2.1.1. Fatty Acids

The beta-oxidation of fatty acids was first demonstrated by Knoop in 1904. FAO represents a predominant fuel source for the myocardium and typically exhibits high flexibility following changes in substrate availability [27]. FAs pass the plasma membrane via tissue-specific transporters and are processed to enter mitochondria for oxidation. Each cycle of FAO produces acetyl-CoA, NADH, and FADH<sup>2</sup> for ATP production, and this production is estimated to equal 50% to 70% of the ATP consumed during contraction [28]. In HF, a reduction in the expression of genes encoding FAO enzymes was observed in patients and animal models [29], and this decrease is linked to a reduction in mitochondrial respiration; both of these effects are directly correlated with the stage (worsening from early to advanced) and the cause of HF. Accordingly, one of the most consistent metabolic changes in HF is the marked downregulation of fatty acid utilization, which is mainly described in end-stage HF studies [24,30,31] (Figure 1). Cardiac lipid accumulation is also observed in failing hearts from patients who are affected by diabetes and obesity-related metabolic complications, triggering a phenomenon known as "lipotoxicity" [32]. Reduced levels of FAO pathways and the accumulation of incompletely oxidized fatty acids trigger a mismatch between the supply and oxidation of FAs [33]; moreover, the oxidation of glucose and lactates becomes impaired, which leads to an uncoupling of yet increased glycolysis and less cardiac efficiency and function.

In addition to respiratory inhibition, lipotoxicity leads to other mitochondrial dysfunctions caused by a free FAs-dependent drastic permeability of mitochondrial membranes increasing proton conductance [34] and, being long-chain acyl-CoA esters, potent inhibitors of adenine nucleotide translocator (ANT) [35], with serious implications in ROS generation. In the first scenario, free FAs lead to the dissipation of the membrane potential by continuously passing across the IMM from the intermembrane space (IMS) to the matrix and there, releasing protons. Indeed, the matrix has higher pH than IMS. In a second step, FAs come back to the IMS as anions taking advantage of ANT; here, for each of these cycles, the mitochondrial matrix is enriched by one proton. This action would affect membrane permeability and bioenergetics, and trigger cell death [36,37]. Recently, it was demonstrated that introducing omega-3 FA into the diet modifies the composition of free FA, shifting from the precursors of inflammatory states (arachidonic acid) to those involved in their resolution (eicosapentaenoic acid), thus ameliorating cardiac dysfunction [38].

In addition to their impact as energy-providing substrates, FAs act as mediators of signal transduction and as ligands for nuclear receptor peroxisome proliferator-activated receptors (PPARs)-α. The PPAR-α/PXRα complex (retinoid X receptor pathway) and its transcriptional partner Peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC1-α) are considered master regulators of mitochondrial biogenesis [39] and FAO by sensing dietary needs and pathological states. Interestingly, in almost all studies in the field, these two compounds were found to be significantly downregulated and thus become the culprits of the metabolic shift toward glycolysis in HF [40–42]. In support of this finding, preclinical studies observed repression of OXPHOS, increased oxidative stress, and accelerated HF following pressure overload in cardiac PGC1-α-knockout (KO) transgenic animals.

#### 2.1.2. Glucose

In an effort to counteract the decrease in ATP generation due to OXPHOS impairment, glycolysis rates become elevated in HF [43].

In 1907, Locke and Rosenheim were the first to study myocardial glucose uptake in an isolated rabbit heart model [44]: Glucose and its metabolites play several roles in cardiac myocytes, and their uptake is regulated by the membrane translocation of glucose transporter (GLUT) 1 and GLUT4 followed by phosphorylation into 6-phosphoglucose (G6P). Of note, glycolysis and glucose oxidation are differentially regulated in the heart. Therefore, in failing hearts, the elevated rate of glycolysis is not needed to translate into enhanced glucose oxidation [45]. A plethora of studies support this notion: Patients with HF present enhanced levels of cardiac glycolysis without increases in glucose oxidation, lactate, and pyruvate accumulation [46]. Instead, significant abolition of glucose oxidation was detected in many animal models of HF and in a pool of patients with congestive HF [47] and is currently considered a metabolic marker that proceeds with the development of cardiac deterioration in HF [48]. This feature is due to deregulation of the overall

mitochondrial oxidative capacity and alterations in pyruvate dehydrogenase (PHD) activity, which lead to a reduction in the conversion of pyruvate into acetyl-CoA [49–51].

Nevertheless, the literature reports apparently conflicting evidence for glucose utilization in HF over time. Additional studies showed that the glucose oxidation rates either remained unchanged in a compensated HF rat model [21] or were elevated in canine cardiac-pacing experiments [41]. The reason for this contradictory evidence is unclear. The most accredited hypothesis is that glucose oxidation varies according to the severity of HF (and to a small extent to the cause); it starts to increase at initial stages and then remains unchanged until it decreases at advanced stages [52].

#### 2.1.3. Ketones

New metabolic changes, including ketone bodies and branched-chain amino acids (BCAAs), were recently identified as alternative substrates in end-stage HF (Figure 1).

Ketone bodies are small molecular energy substrates that are rapidly mobilized and produced during fasting and starvation through hepatic ketogenesis and provide three types of ketones: acetone, β-hydroxyburate (βOHB), and acetoacetate. Ketones can enter cardiomyocytes and be translocated to the mitochondrial matrix, where βOHB is oxidized into acetoacetate by the key enzyme βOHB dehydrogenase (BDH1); thus, acetoacetate is stimulated by succinyl-CoA to acetoacetyl-CoA and then converted to acetyl-CoA by acetyl-CoA acetyltransferase. Consequently, acetyl-CoA is able to enter the tricarboxylic acid cycle and ETC to produce ATP [53,54].

Recent studies by Aubert and associates found that BDH1 was significantly upregulated at least two- to three-fold in an HF mouse model, and this increase is accompanied by significant stimulation of ketone metabolism, which suggests the role of ketone bodies as alternative substrates when glucose oxidation is downregulated in failing hearts [55]. Consistent with previous findings, enhanced myocardial ketone body oxidation rates were found in an ex-vivo-isolated murine heart [56], and significantly higher ketone body levels were detected in the peripheral blood of patients affected by HF than in normal subjects [57]. Although metabolomics studies confirmed that HF utilizes ketones as an alternative substrate in an effort to supply energy at the end stage of human HF, this change does not apport beneficial effects to the evolution of the pathology. Indeed, the plasma ketone concentration and acetone are reportedly linked with malignant prognosis in patients with chronic HF [58,59]. These findings might provide new insights for therapeutic approaches toward HF [60].
