Murashige et al. (19) analyzed plasma samples obtained from the radial artery, coronary sinus, and femoral vein of 110 patients including heart failure patients and controls through targeted metabolomics (600 nutrient metabolites) and detected a total of 227 nutrient metabolites in the samples. Of these nutrient metabolites, ketone bodies, acetate, and glutamate were taken up in large amounts in both the heart and legs, with the ketone-body uptake being particularly high in patients with heart failure. Ketone bodies are energetically efficient fuels because they produce large amounts of adenosine triphosphate (ATP) molecules with low oxygen demand (20). Therefore, an increased uptake of ketone bodies is believed to protect against heart failure. A significant correlation between the plasma BNP levels and the cardiac uptake of ketone bodies suggests that cardiac utilization of ketone bodies increases with the deterioration of the left ventricular function (21). It has been shown that advanced stages of heart failure are characterized by the myocardial utilization of ketone bodies, despite impaired ketone-body utilization by skeletal muscles (22).

Ketone bodies not only serve as an energy source but are also involved in intracellular signaling. β-Hydroxybutyric acid is an endogenous inhibitor of histone deacetylase (HDAC). HDAC affects the acetylation state of histones and other important cellular proteins and regulates gene expression, including therapeutic targets for pathological cardiac remodeling (23). Lkhagva et al. (24) examined the effects of HDAC in a rat model of isoproterenol-induced heart failure. They demonstrated that oral HDAC inhibitors reduced the amplitude of the increase in inflammatory factors, reduced ventricular volume and improved cardiac function. Hence, it can be hypothesized that ketone bodies can reduce cardiac inflammation by inhibiting HDAC. β-Hydroxybutyric acid is a ligand for G-protein-coupled receptor 41 (GPR41), which is highly expressed in sympathetic ganglia. Its inhibition by β-hydroxybutyric acid suppresses both the sympathetic nervous system activity and heart rate (25). Hence, it can be suggested that ketone bodies directly modulate sympathetic nerve activity, thereby controlling the energy expenditure required for maintaining metabolic homeostasis, which may ameliorate heart failure.

Li et al. (26) showed that, in advanced stages of heart failure, where increased glycolysis adversely affects mitochondrial uncoupling, oxidative phosphorylation, and cardiac energy supply, the binding of empagliflozin to GLUT1 and GLUT4 may help slow down the progression of the disease. Mustroph et al. (27) investigated the direct metabolic effects of SGLT2 inhibitors on cardiomyocytes and demonstrated that when isolated resting cardiomyocytes obtained from human end-stage heart failure transplanted hearts or transverse aortic constriction (TAC)-induced heart failure mouse hearts were treated with 1 μM of empagliflozin for 24 h in the presence of albumin, the expression of the glucose transporter protein GLUT1 increased. However, this effect was not observed in GLUT4.

SGLT2 inhibitors stimulate lipolysis, lipid oxidation, and ketogenesis, all of which reduce body fat (28). Sawada et al. (29) studied the anti-obesity mechanism of SGLT2 inhibitors. They divided diet-induced obese mice into two groups: experimental and control. Then, they subjected the experimental animals to hepatic vagotomy (HVx) and the control animals to a sham operation. Both animal groups were then fed a high-fat diet containing 0.015% tofogliflozin for 3 weeks. In contrast to the sham-operated control group, the HVx group demonstrated reduced adiposity and attenuated levels of protein kinase A (PKA) phosphorylation in their white adipose tissues. This is because the PKA pathway can act as an effector of the liver–brain–adipose axis and activate triglyceride lipase in adipocytes. Yang et al. (30) experimentally demonstrated that canagliflozin promotes fat thermogenesis and catabolism via the β-adrenergic receptor-cyclic adenosine 3′,5′-monophosphate-protein kinase A pathway. These results suggest that SGLT2 inhibition triggers glycogen-depletion signaling and initiates the liver–brain–adipose axis, leading to PKA activation within adipocytes. It can also be suggested that the weight-loss effects of SGLT2 inhibition are mediated, in part, through the liver–brain–adipose neural circuit.

Lee et al. (31) confirmed that the effect of empagliflozin on triglycerides in obese adults varies according to the visceral adipose tissue (VAT). Empagliflozin inhibits the synthesis of triglycerides in subjects with low VAT, whereas it increases triglycerides in subjects with high VAT. Empagliflozin shows a unique effect on lipid metabolism in obese nondiabetic adults, depending on VAT levels in such adults. It inhibits triglyceride synthesis in subjects with low VAT but causes weight loss in those with high VAT even though it increases triglycerides in such individuals. We speculate that the trend toward elevated triglycerides in the latter may be transient and related to VAT mobilization.

Nakamura et al. (32) demonstrated that NHE1 activation increases intracellular Na⁺ loading, which reduces or reverses the drive for Na⁺/Ca²⁺ exchanger-mediated Ca²⁺ efflux and leads to intracellular Ca²⁺ overload and contractile dysfunction. In experimental models of heart failure, inhibition of NHE has been shown to reduce cardiac hypertrophy, fibrosis, remodeling, and contractile dysfunction (33). Trum et al. (34) reported abundant expression of NHE1 in human atrial and ventricular tissues. They also noted a higher expression of atrial NHE1 in HFpEF and atrial fibrillation compared to that in patients without heart failure, especially in end-stage heart transplant recipients. Significant increases in atrial and ventricular NHE1 expressions were observed in patients with heart failure. The group of Zuurbier were the first showing that SGLT2i's can inhibit the NHE1 and lower intracellular Na+ and Ca2+ in cardiomyocytes from rabbit and mouse (35, 36) and in human endothelial cells (37). Significant inhibition of NHE activity was observed when human atrial myocytes and mouse ventricular myocytes were acutely exposed to empagliflozin (1 μmol/L for 10 min). Baartscheer et al. (35) reported that in the absence of albumin, acute (<10 min) or prolonged (3 h) incubation of ventricular myocytes with empagliflozin (0.2–1.0 µM) reduced the cytoplasmic levels of Na⁺ in healthy rabbit cardiomyocytes. In the absence of albumin, 1 µM of empagliflozin attenuated late I_Na in mouse cardiomyocytes with heart failure or sodium-channel mutations (38). Late attenuation of I_Na helps to reduce cardiac action potential prolongation and prevent arrhythmias associated with action potential prolongation.

Although most studies could not detect SGLT2 expression in the human heart (39), the aforementioned experimental SGLT2 inhibitors function independently of glucose levels. Hence, it can be suggested that this is a class effect of SGLT2 inhibitors (36). That SGLT2 inhibitors results in cardioprotection independent of SGLT2 was also supported by a study showing that empagliflozin could lower infarct size followinh a cardiac ischemia-reperfusion insulin in wild-type and SGLT2 knockout animals alike (40). Therefore, SGLT2 inhibitors may confer their protective effects by inhibiting the NHE directly in the myocardium, irrespective of SGLT2 inhibition.

Ca²⁺ homeostasis in cardiomyocytes is maintained by the inflow of mitochondrial Ca²⁺ unidirectional (MCU) transport proteins and the outflow of the mitochondrial Na⁺/Ca²⁺ exchanger (mNCE) (41). It has now been demonstrated that an increased expression of NHE1 in the state of heart failure leads to an overload of intracellular Na⁺, which induces a decrease in the [NAD(P)/NAD(P)⁺] redox potential and an increase in oxidative stress. An increase in the concentration of mitochondrial Na⁺ decreases the fluidity of the inner mitochondrial membrane, which promotes the production of mitochondrial ROS (42). In contrast, a decrease in the concentration of mitochondrial Ca²⁺ hinders the tricarboxylic acid-derived production of NADH through the inhibition of dehydrogenase or F0F1 ATP synthase, which leads to energy deficiency and a reduction in the NADPH-generating ability of antioxidants (43). In addition, an elevated concentration of cytoplasmic Na⁺ in the heart leads to a reduction in the concentration of mitochondrial Ca²⁺ and/or an increase in that of cytoplasmic Ca²⁺. Consequently, cardiac metabolism can shift from fatty acid metabolism to glucose metabolism (44). In addition to inhibiting NHE to indirectly increase the concentration of mitochondrial Ca²⁺ (35) to improve cardiac function, SGLT2 inhibitors increase the activity of the sarcoplasmic/endoplasmic reticulum Ca²⁺–ATPase responsible for Ca²⁺ re-uptake (SERCA2a). As a result, LV diastolic function improves (4). Another way in which empagliflozin regulates Ca²⁺ involves the inhibition of calcium/calmodulin-dependent protein kinase II (CaMKII). Using a mouse model, it has been shown that this inhibition slows down calcium loss from the sarcoplasmic reticulum, resulting in stronger myocardial contractions (45).

In cardiomyocytes isolated from animals fed a high-fat diet, the use of empagliflozin (1 µM) for 2 h in the absence of albumin reduces the ROS levels in cardiomyocytes under metabolic stress through the activation of AMP-activated protein kinase (AMPK) and sestrin2. Kolijn et al. (49) showed that in patients with HFpEF who have undergone left ventricular biopsies, 1-h treatment with empagliflozin (0.5 µM) significantly attenuated oxidized protein kinase GIα (PKGIα), improved the myofilament protein phosphorylation, decreased the cardiomyocyte stiffness, and reduced the cytokine levels. Liu et al. (50) performed the ligation of the left anterior descending artery in C57BL/6 mice, which induced myocardial infarctions in the mice. The animals were postoperatively randomly assigned to two groups for administering either saline or empagliflozin. Compared to the saline-administered group, the empagliflozin -treated group showed improved cardiac function, reduced infarct size, and a decrease in interstitial fibrosis. Hence, it can be suggested that empagliflozin inhibits cardiomyocyte apoptosis by alleviating oxidative stress and restoring the mitochondrial membrane potential.

Various in vitro studies using human and murine models have shown that SGLT2 inhibitors exhibit anti-inflammatory effects by directly reducing endothelial cell and macrophage inflammatory molecules. Gaspari et al. (51) were the first to report the direct endothelial effects of SGLT2 inhibitors. They demonstrated that in the presence of albumin, low concentrations of dapagliflozin (<5 nM) decreased the TNFα-induced NF-κB mRNA expression in human umbilical vein endothelial cells. They also reduced the stretch-induced ROS production in human coronary artery endothelial cells. However, this effect was lost with the intervention of the NHE-1 inhibitor cariporide (52). Hence, it can be suggested that the TNFα-induced activation of NHE-1 produces ROS, which increases the intracellular Na⁺ (37). Zhou et al. (53) reported that dapagliflozin can reverse the H₂O₂-mediated inhibition of serine phosphorylation of endothelial nitric oxide synthase (eNOS) and the expression of sirtuin 1 (SIRT1) in endothelial cells. The function of such endothelial cells can be improved by restoring the eNOS activity, restoring NO bioavailability, and activating SIRT1. As a result, the ROS production is reduced, which ameliorates the endothelial dysfunction. Heme oxygenase-1 (HO-1) is an endogenous antioxidant enzyme. Peyton et al. (54) treated human endothelial cells with canagliflozin and found that it stimulated a concentration- and time-dependent increase in HO-1, which correlated with a significant increase in HO activity. In addition, they reported that canagliflozin induced a concentration-dependent inhibition of endothelial cell proliferation, DNA synthesis, and migration, independent of the inhibition of HO-1 activity and/or expression.

Sustained inflammation and oxidative stress result in the onset and progression of various vasoproliferative diseases. Notably, the proliferation and migration of SMCs play a key role in the pathogenesis of these diseases. Sukhanov et al. (55) reported low levels of SGLT2 expression in human aortic SMCs. They also demonstrated that pretreatment of these SMCs with empagliflozin increases the expressions of SGLT2 and NLRP3. In addition, empagliflozin impacts the production of IL-17A/TRAF3IP2-dependent superoxide and hydrogen peroxide, the expression of NLRP3, the activation of caspase-1, and the release of mitogenic and migratory IL-1β and IL-18. All these effects inhibit the proliferation and migration of SMCs at normal glucose levels and without inducing cell death. Benetti et al. (56) found that empagliflozin also inhibits SGLT2 expression by suppressing NLRP1 expression, which, in turn, reduces the levels of pro-inflammatory and pro-atherosclerotic cytokines IL-18β and IL-3 in SMCs.

Yumei et al. (57) reported that dapagliflozin increases P-AMPK in cardiac fibroblasts exposed to lipopolysaccharide. Dapagliflozin attenuates the increase in NHE-1 mRNA and the association between NHE-1 and heat shock protein-70 (Hsp70). This effect depends on the activation of AMPK.