An Updated Insight Into Molecular Mechanism of Hydrogen Sulfide in Cardiomyopathy and Myocardial Ischemia/Reperfusion Injury Under Diabetes

Cardiovascular diseases are the most common complications of diabetes, and diabetic cardiomyopathy is a major cause of people death in diabetes. Molecular, transcriptional, animal, and clinical studies have discovered numerous therapeutic targets or drugs for diabetic cardiomyopathy. Within this, hydrogen sulfide (H2S), an endogenous gasotransmitter alongside with nitric oxide (NO) and carbon monoxide (CO), is found to play a critical role in diabetic cardiomyopathy. Recently, the protective roles of H2S in diabetic cardiomyopathy have attracted enormous attention. In addition, H2S donors confer favorable effects in myocardial infarction, ischaemia-reperfusion injury, and heart failure under diabetic conditions. Further studies have disclosed that multiplex molecular mechanisms are responsible for the protective effects of H2S against diabetes-elicited cardiac injury, such as anti-oxidative, anti-apoptotic, anti-inflammatory, and anti-necrotic properties. In this review, we will summarize the current findings on H2S biology and pharmacology, especially focusing on the novel mechanisms of H2S-based protection against diabetic cardiomyopathy. Also, the potential roles of H2S in diabetes-aggravated ischaemia-reperfusion injury are discussed.


INTRODUCTION
The International Diabetes Federation has estimated that global diabetic prevalence might rise to 10.2% by 2030 and 10.9% by 2045, respectively (Saeedi et al., 2019). On a global scale, increasing epidemic of diabetes exerts serious economic and social implications in many countries, and 77% of diabetic individuals live in low and middle income countries (Standl et al., 2019). Apart from overt diabetic patients, 352.1 million people are estimated to be at risk of diabetes globally (also known as pre-diabetes) (Standl et al., 2019). Diabetic pandemic seems to be one of global public health problems that need to be solved urgently, especially considering the increasing number of deaths caused by diabetic complications (Ayensa-Vazquez et al., 2020;Magliano et al., 2020). Cardiovascular disorders are the most complications of diabetes with a serious impact on healthcare resources, and they are also the main causes of death in diabetic patients (International Hypoglycaemia Study Group, 2019). Diabetesinduced cardiovascular diseases encompass peripheral artery disease, cerebrovascular disease, coronary artery disease, heart failure, and cardiomyopathy (International Hypoglycaemia Study Group, 2019). A host of clinical studies have demonstrated that cardiovascular disease risks are 3-to-4-fold higher in diabetic patients (Xie et al., 2017). Diabetic cardiomyopathy, a unique type of cardiac damage, exhibits abnormal myocardial structures independent of hypertension, coronary artery disease, valvular heart disease or other known precipitating factors for heart failure (Rubler et al., 1972;Adeghate and Singh, 2014;Zamora and Villena, 2019). A population-based study has reported that diabetic patients are at a higher risk to develop diastolic or systolic dysfunctions when compared to normal individuals (Devereux et al., 2000;Bertoni et al., 2006). Diabetic cardiomyopathy has gained increased attention due to its high morbidity and poor prognosis (Oktay et al., 2000;Pikkemaat et al., 2017). Studies have confirmed that the pathogenesis of diabetic cardiomyopathy is multifactorial and mainly include, but not limited to, oxidative/nitrosative stress, impaired calcium homeostasis, mitochondrial dysfunction, and accumulation of advanced glycation end products (AGEs), as well as increased inflammation (Borghetti et al., 2018;Al Hroob et al., 2019). Although several anti-hyperglycemic drugs are commonly used for the treatment of diabetes, they can hardly improve diabetic cardiomyopathy (Borghetti et al., 2018). As a result, there is an urgent unmet medical need to discovery novel and effective therapies directly against diabetic cardiomyopathy.
Historically, hydrogen sulfide (H 2 S) was viewed as a toxic gas with foul smelling, but more and more studies have identified H 2 S as a member of the gasotransmitter family, along with nitric oxide (NO) and carbon monoxide (CO) (Beck and Pfeilschifter, 2020). In the cardiovascular system, H 2 S is endogenously generated by three endogenous enzymes, including cystathionine β-synthase (CBS), cystathionine c-lyase (CSE), and 3-mercaptopyruvate sulfur transferase (3-MST)/cysteine aminotransferase (CAT) . Besides, H 2 S is also generated from D-cysteine by D-amino acid oxidase (DAO) (Wen et al., 2018). In the last several decades, the biological and pharmacological effects of H 2 S on various systems are gradually revealed after the measurement of brain H 2 S content in 1989 (Goodwin et al., 1989;Wang, 2002;Wang, 2012). Subsequently, H 2 S is emerging as an essential regulator in a diversity of physiological functions in the biological systems (Xie et al., 2016). Similar to NO and CO, H 2 S is critically involved in cardiovascular regulation (Testai et al., 2020). As the third gaseous signaling molecule, H 2 S is brought into the spotlight of cardiovascular field as H 2 S exhibits multiple cardiovascular protective effects, including vascular relaxation, pro-angiogenesis, anti-hypertension, anti-atherosclerosis, attenuation of oxidative stress and inflammation, as well as amelioration of myocardial ischemia-reperfusion injury (Gorini et al., 2020;Teoh et al., 2020). Coincidentally, accumulating evidence has suggested that H 2 S could modulate numerous molecular cascade events to ameliorate diabetesinduced cardiac damage (Kar et al., 2019a). Thereafter, we will highlight recent advances in the current understanding of the ability of H 2 S to relieve diabetic cardiomyopathy and diabetes-deteriorated ischaemia-reperfusion injury. Interactions of H 2 S with NO in diabetic cardiomyopathy are also described. Eventually, the possible challenges and directions of H 2 S-based pharmacological therapy in diabetic cardiomyopathy are overviewed and proposed.

PATHOPHYSIOLOGY OF DIABETIC CARDIOMYOPATHY
In 1954, Lundbaek reported a distinct clinical entity with cardiac dysfunction in type 2 diabetic individuals (termed as diabetic cardiomyopathy), independent of hypertension and coronary artery disease (Lundbaek, 1954). Rubler et al. (1972) had observed glomerulosclerosis and heart failure in diabetic individuals, although these diabetic patients had no hypertension, coronary artery disease, valvular or congenital heart disease. A large population study found that the risk of heart failure hospitalization in diabetic patients was still higher even though all cardiovascular risk factors were well controlled (Rawshani et al., 2018). With rapid increases in diabetic incidence and prevalence, diabetic cardiomyopathy, an important cardiovascular complication of diabetes, is becoming one of the main causes of disability and death in diabetic individuals (Zhang et al., 2021). Of all diabetic complications, diabetic cardiomyopathy accounts for nearly 80% of death in diabetic individuals (Wang and Hill, 2015).
Pathologically, diabetic cardiomyopathy is reflected by cardiac hypertrophy, myocardial apoptosis and fibrosis, and cardiac dysfunction (Ernande et al., 2011;Zhang and Sun, 2020). Numerous molecular mechanisms have been suggested to be involved in the development and progression of diabetic cardiomyopathy, such as oxidative stress, nitrosative/nitrative stress, inflammatory response, mitochondrial dysfunction, endoplasmic reticulum stress, impaired autophagy, cardiomyocyte apoptosis and death, and diabetic microangiopathy (Bugger and Abel, 2014;Varga et al., 2015;Kaludercic and Di Lisa, 2020). Additionally, experimental studies have shown that cardiac glycolipid toxicity, impaired insulin signaling, cardiac energetic impairment, activation of the renin-angiotensin system (RAS) and increased formation of angiotensin II (Ang II), cardiac autonomic neuropathy, reduced NO bioavailability, elevations in AGEs, and dysregulated cardiomyocyte calcium handling, play central roles in diabetic cardiomyopathy pathologies ( Figure 1) (Pechánová et al., 2015;Jia et al., 2018a;Jia et al., 2018b;Dhalla and Shah, 2020;Komici et al., 2019;Zamora and Villena, 2019). Although our current understanding of the pathogenesis of diabetic cardiomyopathy is increasing, the etiology of diabetic cardiomyopathy is multifactorial and extremely complicated. There are still gaps in our current knowledge of the precise mechanisms involved in diabetic cardiomyopathy. As a result, more studies are required to deepen the comprehensive mechanisms that underlie the initiation and progression of diabetes-induced cardiac dysfunction, thereby leading to the development of newly effective targets or approaches to reduce the risk of diabetic patients developing life-changing complications, such as heart failure.

CHEMISTRY AND BIOCHEMISTRY OF H 2 S
Previously, H 2 S is taken as a poisonous and occasionally lethal gas, and it might represent an industrial safety hazard since this colorless gas could be formed from the decomposition of various organic materials (Perna et al., 2011). Because of a special characteristic of rotten smell, H 2 S is believed to be a toxic byproduct of microbial metabolism in the atmosphere (Pan et al., 2017). H 2 S could be detected by the human nose at a level of 0.1 ppm, a dose of 400-fold lower than its toxic concentrations (Wang, 2002), while long-term exposure can render desensitization of the olfactory nerve to H 2 S . H 2 S is readily dissolved in water due to its weak acid properties (pKa at 37°C, 6.76), and it will split to generate two dissociation states; the hydrosulphide anion (HS − , pKa 7.04), and sulphide anion (S 2− , pKa 11.96) . As a consequence, approximately 18.5% of the total sulphide exists as the undissociated acid and 81.5% as the HS − and S 2− at physiological pH of 7.4 in aqueous solution (Dombkowski et al., 2004), which are in dynamic equilibrium in the following sequential reactions: H 2 S#HS − +H + #S 2− + 2H + . Actually, the three equilibria in this formula represent the real dynamics of H 2 S in solution. It can be easily predicted that this equilibria will continuously shift to the left in an open system according to Le Chatelier's Principle, thereby causing H 2 S to escape from the solution (Li and Lancaster, 2013). It has been shown that the half of H 2 S could be lost from solutions within 5 minutes in cell culture wells, 3 minutes in a bubbled tissue bath and an even shorter time in the Langendorff heart apparatus (DeLeon et al., 2012). Under aerobic conditions, the half-life time of H 2 S is 2.0, 2.8, and 10.0 min in murine hepatic, renal, and brain homogenates, respectively (Table 1) (Vitvitsky et al., 2012). One should also be aware that the actual concentration of H 2 S might vary in experimental systems containing headspace, and this may also depict to some extent the remarkable variations in H 2 S concentrations in cells, tissues or plasma (Furne et al., 2008;Whitfield et al., 2008;Shen et al., 2011;Kolluru et al., 2013). Whist it remains elusive whether the biological functions of H 2 S are mediated by H 2 S itself or its derived species, including S 2− or HS − , FIGURE 1 | A proposed schematic diagram showing the signaling pathways involved in diabetic cardiomyopathy-related cardiac dysfunction. Hyperglycemia induces ROS overproduction and causes glucotoxicity and inflammation in the heart, and activation of PKC promotes AGEs formation, a critical event involved in diabetes-induced cardiac fibrosis. In cardiomyocytes, high glucose inhibits GLUT expression and glucose uptake, but upregulates PPARα to reduce glucose oxidation, thus resulting in insulin signaling impairment. Moreover, hyperlipidemia stimulates FA uptake and oxidation, and renders cardiomyocytes more sensitive to lipotoxicity. All these above changes synergistically contribute to fibrosis, apoptosis, mitochondrial dysfunction, and hypertrophy during the development and progression of diabetic cardiomyopathy. AGE: advanced glycation end products; FA: fatty acids; GLUT: Glucose transporters; PKC: Protein kinase C; PPARα: Peroxisome proliferator activated receptor α; ROS: reactive oxygen species. it is being accepted that HS− is a potential nucleophile that could react with different electrophilic cellular targets, such as nitrogen species (RSONS) (Hartle and Pluth, 2016). H 2 S is a well-known reducing agent that exhibits various chemical properties, such as its redox activity, acidity, and high nucleophilicity, allowing for reactions with multiple cellular targets (Hartle and Pluth, 2016).
Although it is quite conceivable that the biochemical effects of H 2 S may be dependent on its chemical capacities (especially the reductive and nucleophilic properties), disentangling this chemistry seems to be more complicated (Li and Lancaster, 2013;Nagy et al., 2014). H 2 S is biologically active since its highly lipophilic characteristics allow it freely to penetrate into all types of the cell membranes without facilitation of membrane channels (Mathai et al., 2009). A famous pioneer in H 2 S research, Hideo Kimura, had demonstrated that H 2 S facilitated longterm potentiation in hippocampal tissues at the physiological concentration, pointing this gasotransmitter as an important neuromodulator (Abe and Kimura, 1996). Later, H 2 S is found to regulate both physiological and pathophysiological processes through a wide spectrum of signaling molecules (Kimura, 2020), such as reacting with superoxide anions (Chang et al., 2008), hypochlorite , hydrogen peroxide (Geng et al., 2004), peroxynitrite (Whiteman et al., 2004), metals (Li and Lancaster, 2013), thiol derivatives (Li and Lancaster, 2013), and NO (Ali et al., 2006;Whiteman et al., 2006) (Table 2). It is highly probable that the list of biomolecules affected by H 2 S will grow rapidly in the near future.
Since the discovery of H 2 S production in mammalian cells, many scientists have been working on the biological functions of H 2 S in this emerging field. The critical roles of H 2 S at its physiologically relevant concentrations on cardiovascular homeostasis have been well documented (Pan et al., 2017). Endogenous H 2 S levels are regulated preciously by its generation and elimination in health . It is important to determine the concentrations of H 2 S in cells, tissues, and blood samples before its prosperity in the regulation of physiological/pathophysiological functions. To date, several analytical methods are achieved to quantify H 2 S concentrations in organ tissues and blood, such as fluorescent probes (Zhang et al., 2020b), colorimetry (Sugahara et al., 2016), spectrophotometric analysis (Antoniou et al., 2018), headspace gas determination (Fonseca et al., 2013), polarographic sensor (Doeller et al., 2005), and liquid chromatography-mass spectrometry (Tan et al., 2017). However, different analysis methods have obtained inconsistent range of H 2 S concentrations (Ibrahim et al., 2021). According to the current literatures, the physiological concentrations of H 2 S in human/animal blood or tissues range from 15 nM to 300 μM in vivo (Huang et al., 2015;Hackfort and Mishra, 2016;Pan et al., 2017). Likewise, the physiological concentrations of H 2 S in plasma, serum, and cardiac tissues are considerably inconsistent ( Table 3). The low concentration, short half-life time with fast catabolism, and high reactivity of H 2 S might be challenging for the determination of endogenous H 2 S in vivo (Ibrahim et al., 2021). Additionally, the wide range of H 2 S concentrations might result from variable detection methods as the different analysis technologies have distinct disadvantages, such as low sensitivity and specificity, complex preparation processes, and timeconsuming procedures Wu and Hu, 2018). The striking inconsistency in H 2 S levels under physiological conditions may lead to uncertainty for the exact mechanistic roles of H 2 S in physiological processes. Thence, it is essential to improve detection specificity and lower threshold limitations of these techniques, which may be helpful to obtain the actual physiological compartmentalization of H 2 S in cells, blood, and tissue.

ENZYMATIC PRODUCTION AND METABOLISM OF H 2 S
Endogenous H 2 S in mammalian systems is primarily produced by enzymatic or non-enzymatic pathways (Yu et al., 2014). The enzymatic pathways that generate H 2 S are mediated by CBS, CSE, and the coupling of 3-MST with CAT. The enzymatic pathways of H 2 S formation are illustrated in Figure 2. The intracellular distributions of H 2 S-generating enzymes are different in which CBS and CSE are strictly localized in the cytoplasm, whereas 3-MST and CAT are expressed in both the mitochondrial and the cytosol (but much higher in mitochondria) (Pan et al., 2017). Regarding tissue specificity localization, CSE is abundantly expressed in the cardiovascular system, whereas CBS expression predominates in the brain, liver, and kidneys under normal physiological conditions (Pan et al., 2017;Sun et al., 2019b). 3-MST and CAT are responsible for H 2 S formation in the brain and vascular endothelium (Shibuya et al., 2009;Chen et al., 2015). In spite of these observations, the relative contribution of each of these three enzymes to the tissue or circulating H 2 S levels remains elusive (Donnarumma et al., 2017). In addition to enzymatic pathways, H 2 S is also produced from inorganic polysulfides, organic polysulfides, and elemental sulfur, which are enriched in garlic (Benavides et al., 2007). Therefore, the healthy benefits of garlic may be associated with H 2 S production (Rose et al., 2018;Rodrigues and Percival, 2019). Actually, garlic extracts are effectively slowing the development and progression of atherosclerosis through the generation of H 2 S from S-allylcysteine and S-allylmercaptocysteine (Benavides et al., 2007;Budoff et al., 2009).
The balance of synthesis and metabolism is essential for maintaining an appropriate concentration of H 2 S under physiological conditions. Endogenous H 2 S could be inactivated by several routes (Figure 3). First of all, H 2 S can be converted into thiosulfate through mitochondrial oxidative modification, thiosulfate is further transformed into sulfite and sulfate (Kimura, 2012). Cytosolic methylation is the second pathway to convert H 2 S to dimethylsulfide through thiol S-methyltransferase (Wang et al., Ali et al. (2006), Whiteman et al. (2006) Frontiers in Pharmacology | www.frontiersin.org October 2021 | Volume 12 | Article 651884 4 2020). Also, binding of hemoglobin to H 2 S contributes to the generation of sulfhemoglobin (Olson et al., 2014). The excessive H 2 S could be scavenged by metallo-or disulfide-containing molecules or glutathione disulfide, and could also be released from the lungs (Mani et al., 2014;Donnarumma et al., 2017;Rose et al., 2017;Cao et al., 2019). Although these biosynthesis FIGURE 2 | Endogenous synthesis pathways of H2S. CSE produces H 2 S by using L-cysteine as the main substrate, accompanied by the generation of ammonia, pyruvate, and thiocysteine. The latter would be catalyzed into H 2 S and cysteine. CBS-mediated H 2 S production is dependent on the transformation of homocysteine and L-cysteine to H 2 S and cystathionine. CAT catalyzes the conversion of L-cysteine and α-ketoglutarate to 3-MP and L-glutamate, and 3-MP is then catalyzed to sulfurous acid, pyruvate and thiosulfate by 3-MST. Thiosulfate is then reduced to H 2 S and glutathione disulfide by using reduced glutathione. Moreover, D-cysteine could be converted into 3-MP by DAO, and this reaction is also responsible for the formation of H 2 S in the mitochondria. 3-MP: 3-mercaptopyruvate; 3-MST, 3-mercaptopyruvate sulfurtransferase; CAT, cysteineamino transferase; CBS, cystathionine β-synthase; CSE, cystathionine c-lyase; DAO, D-amino acid oxidase; H 2 S: hydrogen sulfide.
Frontiers in Pharmacology | www.frontiersin.org October 2021 | Volume 12 | Article 651884 5 and degradation pathways of H 2 S have only recently been identified, these findings will undoubtedly promote the clinical translational research of this gaseous transmitter in future studies. Recently, the physiological significance of H 2 S metabolites, such as persulfides, polysulfides, and other reactive sulfur species is emerging (Koning et al., 2015). Moreover, H 2 S and reactive sulfur species could coexist in biological system as they are normally interchangeable (Bolton et al., 2019). It is believed that reactive sulfur species might be responsible for at least some biological activities of H 2 S (Shimamoto and Hanaoka, 2015). However, the comparisons of the protective effects of H 2 S and its metabolites on cells or tissues remain unclear, which merits further research.
The potential of H 2 S metabolites as biomarkers is being appreciated since the levels of H 2 S metabolites in serum and urine may reflect renal disease severity, such as chronic kidney disease (Nakanishi et al., 2002;Koning et al., 2015). In chronic heart failure patients, plasma sulfate concentrations tend to be higher, and sulfate clearance is associated with favorable disease outcome (Koning et al., 2017). Although these observations are promising, it is clear that the recognition of H 2 S metabolites as disease biomarkers is not flawless. Meanwhile, the quantification of H 2 S metabolites in the ischemia-reperfusion conditions or in other cardiovascular debacles has yet to be fully elucidated. Therefore, large-population studies investigating the potential value of H 2 S metabolites as disease biomarkers are underway. Overall, our current understanding of H 2 S metabolites and how their dysfunction contributes to cardiovascular pathogenesis remains largely unclear, thereby requiring more studies.

PHARMACOLOGY AND TOXICOLOGY OF H 2 S
It has been well documented that H 2 S is a crucial signaling molecule in the cardiovascular system because of its physiological and pathological significances in cardiovascular homeostasis Pan et al., 2011;Sun et al., 2011;Liu et al., 2012;Pan et al., 2012;Wang et al., 2015b;Nagpure and Bian, 2016). Plasma and tissue levels of H 2 S are sophisticatedly controlled by its formation and catabolism in health (Elsey et al., 2010;Huang et al., 2015). Under pathological conditions, endogenous H 2 S levels and H 2 S-prodocuing enzyme activities/ expressions are significantly altered (Pan et al., 2017;Sun et al., 2019a;Sun et al., 2019b). Both experimental and clinical evidence has revealed that perturbation of endogenous H 2 S system is obviously linked with the pathologies of cardiovascular disorders, including atherosclerosis, hypertension, diabetic cardiomyopathy, myocardial ischemia/reperfusion injury, endothelial dysfunction, and heart failure (Jiang et al., 2005;Wang et al., 2015c;Gao et al., 2015;Meng et al., 2015;Yang and Wang, 2015;Donnarumma et al., 2017;Kanagy et al., 2017;Li et al., 2017;Kar et al., 2019a). In renal physiology, H 2 S induces vasodilation, and increases renal blood flow and glomerular filtration rate, leading to an indirect increase of the urinary excretion of Na + and K + (Roy et al., 2012). H 2 S exhibits an inhibitory effect on specific Na + and K + transporters in the kidney, thus further increasing the excretion of such electrolyte into the urine (Xia et al., 2009). H 2 S acts as an oxygen sensor in the renal system, especially in medulla (Bełtowski, 2010). Also, H 2 S is found to inhibit renin release in rat models of renovascular hypertension (Lu et al., 2010). Hypertension-related nephropathy, a result of long-term hypertension, is the second leading cause of chronic kidney disease world (Hart and Bakris, 2010). The blood pressure lowering effects of exogenous H 2 S donors have been demonstrated in spontaneous hypertensive rats (Ahmad et al., 2014;Li et al., 2019a), angiotensin II-induced hypertension (Al-Magableh et al., 2015;Hsu and Tain, 2018;Chi et al., 2019), Nwnitro-L-argininemethyl ester (L-NAME)-induced hypertension (Ji et al., 2014;Jin et al., 2017), and renovascular hypertension (Guo et al., 2017a;Feng et al., 2020). It is reasonable that renal protective effects of H 2 S are observed in these hypertensive FIGURE 3 | The proposed rotes for H2S degradation. With the aid of thiol S-methyltransferase, H 2 S is methylated into dimethylsulfide. H 2 S could be oxidized to thiosulfate, coupled with the production of sulfite and sulfate. Hemoglobin interacts with H 2 S, resulting in the generation of sulfhemoglobin. H 2 S can be released from the mammalian lungs. H 2 S, hydrogen sulfide.
Frontiers in Pharmacology | www.frontiersin.org October 2021 | Volume 12 | Article 651884 6 models. Subsequent studies have suggested that suppression of ROS formation and epithelial sodium channel, upregulation of vascular endothelial growth factor might mediate renal protective effects of H2S in hypertension Ahmad et al., 2014;Holwerda et al., 2014;Wang et al., 2015a). Taken together, H 2 S serves as an ideal candidate for the prevention and treatment of hypertension-related renal damage.
The dose of H 2 S donor, sodium hydrosulfide (NaHS), ranging from 1.4 mg/kg/d to 56 mg/kg/d was used in animals, but 1 μM to 10 mM in cells . Similarly, the concentration of other H 2 S donors varies greatly in both animal and cell experiments . Here, it is thought-provoking to determine the exact pharmacological doses of H 2 S that are biologically effective or detrimental. Solving this problem may help to answer whether it is appropriate to use a dissolved exogenous H 2 S solution as a tool to determine the potential physiological function of endogenous H 2 S and its putative therapeutic applications. One should keep in mind that high micro-moles or even milli-moles solutions of dissolved H 2 S in organs or milieubathing cells have been repeatedly used in the literatures (Szabó, 2007;Nicholson and Calvert, 2010;Szabó et al., 2011). However, the solutions of dissolved H 2 S with high micromoles or even milli-moles in vitro are above the concentrations of dissolved H 2 S found in blood and tissues during lethal H 2 S exposure in vivo. In addition, the level of toxicity varies greatly between different cell types, thus the physiological relevance of the data obtained after local or in vitro administration of H 2 S at concentrations of micro-moles or milli-moles is far from uncertain. Also, H 2 S is one of the most toxic mitochondrial poisons, even more toxic than cyanide on a mole-to-mole basis (Cooper and Brown, 2008). The activity of mitochondrial cytochrome c oxidase is diminished by a solution of dissolved H 2 S at a concentration ranging from 10 to 30 μM (Yong and Searcy, 2001;Leschelle et al., 2005). In vivo studies have shown that a severe depression in respiratory medullary neurons and/or cardiac contractility is observed in rodents and large mammals by infusing or inhaling H 2 S at levels yielding blood levels of gaseous H 2 S between 2 and 5 μM Haouzi et al., 2014;Sonobe and Haouzi, 2016). A depression in cardiac contractility is produced by a solutions of H 2 S at concentrations above 50 μM on an isolated heart or cardiomyocytess (Geng et al., 2004;Sun et al., 2008;Yong et al., 2008). It is likely that this decrease in cardiac contractility is pathological and toxic at both the level of individual cardiac cells and the whole heart when exogenous H 2 S concentration reaches higher than the phyiological levels (Judenherc- Haouzi et al., 2016). In all, few micro-moles concentrations of dissolved H 2 S is toxic to most tissues in vivo, impacting vital respiratory and cardio-vascular functions (Haouzi, 2016). However, it seems that high levels of H 2 S are still administrated in the past, present and future research at the animal and cellular levels.
To date, it is merited to answer why the toxic levels of H 2 S are used in physiological studies. The rationality of using high levels (i.e., micro-moles or even milli-moles) of H 2 S to study and predict its physiological effects depends on the initial reports, where the concentration of H 2 S in blood or tissue is at high micromolar in the blood or tissues (Furne et al., 2008). Philippe Haouzi has reviewed that two main reasons could explain why high micro-moles of "endogenous" H 2 S are found in the blood and tissues (Haouzi, 2016). The first reason is related to the nature of the pools of sulfide present in the blood and in tissues, and the second reason for this error is linked with the methodology of measurement of the pool of diffusible/dissolved H 2 S in a biological milieu (Van de Louw and Haouzi, 2012). The subsequent studies have found that low levels of H 2 S donors in vivo are unlikely to increase the concentrations of H 2 S in tissues because of a fact that the majority of H 2 S in the blood after intravenous or intraperitoneal injections in vivo is immediately oxidized or combined with metallo-proteins (hemoglobin), dropping the level of free H 2 S to almost zero, unless the lethal levels are applied (Insko et al., 2009;Wintner et al., 2010;Haouzi and Klingerman, 2013;Klingerman et al., 2013). There is a very small margin between the concentration that does not produce an effect and the concentration that can kill. As reviewed by Szabo et al. (2011), the variable concentrations of H 2 S donors might result in the toxic levels or without chaging H 2 S concentrations in blood and tissues (Szabó, 2007). A growing number of studies support the "physiological" and protective roles of H 2 S by using solution of free H 2 S in the high micro-moles levels that are only found during a lethal in toxication in vivo . As such, high micro-moles or milli-moles levels of H 2 S donors are still used to elucidate the properties of endogenous H 2 S in those ongoing studies. In all, H 2 S donors regulate a multitude of biological processes at the typically used doses (so-called physiological effect), which should belong to the field of toxicology in many cases . Future studies are necessary to quantify the actual H 2 S concentration of each exogenous H 2 S donor used in the tissues and determine what effects are expected to be produced in keeping with sulfide toxicity and its rapid disappearance in the body or solutions.

H 2 S: A Potential Cardioprotective Agent Against Diabetic Cardiomyopathy
Plasma H 2 S levels are remarkably decreased in diabetic patients and animals (King and Lefer, 2011;Guo et al., 2017b;Kesherwani et al., 2017;Suzuki et al., 2017). The levels of H 2 S in plasma, heart tissues, and urine tend to be lower in aging diabetic mice when compared to controls (Jin et al., 2015). Likewise, the myocardium expressions of H 2 S-generating enzymes, CSE and 3-MST, are significantly decreased in the heart samples of diabetic mice, whereas CBS levels are significantly elevated (Jin et al., 2015). Both circulating and myocardium levels of H 2 S are also reduced in high fat diet (HFD)-induced obese diabetic mice (Barr et al., 2015). In streptozotocin (STZ)-induced type 1 diabetic rats, a significant decrease in myocardial H 2 S concentration is also observed (Palamarchuk et al., 2020). These findings characterize a critical role for endogenous H 2 S dysfunction in the etiology of diabetic cardiomyopathy.
Induction of endogenous H 2 S production is also beneficial for diabetic cardiomyopathy. Diallyl trisulfide is reported to afford a protection against high glucose-provoked cardiomyocyte apoptosis by stimulating the formation of CSE-derived H 2 S (Tsai et al., 2015). Metformin, one of the most widely prescribed insulin sensitizer, might benefit diabetic cardiomyopathy by enhancing H 2 S production in the heart (Wiliński et al., 2013). Phosphodiesterase 5 (PDE5) inhibitors have powerful protective effects against diabetic cardiomyopathy, and this may be dependent on H 2 S generation (Das et al., 2015). Exercise training mitigates HFD-induced diabetic cardiomyopathy via promoting cardiac H 2 S biosynthesis and subsequent prevention of cardiomyocyte pyroptosis (Kar et al., 2019b). Pretreatment with curcumin alleviates pathological morphological damage in myocardium tissues from diabetic rats, and the beneficial effects of curcumin might be related with increased myocardium CSE and H 2 S levels (Tong et al., 2018). These exciting findings suggest the cardioprotective actions of H 2 S-based therapeutics in preclinical models of diabetic cardiomyopathy. Modulation of endogenous H 2 S production or application of exogenous H 2 S donors might serve as promising therapies for the management of diabetic cardiomyopathy. Later, we will provide a comprehensive overview on the critical molecular/cellular mechanisms that mediate H 2 S-induced favorable effects on diabetic cardiomyopathy.
Emerging evidence suggests that

STZ-induced diabetic rats
Blockade of endoplasmic reticulum stress and inhibition of myocardial hypertrophy/fibrosis Li et al. (2016) NaHS ( These documents were searched using Pubmed up to May 2021 with the search terms: (diabetic, diabetes, hyperglycemia, OR high glucose) AND (heart OR cardiomyopathy). After that, the titles and abstracts of all possible relevant papers were screened.
Frontiers in Pharmacology | www.frontiersin.org October 2021 | Volume 12 | Article 651884 8 H 2 S-mediated signaling transduction is correlated with production of NO (Paul and Snyder, 2012;Polhemus and Lefer, 2014;Sun et al., 2020b). Thus, the current roles of interactive functions of H 2 S and NO in diabetic cardiomyopathies are also elaborated in this review.

INHIBITION OF MYOCARDIAL FIBROSIS AND HYPERTROPHY
A previous study has found that administration of GYY4137 diminishes rat neonatal cardiac fibroblast migration exposed to high glucose (Zheng et al., 2015). The underlying mechanisms may involve inhibition of reactive oxygen species (ROS)/ERK1/2 signaling, indicating that H 2 S antagonizes cardiac fibrosis in diabetic cardiomyopathy by regulating redox homeostasis (Zheng et al., 2015). Consistent with this finding, NaHS prevents cardiac hypertrophy/fibrosis and ameliorates left ventricular dysfunction in STZ-induced diabetic rats by activating the nuclear factor erythroid 2-related factor 2 (Nrf2)/antioxidant response element (ARE) signaling pathway, accompanied by upregulations of antioxidant proteins haem oxygenase-1 (HO-1) and Nad(p)h: quinone oxidoreductase 1 (NQO1) in myocardium tissues from diabetic rats (Zhou et al., 2015). Exogenous H 2 S reverses myocardial hypertrophy and fibrosis through suppressing cardiac levels of myosin heavy chain 6 and myosin light chain 2 ubiquitination in db/db mice (Sun et al., 2020d). Compared with diabetic rats, the left ventricular hemodynamic parameters and myocardial ultrastructure changes (myocardial hypertrophy and fibrosis) are obviously normalized in diabetic animals treated with NaHS, which may be associated with upregulation of cardiac myosin light chain kinase (MLCK) and downregulation of serum creatine kinase MB isozyme (CK-MB) (Yang et al., 2016a).
The myocardial expressions of pro-fibrotic factors, including matrix metalloprotease 2 (MMP-2), tissue inhibitor of metalloproteinase 2 (TIMP-2), collagens, transforming growth factor (TGF)-β1/SMAD family member 3 (Smad3) signaling pathway, are strikingly changed in diabetic rats . However, an exogenous H 2 S donor, NaHS, ameliorates diabetes-induced myocardial fibrosis by inhibition of the Wnt and TGF-β1/Smad3 pathway and downregulation of myocardial collagen overproduction . Administration of NaHS for 8 weeks prevents myocardial hypertrophy and collagen These documents were searched using Pubmed up to May 2021 with the search terms: (diabetic, diabetes, hyperglycemia, OR high glucose) AND (heart OR cardiomyopathy). After that, the titles and abstracts of all possible relevant papers were screened.
Frontiers in Pharmacology | www.frontiersin.org October 2021 | Volume 12 | Article 651884 9 deposition in STZ-induced diabetic rats by suppression of endoplasmic reticulum stress . Inhibition of the protein kinase C (PKC)/extracellular regulated protein kinase 1/2 (ERK1/2) signaling pathway is required for H 2 S to ameliorate myocardial fibrosis in diabetic rats induced by STZ (Long et al., 2019). Liu et al. (2018) have found that the effects of NaHS on reversing diabetic myocardial fibrosis may be associated with blockade of the Janus kinase/signal transducer and activator of transcription (JAK/STAT) signaling pathway. Hematoxylin and eosin staining and Masson staining results have shown that NaHS is able to improve myocardial hypertrophy and fibrosis in diabetic rats induced by STZ, and the underlying mechanism may be dependent on activation of the phosphatidylinositol 3-kinase (PI3K)/Akt signaling pathway and following autophagy inhibition . These above observations evidently suggest that H 2 S donors function as an attractive alternative for the prevention and treatment of diabetesprovoked myocardial hypertrophy and fibrosis (Kang et al., 2020).

SUPPRESSION OF ENDOPLASMIC RETICULUM STRESS
Nowadays, the involvement of endoplasmic reticulum stress in the development and progression of diabetic cardiomyopathy is well accepted (Figure 4) (Xu et al., 2012). Given the important roles of endoplasmic reticulum stress in the pathologies of diabetic cardiomyopathy, inhibition of endoplasmic reticulum stress would be a therapeutic strategy for intervention of diabetic cardiomyopathy. Interestingly, multiple lines of evidence have proved that the protective effects of H 2 S against diabetic cardiomyopathy are related to inhibition of endoplasmic reticulum stress. For instance, pretreatment of AC16 cardiomyocytes with NaHS could retard palmitic acid-induced cardiac lipotoxicity and apoptosis, which is similar to the effects of an endoplasmic reticulum stress inhibitor 4-phenylbutyric acid (Guo et al., 2017b). The expressions of endoplasmic reticulum stress-associated proteins, including glucose-regulated protein (GRP78), C/EBP-homologous protein (CHOP), and caspase-12, are incremented in diabetic heart tissues, while H 2 S suppresses diabetes-induced endoplasmic reticulum stress and abrogates myocardial hypertrophy and myocardial collagen deposition in diabetic rats . In line with this, intraperitoneal injection of NaHS improvesc cardiac dysfunction and myocardial ultrastructure damage by suppressing the mRNA expressions of endoplasmic reticulum stress markers, such as CHOP, GRP78 and caspase 12 (Yang et al., 2016b). Furthermore, the upregulated expressions of endoplasmic reticulum stressrelated proteins and mitochondrial apoptotic proteins in diabetic cardiac tissues are reduced to normal levels in the presence of NaHS (Yang et al., 2017a).
Deletion of endogenous H 2 S is critically involved in diabetesinduced cardiac damage, whereas exogenous H 2 S donor NaHS improves diabetic cardiomyopathy though inhibition of endoplasmic reticulum stress (Guo et al., 2017b). Molecular experiments have revealed that H 2 S therapy (SG-1002) is capable of relieving HFD-induced cardiomyopathy via suppressing cardiac endoplasmic reticulum stress and restoring cardiac adiponectin levels (Barr et al., 2015). These above findings indicate that H 2 S donors are postulated to serve as novel cardioprotective agents for diabetic cardiomyopathy due to its strong suppression of endoplasmic reticulum stress-dependent apoptosis pathway in the heart.

PROTEIN S-SULFHYDRATION
S-sulfhydration (also called S-persulfidation or Persulfidation), a newly post-translational modification by generating a hydropersulfide moiety (-SSH) or polysulfide in specific cysteine residues of target proteins, is discovered to mediate most of the cellular functions induced by H 2 S (Paul and Snyder, 2012;Paul and Snyder, 2015;Zhang et al., 2017;Sun et al., 2021). Accordingly, several S-sulfhydrated proteins are involved in H 2 S-mediated cardiovascular benefits in diabetes. HMG-CoA reductase degradation protein (Hrd1), an endoplasmic reticulum transmembrane E3 ubiquitin ligase, is vitally involved in ubiquitylation of various substrate proteins and subsequent protein trafficking (Doroudgar et al., 2015). Vesicleassociated membrane protein 3 (VAMP3) is predominantly expressed in the heart and plays a key role in controlling intracellular membrane trafficking and exocytosis (Haglund and Dikic, 2012). Emerging evidence has shown that triacylglycerol accumulation within the myocardium may be responsible for the development of diabetes-induced heart failure, and this process is mainly modulated by transport protein fatty acid translocase CD36 (Chandler et al., 2003;Coort et al., 2004). VAMP3 contributes to CD36 exocytosis in cardiomyocytes, thus leading to increased myocardial long-chain fatty acid uptake and triacylglycerol accumulation (Pulido et al., 2011). On these grounds, Zhang and colleagues hypothesized that H 2 S induced VAMP3 degradation through Hrd1 S-sulfhydration and prevented CD36 translocation, thereby alleviating lipid toxicity in the heart tissues of db/db diabetic mice . As expected, the authors have demonstrated that exogenous H 2 S restores Hrd1 expression by inducing its S-sulfhydration at Cys115 . VAMP3 ubiquitylation and degradation induced by H 2 S is able to delay CD36 exocytosis in cardiomyocytes and reduce droplet formation in the heart tissues from db/db mice . This study leads to a novel finding that H 2 S upregulates Hrd1 expression through its S-sulfhydration at Cys115 and results in VAMP3 ubiquitylation and the stability of membrane CD36 expression in cardiomyocytes, which is necessary to prevent myocardial longchain fatty acid uptake and lipid droplet formation in diabetic heart tissues . Afterwards, the same group has identified that H 2 S promotes the S-sulfhydration of ubiquitin specific peptidase 8 (USP8) and muscle RING finger-1 (MuRF1), thereby preventing cardiac structural injury in diabetes (Sun et al., 2020d;Sun et al., 2020e). These observations suggest that protein S-sulfhydration is believed to participate in the protective roles of H 2 S donors in diabetes-induced cardiomyopathy. In this regard, more S-sulfhydrated proteins will be identified to be responsible for the therapeutic effects of H 2 S in the context of diabetic cardiomyopathy.

INHIBITION OF CARDIOMYOCYTE APOPTOSIS, OXIDATIVE STRESS, AND INFLAMMATION
Evidence from cell cultures, animal models, and clinical studies has confirmed that H 2 S grants an anti-diabetic cardiomyopathy characteristic through anti-inflammation, anti-apoptosis, antioxidant stress pathways (Powell et al., 2018). Pretreatment of H9c2 cells with NaHS obviously counteracts high glucoseinduced cardiac cell apoptosis and inflammation, as evidenced by measurement of caspase-3, inducible nitric oxide synthase (iNOS), cyclooxygenase 2 (COX-2), interleukin 1β (IL-1β), and IL-6 (Huang et al., 2016a). Inhibition of p38 MAPK/NF-κB signaling might mediate these cytoprotective effects of H 2 S in diabetes-induced cardiac apoptosis and inflammation (Huang et al., 2016a). Moreover, NaHS inactivated the toll-like receptor 4 (TLR4)/nuclear factor kappa-B (NF-κB) pathway to attenuate high glucose-induced nucleotide-binding oligomerization domain-like receptor protein 3 (NLRP3) inflammasome activation and cardiotoxicity in H9c2 cells (Huang et al., 2016b). In addition, inactivation of Wnt/β-catenin signaling pathway and the signal transducer and activator of transcription 3 (STAT3)/hypoxia-inducible factor-1α (HIF-1α) pathway, upregulations of HO-1 and NQO1, induction of forkhead box O1 (FoxO1) phosphorylation and nuclear exclusion, are proposed to be main mechanisms for H 2 S donors to attenuate hyperglycemia-induced myocardial apoptosis and oxidative stress (Ye et al., 2018;Li et al., 2019b;Zhang and Ye, 2019). Moreover, H 2 S donor NaHS is reported to lessen hyperglycemiainduced inflammation, apoptosis, and oxidative stress in cardiac tissues by inhibiting c-Jun N-terminal kinase (JNK)/p38 mitogenactivated protein kinase (MAPK) pathways and activating PI3K/ Akt signaling (Zhou et al., 2015). In the heart tissues, hyperglycemia significantly induces cardiac dysfunction with concomitant increases in redox perturbation and inflammatory reactions (Jia et al., 2020). However, treatment with NaHS significantly ameliorates these abnormalities, whereas treatment with propargylglycine (a specific inhibitor of H 2 S) further aggravated such alterations (Jia et al., 2020). These studies highlight the protective roles of H 2 S in diabetesinduced cardiac damage, and its potential mechanism may involve the negative regulation of cardiomyocyte apoptosis, oxidative stress, and inflammation.
Different from apoptosis, pyroptosis is a special pattern of cell death whereby the canonical pathway is reliant on caspase-1 and the noncanonical pathway is specifically dependent on caspase-11 (Bergsbaken et al., 2009). Initially found in monocytes and macrophages, pyroptosis is recently found to be involved in pathological inflammation-related cell death (Miao et al., 2010). NLRP3 inflammasome activates caspase-1 and results in cell swelling and lysis, and consecutive release of IL-1β and IL-18, characteristics of pyroptosis (Bergsbaken et al., 2009). Pyroptosis and its downstream inflammatory cytokines are closely linked with cardiovascular diseases, including diabetic cardiomyopathy (Kar et al., 2019a). NLRP3-dependent pyroptosis is detected in diabetic heart tissues, and it participates in the pathogenesis of diabetic cardiomyopathy (Luo et al., 2014;Luo et al., 2017). As a cardioprotective gaseous signaling molecule, H 2 S donor treatment suppresses caspase-1 activity and caspase-1 recruitment via the apoptosis-associated speck-like protein containing A CARD (ASC) adapter protein in a rat model of myocardial ischemic and inflammatory injury (Toldo et al., 2014). Recently, induction of H 2 S biosynthesis by exercise ameliorates HFD-induced cardiac structural and metabolic by inhibition of pyroptotic signaling (Kar et al., 2019b). However, further evidence is required to extrapolate the benefits of H 2 S donors-induced inhibition of pyroptosis on diabetic cardiomyopathy.

Inhibition of the Leptin-p38 Mitogen Activated Protein Kinase Pathway
Leptin, a 16-kD hormone primarily secreted by adipocytes, exhibits a wide range of biological activities through binding to the leptin receptors (Friedman and Halaas, 1998). It is found that diabetes-induced cardiomyocyte hypertrophy may be mediated by leptin and its receptors (Majumdar et al., 2009). Aberrant expression of p38 MAPK is detected in the heart tissues from diabetic animals, and inhibition of p38 MAPK markedly Frontiers in Pharmacology | www.frontiersin.org October 2021 | Volume 12 | Article 651884 delays the progression of diabetic cardiomyopathy (Zuo et al., 2019). In light of its capacity to facilitate diabetic cardiomyopathy development, p38 MAPK serves as an attractive target for the treatment of diabetic cardiomyopathy . Indeed, inhibition of the p38 MAPK pathway is responsible for H 2 S-mediated protective effects against high glucose-or oxorubicin-induced cardiotoxicity in H9c2 cells Xu et al., 2013). Venkatesh and coworker have confirmed that leptin and its downstream p38 MAPK signaling pathway contribute to endothelin-1-and angiotensin-II-induced cardiomyocyte hypertrophy (Rajapurohitam et al., 2012). However, it remains to be investigated whether the leptin-p38 MAPK pathway contributed to high glucose-caused cardiomyocyte insult, and whether the pharmacological actions of H 2 S are related to this signaling pathway. To answer this, Liao' group measured the leptin, leptin receptors, and p38 MAPK expression levels in high glucose-challenged cardiomyoblasts . Results have shown that the expression levels of leptin and its receptors are significantly enhanced in cardiomyoblasts (H9c2 cells) upon exposure to high glucose (35 mM) for 24 h, while these increases are markedly attenuated by pretreatment with NaHS (400 μM), a donor for H 2 S . Additionally, high glucoseinduced increased phosphorylated p38 MAPK expression is reversed by either NaHS or leptin antagonist, suggesting the involvement of the leptin-p38 MAPK axis in high glucoseelicited cardiomyoblast injury . Importantly, H 2 S-mediated inactivation of the leptin-p38 MAPK pathway mitigates high glucose-evoked apoptosis, ROS generation, and mitochondrial membrane potential loss in H9c2 cells . Altogether, this study indicates that exogenous H₂S donors protect H9c2 cells against high glucoseinduced injury at least in part by inactivating the leptin-p38 MAPK signaling pathway.

Crosstalk With NO
A plethora of studies have revealed the physiological and/or pathological roles of H 2 S and NO in various systems during the last several decades (Lee and Yen, 2009;El-Mas et al., 2012;Nagpure and Bian, 2016;Magierowski et al., 2018;Wu and Hu, 2018;Mukherjee and Corpas, 2020;Xuan et al., 2020). Importantly, studies on the interaction of H 2 S and NO in diabetic cardiomyopathy are gradually emerging.
Several studies have disclosed that H 2 S regulates NO activities in the cardiovascular system (Zhang et al., 2020a), suggesting that H 2 S might exhibit cardioprotective effects via NO signaling. A phase I clinical trial has demonstrated that administration of a novel H 2 S prodrug, SG1002, increases H 2 S and NO levels in both healthy controls and heart failure patients (Polhemus et al., 2015). Exposure of endothelial cells with H 2 S stimulates NO production from endothelial nitric oxide synthase (eNOS) phosphorylation at Ser 1177 through an Akt-dependent mechanism (Predmore et al., 2011). CSE knockout mice are more susceptible to transverse aortic constriction as CSE-deficient mice display greater cardiac dilatation and dysfunction (Kondo et al., 2013). By contrast, cardiac-specific CSE overexpression or H 2 S therapy with SG-1002 restores cardiac structure and function after transverse aortic constriction through activation of the vascular endothelial growth factor-Akt-eNOS-NO-cGMP pathway (Kondo et al., 2013). Furthermore, CSE-deficient mice exhibit diminished eNOS activities, dysfunctional NO levels, and are more vulnerable to myocardial and hepatic ischemia reperfusion injury (King et al., 2014).
Acute H 2 S therapy improves myocardial and hepatic ischemia reperfusion injury by restoring eNOS function and NO bioavailability (King et al., 2014). NaHS treatment restores the impaired endothelial-dependent relaxation of the aorta in diabetic rats by a NOS-dependent mechanism as inhibition of NOS activity by L-NAME abolishes NaHS-mediated vasodilatation in diabetic aortas (Dorofeyeva et al., 2020a). These findings suggest that H 2 S-mediated cytoprotective signaling may be dependent in large part on eNOS-derived NO formation. As a pro-inflammatory factor and NO producer, iNOS express levels are upregulated in high glucoseincubated H9c2 cells, and H 2 S counteracts high glucose-induced inflammation and cytotoxicity via inhibition of iNOS expression (Huang et al., 2016a). In keeping with this, myocardial iNOS activities and expressions are significantly elevated in diabetic rats, which are suppressed by treatment with NaHS, a donor for H 2 S (Yang et al., 2017b). In sharp contrast, modulation of H 2 S synthesis stimulates eNOS activities and NO production in the heart tissues and aortic rings, thus recovering endotheliumdependent relaxation and arterial elastance, as well as cardiac dysfunction in diabetic rats (Dorofeyeva et al., 2021). Injection of STZ elicits reductions in myocardial CSE expression/activity, H 2 S and NO levels in rats, and moxonidine treatment abrogated cardiovascular dysfunction in diabetic rats through restoring cardiac H 2 S and NO levels (El-Sayed et al., 2016). These above findings imply that H 2 S confers a protective effect on the myocardium in diabetic animals, which may be associated with the regulations of iNOS, eNOS, and NO production.
It should be mentioned here that chemical interaction of H 2 S with NO might yield nitroxyl (HNO) (De Witt et al., 2001;Hrabie and Keefer, 2002;Cortese-Krott et al., 2014), one-electronreduced product of NO, and HNO is emerging as a potential pharmacological agent for various diseases by targeting numerous signal pathways (Irvine et al., 2008;Switzer et al., 2009;Sun et al., 2020a;Sun et al., 2020b). An interesting study has shown that CSE-produced H 2 S scavenges vascular NO levels in peripheral arteries and contributes to blood pressure regulation (Szijártó et al., 2018). Pretreatment of vascular rings from CSEdeficient mice with H 2 S donor ameliorates endothelial vasorelaxant response by decreasing NO levels and restoring HNO levels in arteries (Szijártó et al., 2018). Our group recently found that mixture of H 2 S donor NaHS and NO donor sodium nitroprusside (SNP) gives rise to HNO formation in cardiomyocytes (Sun et al., 2020c). HNO donor Angeli's salt significantly ameliorates hygerglycemia-induced myocardial apoptosis, hypertrophy, fibrosis, ROS generation, and cardiac performance in diabetic mice (Sun et al., 2020c). Induction of caveolin-3/eNOS complex may be required for HNO to ameliorate the development of diabetic cardiomyopathy (Sun et al., 2020c). In consistence with this finding, the left ventricular structural remodeling and diastolic Frontiers in Pharmacology | www.frontiersin.org October 2021 | Volume 12 | Article 651884 dysfunction are limited by chronic administration of HNO donor 1-nitrosocyclohexylacetate (1-NCA) in diabetic FVB/N mice (Cao et al., 2015). It is well known that NO resistance, a state in which the tissue responsiveness to exogenous and endogenous NO is impaired, is frequently occurred in cardiovascular disease states, including diabetic cardiomyopathy . In diabetic rats, acute NO effects induced by diethylamine-NONOate, including vasodilation, myocardial contraction and relaxation, are evidently impaired . In contrast, isopropylamine-NONOate-induced acute HNO effects are still preserved in the setting of diabetes, and the positive inotropic effects of HNO are obviously elevated in diabetic rat heart . This intriguing study provides a possible clue that HNO largely circumvents NO resistance in diabetic heart tissues, and supports a possible role for HNO in the management of ischemia and heart failure in diabetes. These above observations suggest that H 2 S and NO enter a redox reaction with each other, resulting in the formation of HNO, a potential way to antagonize the occurrence and development of diabetic cardiomyopathy. However, we can not be too optimistic on the preclinical studies of HNO in diabetic cardiomyopathy due to our insufficient understanding of HNO biosynthesis, chemistry, biology, and pharmacology. Thus, these aspects warrant further interrogation with the emphasis on HNO detection methods, long-lasting donors with safety and efficiency, pharmacological potentials beyond the cardiovascular system.

OTHER MECHANISMS
Adenosine 5′-monophosphate (AMP)-activated protein kinase (AMPK) is a metabolic master switch that play an indispensable role in energy metabolism and organic lipid homeostasis (Bai et al., 2018). Its activation exerts a beneficial effect in diabetic cardiomyopathy, while its inactivation is involved in cardiometabolic abnormalities, including diabetes-induced cardiac apoptosis, myocardial fibrosis, and ventricular hypertrophy (Haye et al., 2020). AMPK negatively regulates mammalian target of rapamycin (mTOR), and downregulation of mTOR mediates a host of biological effects induced by AMPK . A host of studies have highlighted the importance of AMPK inactivation in the pathogenesis of diabetic cardiomyopathy (Haye et al., 2020). On this basis, it is reasonable to hypothesize that H 2 S might protect cardiomyocytes from hyperglycemia stress through AMPKmediated inhibition of mTOR. Indeed, Wei et al. (2014) have ascertained that GYY4137, a novel H 2 S-releasing molecule, reverses high glucose-induced cell apoptosis, lactate dehydrogenase release, and mitochondrial membrane potential collapse. An AMPK activator, 5-amino-4-imidazolecarboxamide riboside (AICAR), has similar effects to GYY4137 treatment, whereas adenine 9-β-D-arabinofuranoside (an AMPK inhibitor) abolishes GYY4137-mediated cardiac benefits under diabetic conditions (Wei et al., 2014). Most importantly, both GYY4137 and AICAR stimulated AMPK phosphorylation and diminished mTOR phosphorylation in high glucose-challenged H9c2 cells, suggesting that H 2 S ameliorates diabetes-induced cardiac damage in an AMPK/ mTOR pathway-dependent manner (Wei et al., 2014).
The presence of cardiac dysfunction in diabetes is accompanied by overload of intracellular calcium in cardiomyocytes . When cardiomyocytes are excited, increased calcium ion will flow into the cells through L-type Ca 2+ channels, followed by the release of Ca 2+ from the sarcoplasmic reticulum (SR) via the ryanodine receptor (RyR2), leading to actin-myosin interaction. Cellular calcium overload is closely related with SR dysfunction, such as RyR2-FK506-binding proteins (FKBP12.6) disassociation, downregulation of sarcoplasmic reticulum Ca 2+ ATPase (SERCA2a) and calsequestrin 2 (CASQ2) (Cheng et al., 2012;Carneiro-Júnior et al., 2014;Roussel et al., 2015;Tomasova et al., 2015). In STZ-induced diabetic rats, the expressions of calciumhandling proteins in the SR, such as FKBP12.6, SERCA2a, and CASQ2 are downregulated, while the concentration of diastolic free calcium was elevated, indicating a calcium leak in diabetic heart tissues (Cheng et al., 2016). However, these abnormalities are attenuated in diabetic rats when treated with NaHS, indicating that H 2 S donors ameliorate diabetic cardiomyopathy-related imbalance in calcium homoeostasis by reversing abnormal expressions of disordered calcium-handling system in the SR (Cheng et al., 2016).
Autophagy plays a critical role in cellular homeostasis by clearing damaged or unnecessary organelles or macronutrients dependent on the lysosomal degradation pathway (Xie et al., 2011). Dysfunction of autophagy is shown to be involved in diabetic cardiomyopathy (Lavandero et al., 2015). It has been revealed that H 2 S is a regulator in autophagy during the process of diabetic cardiomyopathy. In type 2 db/db mouse heart tissues, NaHS facilitates autophagosome content degradation and promotes ubiquitin aggregate clearance via autophagy, which might exert its cardiovascular effect in diabetic cardiomyopathy . It is likely that H 2 S increases autophagic flux to ameliorate diabetic cardiomyopathy.

H 2 S PROTECTS AGAINST MYOCARDIAL ISCHEMIA/REPERFUSION INJURY IN DIABETES
Ischemia/reperfusion cardiac injury is recognized as one of the leading causes of people morbidity and mortality around the world (Yan et al., 2020). Primarily, ischemia/reperfusion leads to the death of cardiomyocytes and increased cardiac infarct size, which can be responsible for up to 50% of the final infarct size (Yellon and Hausenloy, 2007). It has been revealed that H 2 S donors could reduce myocardial ischemia/reperfusion injury by preserving mitochondrial functions and modulation of the signaling pathways of H 2 S by pharmacological agents is likely to prevent ischemia/reperfusion cardiac injury (Andreadou et al., 2020). In addition, the protective roles of H 2 S in myocardial ischaemia/reperfusion injury and cardioprotection by preconditioning or postconditioning have been reviewed (Andreadou et al., 2015). These published findings demonstrated that modulation of endogenous H 2 S production may be of clinical benefit in cardiac ischaemia/reperfusion injury. Mounting evidence has found that diabetic hearts are more susceptible to ischemic injury under ischemia, hypoxia or anoxia conditions (Hearse et al., 1975;Feuvray et al., 1979;Feuvray and Lopaschuk, 1997). This notion is further supported by clinical studies that diabetic patients complicated with coronary heart disease and post-myocardial infarction might have a worse prognosis with higher incidence of heart failure or sudden death (Jaffe et al., 1984;Gwilt et al., 1985;Stone et al., 1989;Saeid et al., 2018). Studies have shown that hyperglycemiainduced decrease in coronary collateral blood flow and impaired responses of coronary microcirculation to ischemia, might cooperatively lead to exacerbated cardiac dysfunction during myocardial ischemic states (Gu et al., 2003;Badalzadeh et al., 2015). It is likely that hyperglycemia might aggravate myocardial damage in the heart when combined other heart diseases, such as myocardial ischemia-reperfusion injury (Lejay et al., 2016). Interestingly, evidence is emerging that H 2 S is a strong protector against myocardial ischemia-reperfusion injury in diabetes (Jeddi et al., 2020). Therefore, in the sections below, we will provide a comprehensive overview of the therapeutic roles of H 2 S signaling in diabetes-aggravated myocardial infarction or myocardial ischemia-reperfusion injury.
Diabetes not only directly induces cardiac cardiomyopathy (a cardiac disease independent of coronary artery disease, hypertension, and hyperlipidemia), but also worsens cardiac damage in response to myocardial infarction or ischemia/ reperfusion (Schramm et al., 2008). To date, various pharmacological interventions are documented to attenuate myocardial ischemia/reperfusion injury in the setting of diabetes (Chin et al., 2017). Among them, H 2 S-based therapies have attracted tremendous attention because of its great potential in limiting cardiac injury during myocardial ischemiareperfusion under diabetic state (Donnarumma et al., 2017).
Previously, Bridgette et al. had sought to examine whether H 2 S granted a cardioprotection against ischemia-reperfusion injury in the setting of type 2 diabetes (Peake et al., 2013). In this work, they found that H 2 S donor sodium sulfide (Na 2 S) precondition for 24 h or 7 days all improved myocardial injury in db/db diabetic mice (Peake et al., 2013). In an effort to evaluate the signaling mechanism responsible for the observed cardioprotection, they demonstrated that exogenous H 2 S ameliorated myocardial ischemia-reperfusion injury in db/db diabetic mice by activating Nrf2 signaling in an ERK1/2-dependent manner (Peake et al., 2013). Later, the same group further validated that H 2 S provides cardioprotection against myocardialischemia reperfusion injury in db/db diabetic mice by activating the ERK1/2 arm of the Reperfusion Injury Salvage Kinase (RISK) pathway (Lambert et al., 2014). Subsequently, several groups have verified that H 2 S-associated protection against ischemia-reperfusion injury in diabetic heart may also involve eNOS-derived NO production (Jeddi et al., 2020), PI3K/ GSK3β pathway (Ansari and Kurian, 2020), and preservation of mitochondria (Ansari and Kurian, 2019;Mahalakshmi and Kurian, 2020). These findings provide ample evidence that H 2 S renders cardioprotection against ischemia/reperfusion injury in diabetes. However, more studies are needed to assess whether pre-and post-conditioning the heart with different H 2 S donors could behave differently in terms of ischemia-reperfusion injury limitation.

CLINICAL USE OF H 2 S DONORS IN DIABETES
The preclinical studies have provided a robust indication that delivery of H 2 S donors might represent an effective approach for the prevention and treatment of cardiomyopathy and myocardial ischemia/reperfusion injury under diabetes. These studies await further clinical translation. Until May 2021, a total of 15 clinical studies on H 2 S are found at clinicaltrials.gov after typing H 2 S in the search box of "Condition or disease" (https://clinicaltrials. gov/). These studies are performed to test the clinical utilities of H 2 S in various diseases, including chronic kidney disease (ClinicalTrials.gov Identifier: NCT01232257), hypertension (ClinicalTrials.gov Identifier: NCT03179163), heart failure (ClinicalTrials.gov Identifier: NCT01989208), myocardial Infarction (ClinicalTrials.gov Identifier: NCT03829605, NCT02899364), carotid stenosis (ClinicalTrials.gov Identifier: NCT03303534), peripheral arterial disease (ClinicalTrials.gov Identifier: NCT01407172), cardiovascular disease (ClinicalTrials.gov Identifier: NCT02180074). In a clinical trial (ClinicalTrials.gov Identifier: NCT01989208), the authors investigated the effects of H 2 S donor SG1002 on heart failure, and they found that administration of SG1002 increased H 2 S levels and circulating NO bioavailability, but attenuated BNP levels in heart failure patients (Polhemus et al., 2015). However, no results were posted regarding the effects of H 2 S on myocardial infarction, carotid stenosis, peripheral arterial disease, and hypertension. Likewise, the clinical value of H 2 S in diabetic cardiomyopathy and diabetes-aggravated cardiac ischemia/ reperfusion injury might attract increased interest in the scientific community. Nevertheless, to the best of our knowledge, no clinical studies have been implemented to test the therapeutic potential of H 2 S in patients with diabetic cardiomyopathy and diabetes-aggravated cardiac ischemia/ reperfusion injury. We anticipate that more retrospective and comparative clinical trials will be performed to investigate the therapeutic value of H 2 S in diabetic cardiomyopathy and its progressed heart failure. Of note, more well-designed preclinical or clinical studies are required to assess the safety, effectiveness, pharmacodynamics of H 2 S-based therapeutics on the broader population with diabetic cardiomyopathy and diabetes-aggravated cardiac ischemia/reperfusion injury. Tremendous efforts aiming to translate those basic research results into the clinical arena are underway by using H 2 S-releasing compounds.

CONCLUSION AND PERSPECTIVES
In this review, we provided the recent advances in our knowledge on the roles of endogenous H 2 S or pharmacologically administered H 2 S donors in diabetes-related cardiomyopathy and myocardial ischemia/reperfusion injury. The present review suggests that H 2 S acts as a cardioprotective mediator in diabetic heart by regulating various signaling pathways ( Figure 5). Both endogenously generated H 2 S and exogenously supplied H 2 S donors confer cardioprotection in a variety of settings. Thus, H 2 S might be recommended as a therapeutic agent against diabetic cardiomyopathy and diabetes-aggravated cardiac ischemia/reperfusion injury in pre-clinical studies.
Given that the therapeutic effect of H 2 S on diabetic cardiomyopathy were mediated by multiple signaling pathways, deciphering into the pros and cons of these cellular events induced by H 2 S might facilitate the knowledge of pharmacological activities and drug development of H 2 S. Also, the stable and controllable H 2 S donors should be designed for clinical management of diabetic cardiomyopathy. Eventually, a combination therapy of H 2 S donors with other standard medications might be considered as novel strategies to treat diabetic cardiomyopathy or reduce the risk of heart failure in diabetic subjects.