P5398CircRNAs deregulation in heart failure patients

Abstract

Abstract Background Circular RNAs (circRNAs) are an emerging class of noncoding RNAs stemming from the splicing and circularization of pre-mRNAs exons. CircRNAs can regulate transcription and splicing, sequester microRNAs acting as “sponge” and inducing the respective targets, and bind to RNA binding proteins. Recently, they have been found deregulated in dilated cardiomyopathies (DCM), one of the cardiovascular diseases with the worst rate of morbidity and mortality, and whose molecular mechanisms are only partially known. Purpose Therein, we will evaluate in ischemic DCM patients the modulation of 17 circRNAs, 14 out of them obtained from literature data on DCM ischemic or not, while the other 3 were circRNAs not characterized in the heart previously. The study aims to identify circRNAs candidates for further functional characterization in DCM. In addition, as differential expression (DE) analysis is not easily performed for circRNAs in RNA-seq datasets, the validated circRNAs will be used to set up the most specific and sensitive bioinformatics pipeline for circRNA-DE analysis. Methods We designed divergent and convergent specific primers for 17 circRNAs and their host gene, respectively, and their amplification efficiency was measured by RT-qPCR. Transcripts expression was measured in left ventricle biopsies of 12 patients affected by non end-stage ischemic HF and of 12 matched controls. Results We identified cPVT1, cANKRD17, cBPTF as DE, and validated the modulation of 5 out of the 14 DCM-related circRNAs (cHIPK3, cALPK2, cPCMTD1, cNEBL, cSLC8A1), while cPDRM5, cTTN1 showed opposite modulation, which may be due to the specific disease condition. All of them were modulated differently from the respective host gene. CircRNA/miRNA interactions were predicted using Starbase 3.0. Next, mRNAs-targets of the identified miRNAs were predicted by mirDIP 4.1 and intersected with gene expression datasets of the same patients, previously obtained by microarray analysis. We found that cBPTF and cANKRD17 might sponge 12 and 2 miRNAs, respectively. Enrichment analysis of the relevant targets identified several important pathways implicated in DCM, such as MAPK, FoxO, EGFR, VEGF and Insulin/IGF pathways. In addition, deep RNA-Seq analysis that is currently ongoing and the validated circRNAs will be used to optimize the bioinformatics pipeline for circRNA DE analysis. Conclusions We identified a subset of circRNAs deregulated in ischemic HF potentially implicated in HF pathogenesis.