H OH OH metabolites Article Enhancing Glycolysis Protects against Ischemia-Reperfusion Injury by Reducing ROS Production Claudia Beltran 1 , Rosario Pardo 2 , Diana Bou-Teen 1 , Marisol Ruiz-Meana 1 , Josep A. Villena 2,3 , Ignacio Ferreira-González 1,4, * and Ignasi Barba 1,5,6, * 1 Cardiovascular Diseases Research Group, Department of Cardiology, Vall d’Hebron University Hospital and Research Institute, Universitat Autònoma de Barcelona, 08025 Barcelona, Spain; claudia.beltran@vhir.org (C.B.); diana.bou@vhir.org (D.B.-T.); mruizmeana@gmail.com (M.R.-M.) 2 Laboratory of Metabolism and Obesity, Vall d’Hebron Research Institute, Universitat Autònoma de Barcelona, 08025 Barcelona, Spain; rosario.pardo@vhir.org (R.P.); josep.villena@vhir.org (J.A.V.) 3 Centro de Investigación Biomédica en Red sobre Diabetes y Enfermedades Metabólicas Asociadas (CIBER-DEM), 28029 Madrid, Spain 4 Centro de Investigación Biomédica en Red sobre Epidemiología y Salud Pública (CIBERESP), 28029 Madrid, Spain 5 Centro de Investigación Biomédica en Red sobre Enfermedades Cardiovasculares (CIBER-CV), 28029 Madrid, Spain 6 Facultat de Medicina. Universitat de Vic – Universitat Central de Catalunya (UVic- UCC), 08500 Vic, Barcelona, Spain * Correspondence: nachoferreira@secardiologia.es (I.F.-G.); ignasi.barba@vhir.org (I.B.)  Received: 19 March 2020; Accepted: 27 March 2020; Published: 30 March 2020  Abstract: After myocardial ischemia-reperfusion, fatty acid oxidation shows fast recovery while glucose oxidation rates remain depressed. A metabolic shift aimed at increasing glucose oxidation has shown to be beneficial in models of myocardial ischemia-reperfusion. However, strategies aimed at increasing glucose consumption in the clinic have provided mixed results and have not yet reached routine clinical practice. A better understanding of the mechanisms underlying the protection afforded by increased glucose oxidation may facilitate the transfer to the clinic. The purpose of this study was to evaluate if the modulation of reactive oxygen species (ROS) was involved in the protection afforded by increased glucose oxidation. Firstly, we characterized an H9C2 cellular model in which the use of glucose or galactose as substrates can modulate glycolysis and oxidative phosphorylation pathways. In this model, there were no differences in morphology, cell number, or ATP and PCr levels. However, galactose-grown cells consumed more oxygen and had an increased Krebs cycle turnover, while cells grown in glucose had increased aerobic glycolysis rate as demonstrated by higher lactate and alanine production. Increased aerobic glycolysis was associated with reduced ROS levels and protected the cells against simulated ischemia-reperfusion injury. Furthermore, ROS scavenger N-acetyl cysteine (NAC) was able to reduce the amount of ROS and to prevent cell death. Lastly, cells grown in galactose showed higher activation of mTOR/Akt signaling pathways. In conclusion, our results provide evidence indicating that metabolic shift towards increased glycolysis reduces mitochondrial ROS production and prevents cell death during ischemia-reperfusion injury. Keywords: metabolic shift; heart; myocardial infarction 1. Introduction Myocardial infarction, and in particular STEMI (ST segment elevation myocardial infarction), is one of the major contributors to mortality and morbidity worldwide [1]. While the implementation Metabolites 2020, 10, 132; doi:10.3390/metabo10040132 www.mdpi.com/journal/metabolites Metabolites 2020, 10, 132 2 of 16 of reperfusion strategies, such as percutaneous angioplasty, has resulted in a reduction in mortality over the last couple of decades, this has been at the expense of an increased burden of heart failure. It has been suggested that the reduction of infarct size would result in improved long term cardiovascular health and fewer events. Thus, treatments addressed to reduce the extension of myocardial infarcts are required [2]. Reperfusion, i.e., the restoration of coronary blood flow, is required for the treatment of myocardial infarction; however, it induces tissue damage that, at least in theory, could be prevented [3–5]. The identification of the mechanisms involved in ischemia-reperfusion injury is essential for the successful translation of cardioprotective strategies to the clinical practice. The heart is an organ with high energy demand due to its mechanical function. On average, a healthy heart obtains most of its energy (~70%) from fatty acids through β-oxidation and the remaining from glucose oxidation [6]. However, in a pathological environment, there may be changes in substrate utilization [6,7]. In particular, in the case of ischemia, there is a relative increase of fatty acid oxidation that continues once perfusion is restored [8,9]. In diabetic patients, the levels of circulating glucose at admission have been associated with clinical outcome after acute myocardial infarction (AMI) [10], suggesting that myocardial metabolism may play a role in it. Metabolic shift from β-oxidation to glycolytic metabolism can reduce the need of oxygen by 11–13% [11] and it has been shown that NAD+ precursors induced cardioprotection is mediated, at least in part, via glycolytic stimulation [12]. However, clinical studies targeting metabolism have been so far inconclusive [13]. While there are some promising results showing reduced in-hospital death and cardiac arrest from glucose/insulin/potassium (GIK) infusion when administered during transfer to hospital, [14] this benefit is not evident when GIK is administered once the patient arrives at the hospital [15]. Similarly, the modulation of the availability of glucose and free fatty acids in patients with hibernating myocardium or chronic ischemia has not been shown to have a clinical benefit [16]. However, metabolic modulation has been proposed to have therapeutic potential in heart failure and ischemic-reperfused heart [13]. Generation of ROS (reactive oxygens species) is exacerbated upon reperfusion, leading to a cascade of events that eventually precipitates cell death [17], through a variety of mechanisms including the opening of mitochondrial transition pore [18]. On top of acute damage, mitochondrial ROS can also initiate pathological events that develop over time [19]. Furthermore, transient inhibition of the electron transport chain at the onset of reperfusion reduces ROS and decreases cell death and the size of myocardial infarction in animal models [20,21]. In a cell culture model, changing the source of carbon allows modulation of the amount of substrate utilized by aerobic glycolysis [22]. The use of galactose as substrate enhances oxidative phosphorylation and reduces aerobic glycolysis [23–25] and has been used in toxicology experiments [24,26,27] and in studies addressed to asses mitochondrial dysfunction [23]. However, to our knowledge, it has not been applied to the study of the mechanisms involved in the protection afforded by the metabolic shift in ischemia-reperfusion injury. The H9C2 cells have been validated for metabolic studies, in particular during ischemia-reperfusion injury, as they share some important features with adult cardiomyocytes, such as high levels of ATP, p-AMPK, and mitochondrial morphology, among others [28]. The use of 13 C-labeled glucose enables to study aerobic glycolysis and Krebs cycle kinetics by the evaluation of 13 C incorporation in Krebs cycle intermediates, or metabolites in fast exchange with them [29]. Also, because the short chain fatty acid acetate can be incorporated into the Krebs cycle [30], we have used it as a marker of β-oxidation. The objective of the present work was to investigate if ROS are involved in the protection against ischemia-reperfusion injury afforded by changes in the rate of glycolysis, oxidative phosphorylation and β-oxidation. The elucidation of the mechanisms involved in the protection exerted by the metabolic shift may facilitate the development of novel therapeutic strategies against ischemia-reperfusion injury. Metabolites 2020, 10, 132 3 of 16 Metabolites 2019, 9, x FOR PEER REVIEW 3 of 15 2. Results 2. Results 2.1. Characterization of the Galactose-Induced Metabolic Modulation 2.1.Cells Characterization treated for of24the Galactose-Induced h with Metabolic Modulation galactose-containing media did not show any apparent difference in morphology or number compared to those grown in glucose media. Therefore, all experiments were Cells treated for 24 h with galactose-containing media did not show any apparent difference in performed morphology withorcells passaged number and grown compared to thoseingrown glucose-containing in glucose media.media, in which Therefore, allmedia was changed experiments were to performed glucose orwith galactose 24 h prior cells passaged andeach experiment. grown in glucose-containing media, in which media was changed H9C2 cells to glucose treated with or galactose 24 h galactose prior eachfor 24 h showed increased citrate synthase activity and higher experiment. oxygenH9C2consumption thanwith cells treated glucose-treated galactose for cells (Figure 1A,B). 24 h showed Thiscitrate increased is consistent synthasewith galactose-grown activity and higher cells relying oxygen on oxidative consumption thanphosphorylation glucose-treated cells as (Figure a main1A,B). source of is This ATP production. consistent The increase in with galactose-grown oxygen consumption cells relying in galactose-treated on oxidative phosphorylationcells as awas main paralleled source ofby ATPan increase production. in the Theexpression increase inof NDUFB9 and COX IV, in oxygen consumption two constituent proteins galactose-treated of complex cells was paralleledI and by anIV,increase respectively, in theofexpression the oxidative of phosphorylation NDUFB9 and COX system (OxPhos) IV, two (Figure constituent 1C–E).ofSomehow proteins complex I surprisingly, mRNA levels and IV, respectively, of COX IV, of the oxidative asphosphorylation well as mRNA ofsystemother mitochondrial (OxPhos) (Figure genes remained 1C–E). Somehow unchanged (Supplementary surprisingly, mRNA levels Figure S1). IV, of COX Also, as well there wereasnomRNA of other differences mitochondrial in the genes remained content of ALDH2, unchanged a matrix-located (Supplementary mitochondrial Figure protein, S1). between Also, there glucose were no differences and galactose-grown cells in thenot (data content shown).of ALDH2, a matrix-located These results suggest thatmitochondrial protein, although mitochondrial betweenfunction oxidative glucose isand galactose-grown increased cells with in cells treated (datagalactose not shown). These for 24 results h, this is notsuggest due to athat net although increase in mitochondrial oxidative mitochondrial biogenesis. function is increased in cells treated with galactose for 24 h, this is not due to a net increase in mitochondrial biogenesis. Figure 1. Graphs showing (A) citrate synthase activity in cells grown in galactose media compared to Figure cells grown1. Graphs showing in glucose (A) media, citrate and synthase activity (B) mitochondria in cells oxygen grown in galactose consumption media per million compared of cells to in glucose and galactose media. (C–E) correspond to the Western blot of NDUFB9 and COXIV for glucose (n in cells grown in glucose media, and (B) mitochondria oxygen consumption per million of cells = 5) glucose and and (n galactose galactose media. = 4)-treated (C–E) cells correspond and their to the Western quantification. blot ofdifferent * Statistically NDUFB9 p