Rationale Non-coding RNAs (ncRNAs), including microRNAs (miRNAs), circular RNAs (circRNA) and long non-coding RNAs (lncRNA), have been implicated as novel cardiac biomarkers.

Objective To compare the tissue-specificity and release kinetics of ncRNAs and protein biomarkers after induced myocardial injury by transcoronary ablation of septal hypertrophy (TASH).

Methods and results Upon screening the relative abundance of 109 circRNA and 21 lncRNAs in human cardiac tissue, 12 circRNAs and 11 lncRNAs were selected for further analyses. Human myocardial tissue was spiked into plasma from healthy individuals and the expression levels of ncRNAs were compared to a panel of miRNAs, including muscle- (miR-1, miR-133a) and cardiac-enriched miRNAs (miR-208a, miR-208b, miR-499). Curve fitting analyses of each of the three ncRNA classes with the highest R2 values revealed no significant differences in the regression coefficients compared with high sensitive troponin T and I (hs-cTnT, hs-cTnI) and cardiac myosin-binding protein C (cMyC). At low spike-in concentrations, however, significantly higher regression coefficients were observed for all ncRNA species (Mann Whitney test: miRNAs vs. proteins p<0.0001, fold-change 2.6; circRNAs vs proteins p=0.0028, fold change 2.8; lncRNAs vs. proteins p=0.0028, fold-change 1.6). To assess whether ncRNAs allow earlier detection of myocardial injury, circulating ncRNAs were quantified in patients undergoing TASH before and at 1 hour, 8 hours, 24 hours after the procedure. Unlike cardiac proteins, miR-1 and miR-133a showed the steepest rise within the first hour after cardiac injury. These muscle-enriched miRNAs were also more readily detectable in plasma than the cardiac-specific miRNAs, miR-208b and miR-499. In contrast, cardiac circRNAs remained undetectable in plasma even after myocardial injury. Putative cardiac lncRNAs, including the Long Intergenic Non-Coding RNA Predicting Cardiac Remodelling And Survival (LIPCAR) were abundant in plasma but failed to show significant changes after TASH, refuting a predominant cardiac origin. Finally, the stability of ncRNAs was assessed in plasma. Degradation of cardiac/muscle-enriched miRNAs was observed in plasma left at room temperature for 1 hour and reduced by treatment with RNAse inhibitors. As expected, circRNAs were less susceptible to degradation. Mitochondrial-derived lncRNAs such as LIPCAR increased after 1 hour of incubation in plasma, while non-mitochondrial lncRNAs showed a degradation pattern similar to miRNAs.

Conclusions Our results demonstrate that circulating heart-associated ncRNAs may enhance early detection of myocardial injury. All three ncRNA classes demonstrated superior release kinetics in vitro compared with established cardiac protein biomarkers. At early time points after TASH, however, a higher sensitivity was only observed for muscle-enriched miRNAs, but not for circRNAs or lncRNAs.