Abstract 265: Identification of MicroRNA in an Ovine Model of Heart Failure

Abstract

Introduction: The sheep ( Ovis aries ) provides a large animal model in cardiovascular research including heart failure (HF). However, microRNA (miR) related work in the sheep model has been limited due to a paucity of information regarding oar-miR. The aim of this study was to identify novel oar-miRs in myocardium and examine their regulation in HF and HF recovery. Methods: Heart tissue was harvested from sheep undergoing 1. HF induced by rapid left ventricular (LV) pacing at 225bpm for 14 days, 2. HF recovery (HF-R) after discontinuation of pacing for an additional 14 days and 3. Sham. LV miRs were examined using next generation deep sequencing (NGS), miR array and stem-loop qPCR. Sequences were aligned with miRBase v20.0 and mapped to ovine genome and miRBase Mature BLAST search engine. Plasma was collected to assess natriuretic peptides (ANP and BNP). Results: Three miR libraries were generated from NGS and a total of 619 miRs were detected. Of these, 93 were oar-miRs; 49 novel miRs (high confidence) mapped to ovine genome and perfectly aligned to mature miRs in other organisms (miRBase v20.0); 69 putative novel miRs (high confidence) that mapped to ovine genome and aligned partially to mature miRs of other species in miRBase v20.0; 168 miRs (low confidence) that mapped to ovine genome but unaligned to any known mature miRs; and 240 miRs (low confidence) that were unmapped to the sheep genome but aligned to either miRBase mature hsa-/mmu-/rno-miRs. Plasma BNP and ANP increased 19-folds and 18- folds respectively in HF, and returned to baseline in HF-R. MiR levels in HF model were examined using miR array. About 1000 miRs were detected from array and at least 301 of them were overlapped with NGS data. Using miR array profiling followed by stem-loop qPCR validation, we found that myocardial enriched miR-133b-3p, miR-208b-3p, miR-21-5p and miR-125a-5p, -125b-5p, -126-3p, -210-3p, and 29a-3p were significantly upregulated in HF (p<0.05 vs. Sham). All trended downwards towards baseline levels during recovery (HF-R) but only miR-210 was significant (p<0.001 vs. HF). Conclusion: We identified 118 novel oar-miRs with high confidence and 408 potential oar-miRs, which is a solid foundation for miR function studies using sheep models. We also identified miR changes in HF and HF-R.