Abstract

Clinical and experimental congestive heart failure (CHF) is characterized by a program of regulated messenger RNA (mRNA) gene transcripts. Much is known about transcriptional control of cardiac gene expression in CHF, but less is known about the alternate mechanism of mRNA stability and translation by micro RNAs (miR). Specifically lacking is information on possible dynamic regulation of miRs during CHF treatment, and the effect on target mRNAs. Methods : We performed comprehensive microarray analyses of both mRNA transcript and miR levels in myocardial specimens from CHF without (n=16) or after (n=10) mechanical unloading with left ventricular assist devices (LVAD), in comparison to 7 non-failing hearts. Individual myocardial RNA samples were hybridized to NCode v2 arrays (Invitrogen, 467 human miRs) and Affymetrix HuEx v1.0 arrays (>1 million exon clusters representing 290,000 full-length mRNAs). miR data were analyzed by a dye swap, loop design algorithm (NCode Profiler) after quality control; mRNA data were analyzed with Partek GS v6.3. Results : Of 467 miRs represented on arrays, 106 were reliably detected in nonfailing and failing hearts, representing cardiac-expressed miRs. Of these, 21 miRs (19.8%) were significantly (P<0.05) regulated >1.5-fold in CHF; 12 of these were increased >10-fold. Hierarchical clustering showed no specific pattern of regulated miRs. Strikingly, all 21 significantly regulated miRNAs reverted toward nonfailing levels in the post-LVAD hearts (P<0.05 vs CHF) with complete normalization observed for 5 miRs. In contrast, of 755 mRNAs that were altered by >1.3-fold in failing hearts, 38% remained uncorrected post-LVAD. Neither the miR nor mRNA signatures differed between ischemic and nonischemic hearts, suggesting that the RNA profiles reflected advanced CHF, rather than a specific disease etiology. Conclusions : There is a distinct miR signature of end-stage heart failure that is reversible with mechanical unloading. This contrasts with mRNA profiles from the same hearts in which only a fraction of dysregulated mRNAs were normalized by LVAD support. Our findings suggest that regulation of translation by miR may account for some of the discordance between transcriptional and other phenotypic responses to LVAD support.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call