Development of an enzyme-linked immunosorbent assay for versikine detection and first analyses in human cardiovascular disease

Background Extracellular matrix (ECM) remodelling is central to the pathogenesis of cardiovascular disease (CVD), yet clinically feasible tools to quantify ECM turnover remain limited. Versican, a central component for ECM remodelling after injury and inflammation, is cleaved by ADAMTS enzymes and ADAMTS enzyme activity is linked to pathological remodelling. Cleavage generates the fragment versikine, which may serve as a biomarker of pathological remodelling. Circulating versikine levels in human serum have not been characterized in CVD. Purpose We aimed to develop and validate a sandwich ELISA for quantifying versikine in human serum to determine its circulating levels in healthy individuals and in patients with CVD. Further, we examined the correlation between versikine and outcomes and clinical characteristics in the two patient cohorts. In one of the cohorts, we also examined correlation between versikine and markers of fibrosis. Methods A sandwich ELISA was developed using a neo-epitope capture antibody and a detection antibody targeting the G1 domain of versican. Analytical validation included assessment of repeatability, reproducibility, measuring range, linearity and spike and recovery. The assay was applied to three human cohorts: healthy individuals, patients with ST-elevation myocardial infarction (STEMI) and patients with hypertrophic cardiomyopathy (HCM). In the HCM cohort, in addition to clinical characteristics, versikine levels were compared with levels of known markers of fibrosis, including N-terminal propeptide of procollagen I and III (PINP and PIIINP) and C-terminal telopeptide of type I collagen (CITP). Results The assay demonstrated acceptable results on linearity, repeatability (<10%), reproducibility (<15%) and recovery (<±20%). The quantitative range was 0.2-166.7 ng/ml, and all measured serum samples fell within this range. Median (IQR) levels of versikine were lower in patients with HCM (15.2 (8.2–24.4) ng/ml) and STEMI (13.1 (9.6–19.4) at Day 1, 13.1 (9.9–19.2) ng/ml after 4 months), than in healthy individuals (24 (16.6–27.7)). Versikine demonstrated only weak associations with clinical variables, and no correlations with PINP, PIIINP and CITP. Conclusions We have developed and analytically validated a sandwich ELISA that quantify levels of versikine in human serum. In this initial clinical evaluation, versikine levels were lower in patients with CVD compared with healthy controls and were not associated with markers of collagen turnover. Further studies are warranted to clarify its biological and clinical relevance.