Project description:Introduction: Left ventricular assist devices (LVADs) provide significant pressure and volume unloading, which reverse key structural features of heart failure, including hypertrophy, fibrosis, and altered sympathetic innervation. This has led to LVADs increasing utilization as both a bridge or destination therapy for heart failure. While distinct metabolic changes occur in the human myocardium with LVAD placement, the specific molecular mechanisms underlying these changes have not been identified. Objectives: To identify the role of MuRF1 in the metabolic changes in cardiomyocytes unloading in vitro. Methods: HL-1 atrial cardiomyocyte cells were plated on silastic membranes coated with gelatin/fibronectin and transduced with AdshRNA MuRF1 (or AdshRNA Scramble control) to knock-down MuRF1 protein to <25% of controls and subjected to 15% biaxial stretch at 1 Hz using the Flexcell FX-5000™ Compression System in serum free DMEM (or no-stretch). After 6 hours stretch, media was collected in parallel with time-matched no-stretch controls, followed by serial collections at 1, 3, 6, and 12 hours unloading (i.e. after termination of stretch). Media was analyzed by untargeted metabolomics using GC-MS. Results: After 6 hours stretch, control HL-1 cell media (AdshRNA Scramble) had 19 significantly altered metabolites compared to non-stretched control cell media (AdshRNA Scramble) by t-test, involved in: 1) glyoxylate and dicarboxylate metabolism (p=0.00087043, FDR=0.071375), 2) methane metabolism (p=0.0041113, FDR=0.089824), 3) glycine, serine and threonine metabolism (p=0.0043816, FDR=0.089824), and 4) aminoacyl-tRNA biosynthesis(p=0.0060141, FDR=0.09863). To determine the role of MuRF1 in stretch, we additionally compared the AdshRNA Scramble groups to AdshRNA MuRF1 +/- stretch at 6 hours and identified 41 significant metabolites (of 79 identified) by ANOVA, involved in: 1) phenylalanine, tyrosine, and tryptophan biosynthesis (p=0.0036482, FDR=0.05983), 2) aminoacyl-tRNA biosynthesis (p=3.74E-07, FDR=3.06E-05), and 3) valine, leucine, and isoleucine biosynthesis (p=0.0021624, FDR=0.053255). We next compared the 6 hour stretch time point of the four groups to the 1, 3, 6, and 12 hours unloading conditions to identify MuRF1-associated metabolites during the unloading period. At 12 hours of unloading (representative 1, 3, and 6 hours unloading), MuRF1 knock-down cell media had 35 (of 82 named metabolites) significantly different metabolites by ANOVA involved in 1) phenylalanine, tyrosine and tryptophan biosynthesis (p=0.0023668, FDR=0.048519), 2) aminoacyl-tRNA biosynthesis (p=0.0011403, FDR=0.0032631), 3) phenylalanine metabolism (P= 0.0011403, FDR 0.042673),and arginine and proline metabolism (p=0.0015612, FDR 0.042673). A final analysis comparing all four groups across all five time points identified 21 significant metabolites. When MuRF1 was knocked down (no stretch), these 21 metabolites were significantly increased without stretch. In response to stretch and unloading, these metabolites were further decreased in AdshRNA Scramble (control) cell media decreased. In contrast, these stretch and unloading induced decreases were attenuated in AdshRNA MuRF1 cell media. These metabolites included nine (9) amino acids (citrulline, phenylalanine, tyrosine, asparagine, 2-ketovaline, and -ketoleucine/ketoisoleucine, threonine, isoleucine, leucine), three (3) metabolic co-factors (Pantothenic acid, Myoinositol, 5,6-Dihydrouracil), and one fatty acid (stearic acid). Conclusion: These studies identify for the first time a role for MuRF1 in generating key amino acids recently reported in LVAD unloading in human myocardium using an in vitro model of atrial cardiomyocyte unloading.
2017-12-25 | ST000925 | MetabolomicsWorkbench