenous, exercise-inducible therapy for thrombotic disorders.