Post-transcriptional regulation of cathepsin K in atherosclerosis

M. Sachse (Mannheim)1, G. Georgiopoulos (Athens)2, K. Sopova (Mannheim)3, M. Polycarpou-Schwarz (Mannheim)1, G. Mavraganis (Athens)2, S. Tual-Chalot (Newcastle Upon Tyne)4, M. Sigl (Mannheim)5, K. Stamatelopoulos (Athens)2, K. Stellos (Mannheim)6
1Medizinische Fakultät Mannheim der Universität Heidelberg Abteilung für Herz- Kreislaufforschung Mannheim, Deutschland; 2University of Athens Department of Clinical Therapeutics Athens, Griechenland; 3Universitätsklinikum Mannheim GmbH I. Medizinische Klinik Mannheim, Deutschland; 4Newcastle University Biosciences Institute Newcastle Upon Tyne, Großbritannien; 5University Medical Centre Mannheim Department of Medicine VI Mannheim, Deutschland; 6Universitätsmedizin Mannheim der Universität Heidelberg Institut für Herz-Kreislaufforschung Mannheim, Deutschland

Introduction: Adenosine-to-inosine (A-to-I) RNA editing by adenosine deaminase acting on RNA-1 (ADAR1) has emerged as a novel regulatory layer of gene expression in disease. Cathepsin K, a potent collagenase and elastase, contributes to atherosclerosis by promoting extracellular matrix degradation, plaque instability, and vascular inflammation within atherosclerotic lesions. Whether cathepsin K expression is determined by RNA editing in atherosclerosis remains unknown.
Methods: Expression levels of Cathepsin K (CTSK) and ADAR1 were quantified via qRT-PCR in peripheral blood mononuclear cells (PBMCs; n = 367) and carotid plaques (n = 35) from individuals with or without atherosclerotic cardiovascular disease (ASCVD). Carotid and femoral ultrasound imaging was used to evaluate subclinical atherosclerosis through intima-media thickness, maximum wall thickness, and plaque burden. Participants were followed prospectively for major adverse cardiovascular events (MACE). Mechanistic insights were pursued using RNA-sequencing, RNA-editing profiling, gain- and loss-of-function experiments, gene expression assays, individual crosslinking immunoprecipitation (iCLIP), and RNA immunoprecipitation (RIP) in human primary endothelial cells. 
Results: In humans, CTSK mRNA expression was independently associated with the presence of ASCVD (OR=1.84 for highest vs. lower CTSK tertiles), higher C-reactive protein >3mg/dl, increased average maximum wall thickness of the carotid arteries >1.39mm (OR=4.44 for highest vs. lower CTSK tertiles), and multiple diseased vascular beds (OR=4.09 for highest vs. lower CTSK tertiles) (P<0.05 for all). Furthermore, high CTSK concentrations were independently associated with higher MACE incidence across a median follow-up period of 48 months (OR=5.22 for highest vs. lower CTSK tertiles; P=0.008). Mechanistically, RNA-sequencing and RNA-editing studies revealed editing events in CTSK, an inflammation-stimulated extracellular matrix degradation enzyme with an established role in inflammatory diseases including atherosclerosis. CTSK is extensively edited within the Alu regions of intron 5, which is also enriched of HuR binding sites. Silencing of ADAR1 resulted in a 2-fold downregulation of CTSK mature mRNA and a 2-fold upregulation of pre-mRNA while ADAR1 overexpression exerted the opposite. Accordingly, silencing of HuR reduced CTSK expression by >2-fold. Importantly, iCLIP and RIP experiments confirmed that HuR interacts with intronic edited regions of CTSK. In the absence of RNA editing, HuR did not bind to CTSK. ADAR1 and CTSK levels were significantly upregulated in patients compared to healthy subjects. Of note, CTSK levels closely correlated with the expression of ADAR1 in cardiovascular disease patients (p < 0.001, r = 0.71) and carotid artery plaques (p<0.001, r = 0.81). 
Conclusion: Intronic Alu-mediated RNA editing facilitates HuR-dependent pre-mRNA processing, constituting a primate-specific regulatory mechanism that critically governs inflammatory gene expression in atherosclerotic cardiovascular disease.