P2709Wnt5a contributes to human atherosclerosis via novel USP17 redox signalling

Abstract

Abstract Background Wnt5a is a non-canonical Wnt ligand with potential vascular effects, but its mechanistic role in atherosclerosis progression and the underlying downstream mechanisms are poorly explored. Purpose To address the hypothesis that Wnt5a induces vascular NADPH-oxidases activity, endothelial dysfunction and detrimental downstream redox signalling which could propagate atherosclerosis in humans. Methods Study 1 included 70 patients with coronary artery disease (CAD) versus age- and sex-matched non-CAD controls. Study 2 included 1,003 CAD patients undergoing cardiac surgery; internal mammary artery (IMA) and saphenous vein (SV) segments were harvested and used for ex vivo experiments. Study 3 included 68 individuals undergoing two cardiac computed tomography scans 3–5 years apart; calcified plaque burden was assessed by coronary calcium score (CCS). Superoxide (O2·−) generation was measured by lucigenin chemiluminescence with NADPH 100μM stimulation as indicator of NADPH-oxidases activity. Activation of Rac1, a key NADPH-oxidases subunit, was evaluated by a commercially available kit. Primary vascular smooth muscle cells (VSMCs) and HeLa cells were used for in vitro experiments. Circulating Wnt5a and Sfrp5 (a Wnt5a antagonist) were measured by ELISA in fasting plasma samples. Results In Study 1, the presence of CAD was independently linked with increased circulating Wnt5a bioavailability (A), which was, in turn, associated with increased IMA NADPH-oxidases activity in Study 2 (B). Recombinant Wnt5a directly stimulated NADPH-oxidases activity (C) via Rac1 activation (not shown) in human IMA, while inducing endothelial dysfunction evidenced by impaired SV endothelium-dependent acetylcholine (Ach) vasorelaxations (D). Transcriptomic analysis in Wnt5a-treated primary VSMCs versus controls identified USP17, a deubiquitinating enzyme implicated in Rac1 activation, as the top differentially regulated hit (not shown). Indeed, Wnt5a stimulated USP17 upregulation in VSMCs which was reversed by PEGylated superoxide dismutase (peg-SOD) 300U/mL (E), suggesting a redox sensitive effect. USP17 knockdown abolished the ability of Wnt5a to induce Rac1 activation in HeLa cells (F). At a clinical level, plasma Wnt5a was a predictor of plaque progression (defined as ΔCCS≥1, G) and new onset calcification (H) in Study 3. Conclusions We demonstrate for the first time that Wnt5a is elevated in CAD and causally associated with increased vascular oxidative stress and endothelial dysfunction in humans. We further reveal USP17 to be a novel, previously undescribed, link between Wnt5a, Rac1 activation and NADPH-oxidase activity induction in humans. We finally propose that circulating Wnt5a may have a clinically relevant role in predicting atherosclerosis progression. Our findings identify Wnt5a as rational therapeutic target in vascular disease. Acknowledgement/Funding British Heart Foundation; Alexandros S Onassis Public Benefit Foundation