Reduced hypusination of EIF5A impairs free ubiquitin generation resulting in the development of HFpEF

Project summary: Heart Failure with Preserved Ejection Fraction (HFpEF) affects over half of all heart failure patients, yet underlying mechanisms remain poorly understood. Using proteomic and transcriptomic analysis of human HFpEF myocardium, we found significant downregulation in ribosomal structure and protein translation pathways. In particular, activation of a critical translation factor, elongation factor-5A (eIF5A) by a process called hypusination (to form eIF5AHyp) is significantly reduced in HFpEF hearts. Mice with cardiac specific reduction of DHPS, the enzyme that generates eIF5AHyp (ciDHPS-KD), exhibit a HFpEF phenotype that supports a mechanistic role. Ribosome profiling using cardiomyocytes with pharmacologically inhibited eIF5AHyp identified a group of genes involved with ubiquitination as being most inefficiently translated. In particular, the ubiquitin precursor (UBC) and deubiquitinases (USP9X and USP7) responsible for generating free ubiquitin were major inefficient translation targets. USP9X and USP7 protein are downregulated in human HFpEF and ciDHPS-KD myocardium, and in myocytes with eIF5AHyp reduction. Total protein ubiquitination is also reduced in these conditions, and this does not appear due to increased clearance of ubiquitinated proteins by the proteasome or lysosome. This project hypothesizes that reduced hypusination of eIF5A impairs the translation of ubiquitin precursors and deubiquitinases that are all required to provide the substrate for free ubiquitin, and that this is mechanistically related to cardiac HFpEF pathobiology. This is studied in 3 specific aims. First, I test the hypothesis that depressed eIF5AHyp caused by lowering DHPS or as found in human HFpEF heart impedes translation of UBC, USP9X, and USP7 resulting in free ubiquitin deficiency, hypo-ubiquitination, and thereby impaired PQC. This is tested using paired polysome profiling and proteomics in human myocardium and the ciDHPS-KD myocardium. Proteasome flux, free ubiquitin, and protein aggregation will be assessed in ciDHPS- KD mice. Other translation-stalled protein groups/pathways will also be determined and can be pursued subsequently. Secondly, I determine mechanisms by which reduced eIF5AHyp is linked to lower UBC, USP9X, and USP7 translation. This is thought to involve specific amino acid motifs that are hard-to-translate without eIF5aHyp but may also be related to activation of a broader integrated ribosomal stress response. Both of these are tested. Lastly, I will examine this pathobiology in vivo, testing whether reducing the free ubiquitin pool by genetic knock-down of UBC is sufficient to induce a HFpEF phenotype, or that replenishing this pool by cardiac over-expression of a single ubiquitin or enhancing DHPS activity with spermidine supplementation can rescue the HFpEF phenotype in ciDHPS-KD mice. Together these studies will advance our understanding of a new previously unknown feature of myocardial etiology in HFpEF, and the role that depressed protein translation from reduced eIF5a activity plays in ubiquitin-related protein quality control.