Cardiac REDD1 alters glucose and fatty acid metabolic gene expression via an mTORC1-independent, PPAR alpha-dependent mechanism and drives hypertrophic growth

Abstract

BackgroundRegulated in development and DNA damage 1 (REDD1) is a highly inducible molecule that plays a role in numerous physiological and pathophysiological processes. It is a well-established negative regulator of mammalian target of rapamycin complex 1 (mTORC1), which is critical for maintaining elevated fatty acid-to-glucose oxidation ratio in the heart. In addition, REDD1 deletion results in hyperglycemia, suggesting that REDD1 is critical for tissue glucose metabolism. The role of REDD1 in regulating cardiac glucose and/or fatty acid metabolism in response to physiologic or pathophysiologic cues, however, remains unexplored. MethodsHerein, we utilize AC16 cardiomyocytes with REDD1 deletion, as well as mice with global or cardiomyocyte-specific deletion of Redd1, and their respective controls. We also subject these mice cardiac pressure overload using transverse aortic constriction (TAC) for 2 weeks or sham operation as a control. To examine the molecular regulators of glucose oxidation, we utilized qPCR and western blotting to evaluate pyruvate dehydrogenase (PDH) kinase (PDK) and phospho-PDH (pPDH) levels, respectively. We also directly measured PDH activity and glucose-driven cellular respiration. To investigate the complete REDD1-dependent transcriptome and metabolome, we performed RNA-sequencing (RNA-Seq) and untargeted metabolomics, respectively. To determine if the observed gene expression changes were dependent upon transcription factor peroxisome proliferator-activated receptor alpha (PPAR), we utilized an established pharmacologic PPAR inhibitor, GW6471. Here, we measured PPAR activity directly, as well as the expression of its target genes. In order to determine if our observed effects were mTORC1-dependent, we utilized mTORC1-specific inhibitor, everolimus. Finally, we measured cardiac hypertrophy using gravimetric analyses (heart weight (HW)-to-body weight (BW) or HW-to-tibia length (TL) ratios) and histological analyses of cardiomyocyte cross sectional area (CSA). We also measured mRNA and protein levels of pathological hypertrophic markers Natriuretic Peptide B (Nppb) and Cardiac Ankyrin Repeat Protein (CARP), respectively. ResultsOur data demonstrate that physiological levels of glucose induce REDD1 expression in cardiomyocytes. Further, we show that in cardiomyocytes or the hearts of mice with REDD1 deletion, there is elevated PDK4 expression, as well as increased levels of pPDH (S300 and/or S293) and reduced PDH activity. Interestingly, everolimus treatment has no effect on these alterations. In vitro, we also observe elevated glycolysis and glycolytic capacity, and reduced maximal respiratory capacity (MRC) in the presence of glucose. Interestingly, our RNA-Seq data reveals the upregulation of genes involved in fatty acid catabolism. Further, we demonstrate that PPAR activity is enhanced, and everolimus treatment also has no effect on this parameter. Additionally, we show that treatment of cardiomyocytes with GW6471 normalizes the expression of its target genes (PDK4, ACSL1) and levels of pPDH (S300), that are elevated in cells with REDD1 deletion. Finally, we observe elevated REDD1 in the hearts of mice following TAC. Moreover, we show reduced HW/BW, HW/TL, cardiomyocyte CSA, and levels of cardiac Nppb and CARP in mice with cardiomyocyte Redd1 deletion subjected to TAC versus controls also subjected to TAC. Importantly, TAC-induced reductions in cardiac Pdk4 and pPDH (S293 and S300), are normalized to control levels in mice with Redd1 deletion subjected to TAC. ConclusionsTogether, our findings suggest that physiological glucose-induced and pathological pressure overload-induced REDD1 is required for enhancing glucose oxidation and suppressing fatty acid oxidation in cardiomyocytes. In this way, REDD1 supports cardiac hypertrophic growth. We also outline a mechanism whereby REDD1 inhibits PPAR activity, thereby inhibiting the expression of its target genes, including PDK4 and those involved in fatty acid oxidation. Finally, we demonstrate that these effects are independent of REDD1s ability to inhibit mTORC1.