Sodium-glucose cotransporter-2 inhibitors in cardiovascular disease: a gaseous solution

Sodium-glucose cotransporter-2 inhibitors (SGLT2i) are new oral hypoglycemic drugs developed for the management of type 2 diabetes mellitus (T2DM). These drugs target sodium-glucose cotransporter-2, the major glucose transporter expressed in the kidney, which is responsible for approximately 90% of glucose reabsorption from the urine. By blocking renal glucose uptake, SGLT2i induces glycosuria and decreases fasting and postprandial glycemia. The first SGLT2i, phlorizin, was isolated from the root bark of apple trees, but its low water solubility and poor bioavailability precluded its development as an antihyperglycemic agent. Instead, novel phlorizin-based analogs derived from c-aryl glycosides possessed the required solubility, bioavailability, selectivity, and stability required for clinical use. To date, five SGLT2i, canagliflozin, dapagliflozin, empagliflozin, ertugliflozin, and bexagliflozin are certified for use in the United States, with canagliflozin being the first SGLT2i approved for the treatment of T2DM back in 2013. While these drugs were initially developed for their glucose-lowering effects, a series of landmark cardiovascular outcome trials demonstrated that SGLT2i reduces cardiovascular death and hospitalization for heart failure compared to other anti-hyperglycemic agents in patients with T2DM. Moreover, a meta-analysis of these outcome trials found that SGLT2i have beneficial effects on adverse atherosclerotic events (stroke and myocardial infarction) in patients with established atherosclerosis.1 Intriguingly, the modest effect on glycemic control is insufficient to account for the scope of clinical benefit exhibited by SGLT2i, suggesting alternative modes of action for these drugs. Numerous hypotheses have been proposed to explain the cardiovascular benefits of SGLT2i, including reductions in blood pressure, body mass, adipose tissue, uric acid, and inflammation, increases in natriuresis, and improvements in renal function, liver steatosis, and cardiac structure, and alterations in energy metabolism.2 In this perspective, I propose that the synthesis of the signaling gases, nitric oxide (NO), and carbon monoxide (CO), by SGLT2i, contributes to their protective actions in the cardiovascular system and that strategies that target these gases or their downstream effectors will improve the efficacy of SGLT2i in mitigating vascular and cardiac disease.

NO and CO are small, endogenously produced, lipid-soluble gases that play a significant role in the cardiovascular system. Although historically considered as inconsequential or potentially toxic reaction products, these gases have emerged as key signaling molecules that regulate diverse physiological processes in the immune, nervous, respiratory, and cardiovascular systems. NO is formed from arginine by three distinct isoforms of NO synthase (NOS): inducible NOS, neuronal NOS, and endothelial NOS (eNOS).3 eNOS performs a critical task in preserving vascular and cardiac health. The calibrated release of NO by eNOS promotes blood flow and fluidity by dilating blood vessels and blocking platelet aggregation, respectively. Additionally, eNOS-derived NO maintains vascular smooth muscle cells (VSMCs) in a non-proliferative and non-migratory quiescent state and limits the recruitment, infiltration, and activation of leukocytes within the heart and vasculature. eNOS-derived NO also maintains cardiac myocyte function by tempering mitochondrial oxygen consumption, calcium influx, β-adrenergic responsiveness, and myofilament calcium sensitivity. In contrast, inducible NOS, a key player in host defense, provokes vascular and cardiac dysfunction by generating copious quantities of NO that trigger the activation of inflammatory and apoptotic pathways. CO is generated by the catabolism of heme by two discrete isozymes of heme oxygenase (HO).4 HO-1 is a highly inducible enzyme that operates in an adaptive fashion to sustain cellular redox balance, whereas HO-2 is a constitutively expressed form of the enzyme that regulates neurotransmission in both the central and peripheral nervous systems. Alongside CO, the metabolism of heme yields the bile pigment biliverdin which is rapidly converted to the potent antioxidant bilirubin. The HO-1-mediated formation of CO and bilirubin have been demonstrated to exert salutary actions in the cardiovascular system by influencing various cellular processes in both vascular cells and cardiomyocytes.

Considerable evidence indicates that SGLT2i stimulate endothelial NO synthesis.2 Initial work showing that phlorizin restores NO production in human endothelial cells (ECs) treated with palmitic acid has been extended to demonstrate that clinically relevant SGLT2i also increase the generation of NO by ECs isolated from various vascular beds. The elevation of NO synthesis by SGLT2i occurs via the activation and phosphorylation of eNOS by the phosphatidylinositol-3-kinase-AK strain transforming (Akt) or adenosine monophosphate (AMP)-activated protein kinase signaling pathways. Alternatively, SGLT2i increase eNOS levels by reducing the formation of inflammatory cytokines which are known suppressors of eNOS expression. SGLT2i also raise NO levels by interfering with the scavenging of NO by reactive oxygen species. In addition, SGLT2i-mediated increases in arterial wall shear stress, secondary to a rise in blood hematocrit and viscosity, may stimulate eNOS phosphorylation and activation. Moreover, the ability of SGLT2i to increase endothelial glycocalyx thickness may optimize mechanotransduction events that lead to the production of NO. Furthermore, SGLT2i-induced elevations in the arginine/asymmetric dimethyarginine ratio provides additional substrate to drive eNOS-mediated NO synthesis. Thus, SGLT2i promote the synthesis of NO by eNOS via multiple mechanisms.

Growing evidence suggest that SGLT2i also upregulate the production of CO. We recently reported that canagliflozin stimulates the activity of the CO-generating enzyme HO-1 in VSMCs.5 The increase in HO-1 activity by canagliflozin is associated with a rise in HO-1 mRNA and protein that is concentration- and time-dependent and relies on de novo RNA synthesis. The stimulation of HO-1 gene expression is mediated by the reactive oxygen species-nuclear factor erythroid 2-related factor (Nrf2) signaling pathway since it is blocked by antioxidants or by overexpressing a dominant-negative mutant of Nrf2. The source of reactive oxygen species responsible for stimulating HO-1 transcription arises from the mitochondria since HO-1 expression is suppressed by the mitochondrial-targeted antioxidant (2-(2,2,6,6-tetramethylpiperidin-1-oxyl-4-ylamino)-2-oxoethyl) triphenylphosphonium chloride (mito-TEMPO) and by the mitochondrial electron chain complex I inhibitor rotenone. Interestingly, empagliflozin also induces the expression of HO-1 in VSMCs via the activation of AMP-activated protein kinase, indicating that SGLT2i stimulate HO-1 gene expression via multiple signaling cascades.6 However, variable effects on VSMC HO-1 expression have been noted with empagliflozin depending on the vascular source of cells, animal species, and/or culture conditions.5,6 Significantly, the induction of HO-1 contributes to the antiproliferative and antimigratory actions of canagliflozin in VSMCs. Knockdown of HO-1 increases the proliferation and migration of canagliflozin-treated cells. Moreover, the exogenous administration of CO can substitute for HO-1 and fully restore the inhibitory response of canagliflozin, suggesting the HO-1-derived CO mediates the actions of the drug. In other work, empagliflozin was found to attenuate vascular calcification in a mouse model of chronic kidney disease by blocking calcification of VSMCs via the Nrf2-HO-1 pathway.6 More recently, we determined that canagliflozin also causes the induction of HO-1 in ECs, and that HO-1 contributes to the anti-inflammatory action of this drug in these cells, further emphasizing the importance of this enzyme in mediating the effects of canagliflozin in the vasculature.7

The stimulation of gaseous monoxide production by SGLT2i elicits important effects in the cardiovascular system that may contribute to the favorable pharmacological profile of these drugs in T2DM. Both NO and CO function as vasodilators and may mediate the improvement in endothelium-dependent vasodilation noted following the use of SGLT2i in both animals and humans. By inhibiting vascular tone, these gases may also normalize organ perfusion and blood pressure in diabetic patients. Importantly, restoration of NO synthesis by SGLT2i will correct diabetes-associated endothelial dysfunction, a key initial event in atherosclerosis and an important contributor to other vascular diseases. The ability of SGLT2i to induce HO-1 expression is also noteworthy as its protective role in atherosclerosis is well appreciated.4 Aside from inhibiting the development of atherosclerotic lesions, HO-1 and CO may prevent plaque rupture by reducing lipid accumulation within arterial lesions while increasing fibrous cap thickness by blocking inflammation and VSMC apoptosis.4 Collectively, these actions of CO may promote the reduction in myocardial infarction seen in diabetic patients treated with SGLT2i. In addition, the SGLT2i-mediated production of NO and CO may lower the risk and improve outcomes in atherosclerosis-related ischemic stroke by increasing cerebral blood flow, reducing oxidative stress, inflammation, and thrombosis, and improving nerve cell viability and function.3,4 Furthermore, the generation of these signaling gases by SGLT2i may explain the capacity of these drugs to reduce the occurrence of in-stent stenosis since these gaseous monoxides uniquely affect the remodeling response after stent implantation by inhibiting the proliferation and migration of VSMCs while simultaneously stimulating the growth and migration of ECs resulting in optimal healing of stented blood vessels.3,4

SGLT2i are the newest addition to guideline-directed medical therapy in heart failure with recent trials demonstrating a significant reduction in morbidity and mortality in patients with heart failure across the complete ejection fraction spectrum.2 Thus far, SGLT2i have been shown to reduce the risk of heart failure-related hospitalizations, improve cardiomyopathy questionnaire scores, and increase timed walk distances. However, the mechanisms mediating these beneficial effects are not fully known. While several processes are likely to be involved, the ability of SGLT2i to stimulate NO and CO synthesis is highly pertinent. In fact, sublingual and parenteral formulations of NO donor compounds have been utilized for over a century to treat heart disease. Moreover, preclinical studies indicate that the eNOS-NO pathway improves cardiac function in a murine left ventricular pressure overload model treated with empagliflozin by suppressing apoptosis and enhancing myocardial capillarization.8 Interestingly, the induction of HO-1 by canagliflozin in a mouse model of heart failure is paralleled by a decline in cardiac oxidative stress, inflammation, fibrosis and high-output inducible NOS expression.9 There are numerous mechanisms by which NO and CO may obviate heart failure including augmentation of cardiac perfusion, compliance, and contractility, enhancement of mitochondrial biogenesis, and attenuation of endoplasmic reticulum stress and ischemia/reperfusion injury.3,4

A model depicting the effects of SGLT2i on cardiovascular function is shown in Figure 1. Of paramount importance, SGLT2i stimulate eNOS and HO-1 expression and/or activity leading to increases in NO and CO syntheses that reduces atherosclerosis and in-stent stenosis by inhibiting EC dysfunction, vascular tone, oxidative stress, inflammation, thrombosis, vascular cell apoptosis, and VSMC proliferation and migration. The augmented production of NO and CO by SGLT2i also prevents heart failure by mitigating myocardial cell apoptosis, oxidative and endoplasmic reticulum stress, ischemia-reperfusion injury, and fibrosis, while improving myocardial capillarization, cardiac perfusion and contraction, and mitochondrial biogenesis. The identification of a critical role for signaling gases in promoting the cardioprotective actions of SGLT2i raises several options for optimizing the clinical efficacy of SGLT2i. Since the endogenous production of signaling gases is substrate-limited, the combined use of SGLT2i with arginine and heme may boost the endogenous synthesis of NO and CO, respectively.3,4 Homoarginine may also be used since it is structurally similar to arginine and serves as a substrate for eNOS. The deployment of heme is particularly attractive since it functions as both a substrate and inducer of HO-1 and is already approved for the treatment of acute porphyria and other red blood cell-related disorders. In addition to increasing endogenous synthesis of NO and CO by SGLT2i, the biological potency of these gases could be amplified. As many of the favorable effects of NO and CO in cardiovascular tissue arise from elevations in cellular 3′,5′-cyclic guanosine monophosphate, schemes that target this cyclic nucleotide may be advantageous. Accordingly, the use of phosphodiesterase type 5 inhibitors such as sildenafil which prevents the breakdown of 3′,5′-cyclic guanosine monophosphate may potentiate the beneficial actions of SGLT2i. The use of sildenafil is especially attractive given its potential to also enhance the synthesis of NO and CO.10 Finally, the application of soluble guanylate cyclase stimulators that both sensitizes the enzyme to the gaseous ligands NO and CO and stimulates the expression of gaseous monoxide-generating enzymes may further promote the positive effects of SGLT2i.11

Figure 1.

The beneficial effect of sodium-glucose cotransporter-2 inhibitors (SGLT2i) in cardiovascular disease via the production of signaling gases.

SGLT2i stimulate endothelial nitric oxide synthase (eNOS) and heme oxygenase-1 (HO-1) expression and/or activity leading to increases in nitric oxide (NO) and carbon monoxide (CO) synthesis that reduces atherosclerosis and in-stent restenosis by inhibiting endothelial cell (EC) dysfunction, vascular tone, oxidative stress, inflammation, thrombosis, vascular cell apoptosis, and vascular smooth muscle cell (VSMC) proliferation and migration. The augmented production of NO and CO by SGLT2i also prevents heart failure by mitigating myocardial cell apoptosis, oxidative and endoplasmic reticulum (ER) stress, ischemia-reperfusion (IR) injury, and fibrosis, while improving myocardial capillarization, cardiac perfusion and contraction, and mitochondrial biogenesis. Created with BioRender.com.

Aside from increasing endogenous gas synthesis or the potency of downstream gas effectors, dual therapy with SGLT2i and signaling gases should be explored. Although inhalation of NO is approved to treat pulmonary hypertension, questions remain regarding the feasibility of this delivery approach in ameliorating cardiovascular disease. Owing to the high affinity for hemoglobin, the delivery efficiency of NO and CO from the lung to diseased tissues such as the heart and vasculature may be compromised. Moreover, in the case of CO, the potential toxicity following inhalation secondary to the formation of carboxyhemoglobin further detracts from this approach. Other concerns related to the use of inhaled gases include the need for unwieldy pressurized cylinders, risk of exposure to healthcare workers, and lack of target organ selectivity and dosage control. In this regard, the use of donor molecules provides a more promising path for the systemic delivery of signaling gases.8 Many NO and CO releasing molecules have been manufactured that have distinct chemical properties and release kinetics. In addition, gas-releasing molecules have been developed that liberate a defined quantity of gas in response to various biophysical and biochemical stimuli such as ultraviolet light, ultrasound, heat, oxidative stress, and pH. These stimulus-responsive gas releasing molecules may enhance gas targetability while reducing systemic toxicity, permitting the delivery of higher concentrations of the gas to chosen tissues. This is of central importance when dealing with the heart due to its high concentration of myoglobin that can quench NO and CO. Nanoparticle-based gas releasing molecules have also been generated that exhibit improved solubility, pharmacokinetics, toxicity, and specificity for the vasculature. The use of gas-saturated solutions provides another vehicle for the systemic administration of gases, while the recent development of gas-entrapped materials permits the topical application of gases directly to blood vessels and the myocardium. The utility of combining NO and/or CO with SGLT2i has recently been bolstered by a study showing that adjunct therapy with an oral hydrogen sulfide donor provides additional therapeutic benefit beyond that of a SGLT2i in rodent cardiometabolic heart failure models.12 Whether SGLT2i stimulates the synthesis of hydrogen sulfide remains an open question; however, this may provide an additional gaseous pathway by which SGLT2i improve cardiovascular disease. Moreover, SGLT2i may impact the production of other endogenous gases including ammonia, methane, sulfur dioxide, and hydrogen that could also contribute to the clinical profile of these drugs.

In conclusion, there is a growing realization that the generation of signaling gases contributes to the beneficial effects of SGLT2i in the circulation and heart. Strategies that potentiate the endogenous synthesis of gaseous monoxides by SGLT2i, the potency of the downstream effectors of NO and CO, or involve the joint administration SGLT2i with signaling gases represent promising new approaches in preventing and treating cardiovascular disease. Given the large number of delivery platforms available, combination therapy involving signaling gases and SGLT2i may provide a feasible and robust approach in alleviating vascular disease and heart failure in both diabetic and non-diabetic patients.

This work was funded by the National Institutes of Health, National, Heart, Lung, and Blood Institute, No. R01 HL149727.