Role of Endothelial Dysfunction in Cardiovascular Diseases: The Link Between Inflammation and Hydrogen Sulfide

Endothelial cells are important constituents of blood vessels that play critical roles in cardiovascular homeostasis by regulating blood fluidity and fibrinolysis, vascular tone, angiogenesis, monocyte/leukocyte adhesion, and platelet aggregation. The normal vascular endothelium is taken as a gatekeeper of cardiovascular health, whereas abnormality of vascular endothelium is a major contributor to a plethora of cardiovascular ailments, such as atherosclerosis, aging, hypertension, obesity, and diabetes. Endothelial dysfunction is characterized by imbalanced vasodilation and vasoconstriction, elevated reactive oxygen species (ROS), and proinflammatory factors, as well as deficiency of nitric oxide (NO) bioavailability. The occurrence of endothelial dysfunction disrupts the endothelial barrier permeability that is a part of inflammatory response in the development of cardiovascular diseases. As such, abrogation of endothelial cell activation/inflammation is of clinical relevance. Recently, hydrogen sulfide (H2S), an entry as a gasotransmitter, exerts diverse biological effects through acting on various targeted signaling pathways. Within the cardiovascular system, the formation of H2S is detected in smooth muscle cells, vascular endothelial cells, and cardiomyocytes. Disrupted H2S bioavailability is postulated to be a new indicator for endothelial cell inflammation and its associated endothelial dysfunction. In this review, we will summarize recent advances about the roles of H2S in endothelial cell homeostasis, especially under pathological conditions, and discuss its putative therapeutic applications in endothelial inflammation-associated cardiovascular disorders.


INTRODUCTION
Currently, cardiovascular disease is identified to be a major cause of people death around the world, and this situation is estimated to remain for many years to come, thus bringing a considerable burden to the world's health resource (Mathers and Loncar, 2006). It is well known that poor diet, smoking, obesity, and physical inactivity are various modifiable risk factors for cardiovascular diseases, all of which lead to a proinflammatory state (Allende-Vigo, 2010). Actually, a wide range of evidence supports a crucial role of inflammatory response in the pathogenesis of cardiovascular diseases through driving endothelial cell activation/dysfunction (Carter, 2012). Therefore, it is not unexpected that huge efforts have been made to identify therapeutically potential targets to halt endothelial cell inflammation.
The blood vessels are composed of connective tissue, fibroblasts, endothelial cells, and vascular smooth muscle cells (VSMCs). On the innermost side of blood vessels, the normal endothelium is a semipermeable layer between blood stream and blood vessel wall. Due to its tight specialized cell-to-cell junctions, the endothelium forms a barrier that selectively limits the movement of macromolecules (Rahimi, 2017). The barrier is critically involved in vascular tone, fluid homeostasis, and host defense (Zhang et al., 2018b). Activated endothelial cells may release various cytokines, chemokines, and growth factors that promote the proliferation, migration, and permeability of endothelial cells (Park-Windhol and D'amore, 2016). The endothelial cells with inflammatory phenotype cause inflammation in the blood vessels, resulting in endothelial dysfunction and following progression of cardiovascular diseases . In accordance with this notion, endothelial cell inflammation is directly responsible for various cardiovascular diseases, such as hypertension, atherosclerosis, aging, stroke, heart disease, diabetes, obesity, venous thrombosis, and intimal hyperplasia Castro-Ferreira et al., 2018;Haybar et al., 2019;Zhong et al., 2019).
In the endothelium, hydrogen sulfide (H 2 S), the third endogenous gaseous molecule after nitric oxide (NO) and carbon monoxide (CO), is synthesized and observed (Pan et al., 2017). Over the last decade, the roles of H 2 S in the pathogenesis of endothelial dysfunction have grown exponentially. As a result, the current understanding of H 2 Smediated endothelial cell functions in both heath and disease continues to deepen. However, the potential molecular mechanisms that underlie H 2 S-mediated cardiovascular homeostasis, especially endothelial inflammation, are not comprehensively elucidated. The present review focuses on the current progress regarding the roles of H 2 S in endothelial inflammation-related cardiovascular disorders including hypertension, atherosclerosis and diabetes. Furthermore, we will discuss the possible challenges for developing H 2 Sderived therapeutics to treat endothelial dysfunction in cardiovascular disorders.

ENDOTHELIAL DYSFUNCTION AND INFLAMMATION
The dysfunction of endothelial cells in the vasculature is profoundly implicated in the pathogenesis of cardiovascular disorders (Boulanger, 2016). Mounting evidence has shown that endothelial cell dysfunction is characterized by disrupted vascular tone and redox balance, and increased inflammatory reactions within the blood vessel wall (Ooi et al., 2018) ( Figure 1). Most commonly, the impaired endotheliumdependent vasodilatation is defined as a hallmark of endothelial dysfunction, which is critically responsible for several cardiovascular disorders including diabetes mellitus, hypertension, atherosclerosis, aging and heart failure (Leung and Vanhoutte, 2017;Suryavanshi and Kulkarni, 2017). More recently, endothelial activation is also a prominent alteration in endothelial dysfunction, which refers to the upregulations of chemokines and adhesion molecules and other proteins involved in cell-cell interactions (Weber and Noels, 2011;Ng et al., 2018), thus leading to the prothrombotic and proinflammatory circumstance in the blood vessels.
In activated endothelial cells, the expressions of proinflammatory cytokines, chemokines, enzymes, and adhesion molecules are substantially upregulated (Baghai et al., 2018). It is highly possible that endothelial cell inflammation plays an important role in the pathogenesis of endothelial dysfunction in cardiovascular disorders. Therefore, identification of endothelial cell-derived inflammatory factors and its underlying mechanisms may be effective in preventing the progression of cardiovascular diseases.

REGULATION OF ENDOTHELIAL FUNCTION BY H 2 S UNDER PHYSIOLOGICAL CONDITION
As the third endogenous gasotransmitter, H 2 S is primarily synthesized in mammalian tissues through enzymatic or nonenzymatic pathways (Li et al., 2011;Liu et al., 2012). The majority of endogenous H 2 S is produced by three enzymes including cystathionine g-lyase (CSE), cystathionine b-synthase (CBS) and 3-mercaptopyruvate sulfurtransferase (3-MST) in mammalian tissues (Liu et al., 2011) (Figure 2). In the vascular endothelium, H 2 S is synthesized via the enzymatic metabolism of CBS/CSE using cysteine as the substrates (Tao et al., 2017;Mitidieri and Gurgone, 2019). Likewise, the involvement of 3-MST and cysteine aminotransferase (CAT) in endothelial generation of H 2 S has been demonstrated (Wang, 2012).
The regulation of vascular tone by H 2 S may be dependent on endothelium-independent and -dependent manners (Wang et al., 2015b). In the vasculature, H 2 S has been shown to induce vasodilation in aorta (Zhao et al., 2001), gastric artery (Kubo et al., 2007), mesenteric artery (Cheng et al., 2004), and internal mammary artery (Webb et al., 2008). The underlying mechanism by which H 2 S relaxes blood vessels is related with activation of vascular smooth muscle ATP-sensitive K + (KATP) channels (Zhao et al., 2001), independently of the endothelium. The involvement of KATP channels in H 2 S-induced vasodilation is further confirmed by a finding that this relaxation is partially blocked by an inhibitor of KATP channels glibenclamide (Webb et al., 2008). Despite of these results, the exact mechanism of how KATP channels are directly activated by H 2 S still remains unknown. It is also reported that 4-aminopyridine-sensitive K + channels are involved in H 2 S-induced relaxation in the rat coronary artery (Cheang et al., 2010). The H 2 S donor sodium hydrosulfide (NaHS) induces concentration-dependent vasorelaxation in both mesenteric arteries and aortas, which is blocked by the KCNQ-type Kv channel inhibitor XE991, suggesting the involvement of KCNQ channels in H 2 Smediated peripheral artery relaxation (Schleifenbaum et al., 2010). Moreover, Ca 2+ channels or sparks (Jackson-Weaver et al., 2015), Cl(-)/HCO(3)(-) channels (Kiss et al., 2008), the NO pathway (Ali et al., 2006), phospholipase A2 (D'emmanuele Di Villa Bianca et al., 2011), transient receptor potential (TRP) channels (White et al., 2013), and metabolic/mitochondrial effects (Kiss et al., 2008), are also suggested to be implicated in H 2 S-induced vasorelaxation. H 2 S appears to play an important role in vasorelaxation via multidimensional mechanisms. In the endothelium, recent studies have provided several lines of evidence to support that H 2 S might function as an endothelium-derived relaxing factor (EDRF), which shares many common traits with other EDRFs (Wang, 2009). Interestingly, the vasorelaxation actions of H 2 S are more remarkable in peripheral resistance arteries than in largeconduit arteries, the effects require the membrane hyperpolarization of both VSMCs and endothelial cells, as well as activation of endothelial intermediate conductance (IK(Ca)) and small conductance (SK(Ca)) potassium channels (Mustafa et al., 2011;Tang et al., 2013). The definition of H 2 S as an endogenous EDHF might shed light on possible therapeutic effects of H₂S on pathological abnormalities in the vascular system. Still, more extensive and mechanistic studies are needed to determine whether H 2 S is a new EDRF in the future.
The endothelial cells also orchestrate tube formation and angiogenesis (Watson et al., 2017). H 2 S is reported to stimulate endothelial proliferation, migration, and angiogenesis (Wang et al., 2010b) (Figure 3). Furthermore, administration of H 2 S promotes angiogenesis in the Matrigel plug assay (Cai et al., 2007). However, it should be pointed out that high dose of H 2 S loses the ability to induce angiogenesis (Cai et al., 2007). In a rat model of chronic hindlimb ischaemia, intraperitoneal injection of the H 2 S donor NaHS at the lower dose significantly improves capillary density, angiographic scores, thus improving hindlimb blood flow (Wang et al., 2010a). In line with the results discussed earlier, higher dose of the H 2 S donor is found to be ineffective in this model (Wang et al., 2010a). On these grounds, we speculate that the effects of H 2 S donors in angiogenesis may range from physiological, cytoprotective effects (low concentration) to cytotoxic effects (which are generally apparent at higher concentrations) (Szabo and Papapetropoulos, 2011). From a genetic perspective, mutant mice lacking CSE exhibit a variety of pathological features, including delayed wound healing secondary to inhibition of angiogenesis (Papapetropoulos et al., 2009). It has been reviewed that several cellular signaling pathways, such as the PI3K/Akt pathway, the mitogen activated protein kinase (MAPK) pathway, and ATP-sensitive potassium channels, are involved in H 2 S-mediated angiogenic effects (Szabo and Papapetropoulos, 2011). In addition to this, further study has demonstrated that H 2 S specifically disrupts cys1045-cys1024 disulfide bond in vascular endothelial growth factor receptor 2 (VEGFR2) and then stimulates its conformation for angiogenesis (Tao et al., 2013). As a molecular switch, H 2 S is also reported to activate signal transducer and activator of transcription 3 (STAT3) (Kan et al., 2014), mammalian target of rapamycin (mTOR), and the VEGFR2 pathway , then the endothelial cell proliferation and angiogenesis are observed. It is noteworthy that due to its proangiogenic effects, H 2 S might lead to pathological angiogenesis in atherosclerotic plaques, thus facilitating plaque vulnerability (Van Den Born et al., 2016). In spite of this, therapeutic angiogenesis is important for wound healing, organ ischaemia, or the reperfusion of previously ischaemic organs (Caporali and Emanueli, 2011;Dulmovits and Herman, 2012;Ng et al., 2018). For this reason, the reparative angiogenesis by H 2 S may provide novel therapeutic avenues for post-ischemic neovascularization. Due to the physiological importance of H 2 S in the endothelium, further research is indispensable to examine the novel roles of endogenous H 2 S in the regulation of cardiovascular functions.

ROLE OF H 2 S IN ENDOTHELIAL INFLAMMATION
Using intravital microscopy, H 2 S donors are found to attenuate the leukocyte adherence in rat mesenteric arteries induced by aspirin, this effect may be likely dependent on activation of KATP channels (Zanardo et al., 2006;Zuidema and Korthuis, 2015). In accordance with this, blockade of endogenous H 2 S exacerbates leukocyte-mediated inflammation in the endothelium (Zanardo et al., 2006). By contrast, NaHS promotes leukocyte rolling and adherence in mesenteric venules of mice with cecal ligation and puncture (CLP)-induced sepsis (Zhang et al., 2007). These conflicting results imply that H 2 S acts as a pivotal regulator of leukocyte activation under different inflammatory states. However, in recent years, more studies support that H 2 S could inhibit the process of endothelial cell inflammation (Wen et al., 2018). For instance, specific endothelial deletion of CSE is associated with the development of endothelial inflammation and atherosclerosis, effects that are reversed on treatment with a polysulfide donor (Bibli et al., 2019). H 2 S treatment reduces the increases in inflammatory mediators such as vascular cell adhesionmolecule-1 (VCAM-1), intercellular adhesionmolecule-1 (ICAM-1) and monocyte chemoattractant protein-1 (MCP-1) in endothelial cell induced by tumour necrosis factor-a (TNF-a), and the underlying mechanism of this protective effect is primarily mediated by inhibition of soluble TNF-a shedding and its relevant MCP-1 release (Perna et al., 2013). Similarly, exogenous H 2 S attenuates Ang II-induced inflammation response via inhibition of the nuclear transcription factor-kB (NF-kB) signaling pathway in endothelial cells (Hu et al., 2016). Inhibition of the NF-kB pathway is also required for H 2 S to attenuate pulmonary endothelial cell inflammation and subsequent pulmonary hypertension (Feng et al., 2017b). Endogenous H 2 S could directly induce sirtuin1 (SIRT1) sulfhydration and stability, thus reducing aortic inflammation and atherosclerotic plaque formation (Du et al., 2019). Deficiency of CSE increases endogenous sulfur dioxide (SO 2 ) level in endothelial cells, and blockade of endogenous SO 2 aggravates CSE knockdown-induced NF-kB pathway and its downstream inflammatory factors release in endothelial cells (Zhang et al., 2018a), suggesting that the increased endogenous SO 2 generation might act as a compensatory mechanism for the downregulated CSE/H 2 S pathway in endothelial inflammatory response (Zhang et al., 2018a). It is concluded that the anti-inflammatory effects of H 2 S donors show tremendous promise for the treatment of endothelial inflammation-related cardiovascular disorders.
In response to proinflammatory cytokines, the leukocyte or macrophages are activated and recruited to the endothelium, thus causing the development of endothelial dysfunction-related cardiovascular diseases (Fang et al., 2013). Fortunately, H 2 S is found to alleviate vascular inflammation through various signaling pathways, including inhibition of NF-kB and nucleotide-binding oligomerization domain, leucine rich repeat, and pyrin domain-containing protein 3 (NLRP3) inflammasome, activation of KATP channels and voltage-and calcium-gated potassium (BKCa) channels (Fiorucci et al., 2005;Zanardo et al., 2006;Zuidema et al., 2010;Altaany et al., 2014;Bourque et al., 2018;Li et al., 2019). These possible mechanisms of H 2 S may explain that H 2 S can diminish vascular inflammation and attenuate the vascular injury, suggesting that the antiinflammation effect of H 2 S is a benefit for cardiovascular protection. Next, we will discuss the beneficial roles of H 2 Smediated suppression of endothelial dysfunction in cardiovascular disorders including atherosclerosis, diabetic cardiovascular complications and hypertension.

H 2 S-RELATED ENDOTHELIAL DYSFUNCTION IN ATHEROSCLEROSIS
Atherosclerosis, a chronic vascular disease of large and medium arteries, involves various risk factors including lipid deposition, hypertension, inflammatory factors, and hyperhomocysteinemia, which synergistically elicit endothelial dysfunction (Baszczuk et al., 2014). Biochemical effects of these factors on the endothelium could lead to endothelial cell damage and vascular remodeling (Baszczuk et al., 2014). This important event induces endothelial inflammation, macrophage differentiation, foam cell formation, platelet deposition, and thrombus formation (Chistiakov et al., 2017;Rahman and Woollard, 2017). As such, correction of endothelial dysfunction could be a therapeutic strategy for management of atherosclerosis.
In recent years, considerable evidence indicates that the downregulated CSE/H2S pathway plays a pathophysiologic role in the development of atherosclerosis (Kanagy et al., 2017;Fan et al., 2019). CSE-knockout mice fed with atherogenic diet exhibt more severe atherosclerosis, suggesting that the disturbed CSE/ H2S pathway predisposes the animals to the development of atherosclerosis (Mani et al., 2013). Macrophage inflammation directly contributes to necrotic core formation and plaque instability in atherosclerosis (Kavurma et al., 2017). In oxidized low density lipoprotein (ox-LDL)-treated macrophage, the levels of CSE mRNA and protein expression, as well as H2S production are remarkably decreased, thus, this finding indicates that alterations of the CSE/H2S pathway plays an important role in ox-LDL-stimulated m acrophage infl am ma tion a nd atherosclerosis (Wang et al., 2013b). It is worth noting that CBS deficiency may cause hyperhomocysteinemia, which is an independent risk factor for the development of atherosclerosis (Zhang et al., 2012a;Yuan et al., 2017). In transgenic CBSdeficient mice, the emegence of hypercholesterolemia accelerates atherosclerotic lesions via oxidative stress and inflammatory monocyte generation (Zhang et al., 2009). On the other hand, vascular calcification and neointimal hyperplasia are also involved in the progression of atherosclerosis (Durham et al., 2018;Yu et al., 2018). Not surprisingly, the production of H2S and CSE protein expression are obviously decreased in rats with vascular calcification (Wu et al., 2006). The CSE expression and H2S production are impaired during the development of balloon injury-induced neointimal hyperplasia in rats, and this effect is obviously reversed by H2S treatment (Meng et al., 2007). In mice with high fat diet for 16 weeks, it is found that CSE protein level is downregulated in the liver, the lung, and the aortic endothelium, 3-MST was also reduced in the liver (Peh et al., 2014). By contrast, CBS expression was higher in the liver and the kidney (Peh et al., 2014). These results suggest that an abnormal H2S pathway may be an important factor for the pathophysiology of metabolic disorders and atherosclerosis.
The anti-atherosclerotic mechanisms of H 2 S have been gradually described, including anti-inflammatory response, anti-oxidative action, endothelial function preservation, inhibition of foam cell formation and regulation of ion channels (Altaany et al., 2014;Mani et al., 2014;Xu et al., 2014;Wang et al., 2017;Barton and Meyer, 2019). Reduced CSE expressions at both mRNA and protein levels are detected in ox-LDL-treated endothelial cells and in aortas from apolipoprotein E knockout (ApoE -/-) mice (Leucker et al., 2017). In this study, the authors demonstrated that increased histone deacetylase 6 (HDAC6) downregulated CSE and H 2 S production via posttranslational modifications, thus leading to endothelial cell dysfunction and the development of atherosclerosis (Leucker et al., 2017). In cultured vascular endothelial cells, the expressions of miR-455-3p, endothelial nitric oxide synthase (eNOS) protein and NO production are augmented by H 2 S. Besides, H 2 S levels and miR-455-3p expressions are also increased in human atherosclerosis plaque, suggesting that the miR-455-3p/eNOS/NO axis is required to H 2 S to circumvent the development of atherosclerosis . Genetic deletion of CSE exaggerates atherosclerosis in ApoE -/mice, and treatment of CSE-knockout mice with H 2 S inhibits the development of atherosclerosis (Mani et al., 2013), pinpointing that endogenous H 2 S may be of benefit in the treatment of atherosclerosis. In addition, the augmented expressions of selectins (P-selectin and E-selectin) and cell adhesion molecules (ICAM-1 and VCAM-1) are observed in vascular endothelial cells from CSE knockout mice (Mani et al., 2013). GYY4137, a novel slow-releasing H 2 S compound, retards atherosclerotic plaque formation and partially restores endothelium-dependent relaxation in ApoE -/mice through decreasing vascular inflammation and oxidative stress . Preconditioning with NaHS also grants a protection in atherosclerosis, as manifested by decreased atherosclerotic plaque size and aortic ICAM-1 levels . Supplementation with H 2 S ameliorates, while inhibition of H 2 S formation intensified aortic CX3CR1 and CX3CL1 expressions and the formation of atherosclerosis (Zhang et al., 2012b). Recently, H 2 S induces S-sulfhydration of kelch-like ECH-associated protein 1 (Keap1) and nuclear factor erythroid 2-related factor 2 (Nrf2) dissociation from Keap1, followed by Nrf2 nuclear translocation and anti-oxidize effects in endothelial cells, contributing to the ameliorating effect of H 2 S on atherosclerosis in the context of diabetes (Xie et al., 2016). Furthermore, in a mouse model of disturbed flow-induced atherosclerosis, application of H 2 S donor NaHS considerably attenuates the severity of atherosclerosis via upregulating expressions of angiotensin converting enzyme 2 (ACE2), thus converting pro-atherosclerotic Ang II to anti-atherosclerotic angiotensin 1-7 (Ang-(1-7)) (Lin et al., 2017). At the cellular level, NaHS promotes the expression of ACE2 to exert antiinflammatory properties in lipopolysaccharide (LPS)-stimulated endothelial cells, as pretreatment with a selective ACE2 inhibitor DX600 abrogates the anti-inflammatory effect of NaHS (Lin et al., 2017). The results showed that endogenous H 2 S system was involved in the development of atherosclerosis. Exogenous H 2 S could confer beneficial effects on the pathogenesis of atherosclerosis.
Also, H 2 S is involved in shear stress and blood viscosity. The occurrence of atherosclerosis may be initiated due to changed patterns of blood flow and ensuing shear stress (Dunn et al., 2015). It is well established that atherosclerotic plaque formation in the endothelium is site specific, and disturbed blood flow formed at the lesser curvature of the aortic arch and branch points promotes plaque formation, whereas steady laminar flow at the greater curvature is indicated to be atheroprotective (Heo et al., 2016). The branches and curvatures of the blood vessels are predisposed to endothelial dysfunction and atherosclerosis progression . Under oscillatory shear stress, H 2 S treatment inhibits monocyte adhesion to endothelial cells via activating the NO-producing Akt/eNOS signaling pathway (Go et al., 2012). Conversely, H 2 S impairs shear stress-induced dilation of isolated mouse coronary arteries by inhibition of NO generation (Chai et al., 2015). It is likely that both H 2 S and NO are implicated in the shear stress-induced atherosclerosis. However, further investigation is required to help us obtain more novel insights into the underlying mechanisms.

H 2 S-RELATED ENDOTHELIAL DYSFUNCTION IN DIABETIC VASCULAR COMPLICATIONS
Circulating levels of H 2 S are markedly reduced in diabetic animal models, such as diabetic rats (Jain et al., 2010;Suzuki et al., 2011), diabetic mice (Brancaleone et al., 2008), and also in diabetic patients (Jain et al., 2013;Suzuki et al., 2017). However, the mRNA level of CSE in the aortas of diabetic rats is not altered (Denizalti et al., 2011). Likewise, the expressions of CSE, CBS, and 3-MST are unaltered in either high glucose-treated endothelial cells or in the aortas of diabetic rats (Jain et al., 2010;Coletta et al., 2015). On the contrary, it has been demonstrated that both high glucose and palmitate inhibit CSE expression and H 2 S production in rat aortic endothelial cells, while exogenous H 2 S could protect endothelial cells against apoptosis under high glucose and palmitate stimulation via suppressing oxidative stress, decreasing mitochondrial fragments and promoting mitophagy . The CSE expression and H 2 S content are significantly reduced in granulation tissues of wounds in obese diabetic mice when compared with control mice (Zhao et al., 2017). The expression of CSE and H 2 S level are reduced after renal ischemia/reperfusion injury in diabetes mellitus . In comparison with control mice, the H 2 S content and CSE expression in heart tissues of diabetic rats are also markedly lower . In progressive diabetic nephropathy, CSE expression is markedly reduced, whereas CBS expression is unaffected (Yamamoto et al., 2013). By contrast, the protein and mRNA expression of CBS are specifically decreased in the kidney, while CSE expression remains unchanged in obese diabetic mice . Interestingly, CSE expression is upregulated in cerebral microvessels of type I diabetic rats (Streeter et al., 2013). Although the data are conflicting, they raise the possibility that H 2 S may be a double-edged sword under diabetic pathophysiology. Certainly, more research is needed to determine the molecular mechanisms underlying the changed or unchanged expressions of H 2 S-generating enzymes/H 2 S under diabetic conditions.
Despite of the aforementioned results, recent study has demonstrated that 3-MST activity is inhibited in endothelial cells during hyperglycemia, leading to reduced H 2 S level, impaired angiogenesis, and suppressed mitochondrial function (Coletta et al., 2015). It is highly probable that inactivation of 3-MST and elevated H 2 S depletion are putative mechanisms for the decreased circulating H 2 S levels in hyperglycemic endothelial cells. The high glucose-incubated vascular rings exhibit impaired endothelium-dependent relaxation, and this effect is rescued by CSE overexpression or H 2 S supplementation (Suzuki et al., 2011). In the same study, they have also shown that the vascular rings from mice with gene knockout of CSE display an aggravated impairment of endothelium-dependent relaxation in response to hyperglycemia (Suzuki et al., 2011). It is anticipated that genetic modulations of CSE, CBS or 3-MST levels are effective approaches to experimentally investigate the roles of H 2 S in diabetic vascular complications.
Exposure to high glucose results in elevated ROS production and apoptosis, as well as decreased superoxide dismutase activity in endothelial cells, and all the above responses could be eliminated by pretreatment with H 2 S (Guan et al., 2012). Exogenous H 2 S alleviated the ROS overproduction and apoptosis in hyperglycemic endothelial cells through inhibiting necroptosis . In the aortas of diabetic rats, the connexin (Cx) 43 and 40 expressions are downregulated, while protein kinase C (PKC) and nicotinamide adenine dinucleotide phosphate-oxidase (NADPH) oxidase subunits are upregulated, H 2 S appears to be effective in attenuating these abnormalities (Zheng et al., 2010). The novel mitochondria-targeted H 2 S donors AP123 and AP39 are proven to prevent hyperglycemiatriggered oxidative stress and metabolic abnormalities in microvascular endothelial cells, suggesting that these compounds could be useful for the treatment of diabetic vascular complications (Gero et al., 2016). Induction of H 2 S by Ginkgolide B alleviates endothelial dysfunction via inhibiting oxidative stress and increasing NO bioavailability in diabetic rats (Wang et al., 2015a). It is likely that the cardiovascular protective effects of H 2 S in diabetes may be mediated by inhibition of oxidative stress.
Inhibition of the leptin/leptin receptor signal pathway contributes to the protective effects of H 2 S on high-glucoseinduced injuries in endothelial cells . Pretreatment with H 2 S prevents high glucose-induced ICAM-1 levels as well as NF-kB activation in endothelial cells (Guan et al., 2013). Besides, stimulation of endothelial cells with high glucose significantly promotes the secretion of endothelin-1 with the concomitant suppression of H 2 S production, and administration of H 2 S attenuates the release of endothelin-1 induced by high glucose (Guan et al., 2015). The increasing recognitions of protective effects of H 2 S in high glucose-induced endothelial inflammation provide a new avenue of antagonism towards diabetic vascular complications.
In addition, high glucose/palmitate-induced excessive autophagy in endothelial cells is rectified by H 2 S, this may be mediated by the Nrf2-ROS-adenosine 5'-monophosphate (AMP)-activated protein kinase (AMPK) signaling pathway . However, another group demonstrates that exogenous H 2 S inhibits mitochondrial apoptosis and promotes mitochondrial autophagy, thus protecting endothelial cells against apoptosis induced by high glucose and palmitate . These contradictory results suggest that additional research is necessary to ascertain the role of autophagy in H 2 Smediated protective actions on diabetic endothelial dysfunction.

H 2 S-RELATED ENDOTHELIAL DYSFUNCTION IN HYPERTENSION
The abnormal levels of H 2 S have been found to be correlated with hypertension (Szabo, 2007;Whiteman and Winyard, 2011). Specifically, in a clinical study, patients with severe hypertension exhibited lower plasma H2S level (Meng et al., 2015). In subjects with pulmonary hypertension, both CSE expression and H2S level are significantly lower than those in healthy population (Sun et al., 2014). In situ hybridization analysis has shown that the expression of CSE mRNA is downregulated in the pulmonary arteries of rats with pulmonary hypertension (Xiaohui et al., 2005). Likewise, the reduced protein contents of CSE and CBS are detected in pulmonary artery endothelial cells from tobacco smoke-induced emphysema and pulmonary hypertension (Han et al., 2011). It has been revealed a marked reduction in CBS and CSE expression as well as H2S production in mesenteric artery and carotid artery from dexamethasone-induced hypertensive rats (D'emmanuele Di Villa Bianca et al., 2015). By contrast, the suppressed CBS expression and reduced H2S concentration in the kidney are observed in high salt-induced hypertension in Dahl rats (Huang et al., 2015). A decreased CSE/H2S activity is a potential contributor to the pathogenesis of maternal hypertension in preeclampsia (Wang et al., 2013a). Also, the plasma H2S level and CSE protein expression in thoracic aorta are all suppressed in spontaneously hypertensive rats (SHR) in comparison with normotensive rats (Yan et al., 2004;Ahmad et al., 2014). An intriguing study has illustrated that the blood pressure is enhanced by treatment with the combination of CSE inhibitor DLpropargylglycine (PAG) or the CBS inhibitor aminooxyacetic acid (AOA) in rats, while either compound alone has no any effect on the arterial pressure, suggesting that H2S plays a critical role in regulating blood pressure (Roy et al., 2012). Despite that the expression of 3-MST is still uncertain under hypensive condition, 3-MST gene therapy improves renovascular dysfunction in response to hyperhomocysteinemia (Sen et al., 2012). Thus, a better understanding of the biochemical functions of the H2S-producing enzyme 3-MST as well as its roles in hypertension may lead to new therapeutic targets based on modulation of H2S production. Overall, these studies suggest that endogenous H2S dysregulation plays an important role in regulating hypertneison-associated pathological processes.
As mentioned above, a close relationship between H 2 S-related endothelial dysfunction and hypertension is confirmed by an observation that genetic deletion of CSE causes the development of hypertension in mice (Yang et al., 2008). In these CSE knockout mice, the endothelium-dependent relaxation of resistance mesenteric arteries is particularly impaired (Yang et al., 2008). In a mouse model of Ang II-induced hypertension, both aortic endothelial function and NO bioavailability are significantly attenuated, and these are reversed by treatment with H 2 S (Al-Magableh et al., 2015). Conversely, blockade of endogenous H 2 S exacerbates these abnormalities (Al-Magableh et al., 2015). In other studies, application of H 2 S donors decrease blood pressure, reverse vascular remodeling via suppressing VSMC proliferation, and collagen deposition in the blood vessels (Zhao et al., 2001;Li et al., 2008;Wang, 2012;Meng et al., 2015;Tomasova et al., 2015). H 2 S therapy markedly restores eNOS function and NO bioavailability in Nw-nitro-l-arginine methyl ester (L-NAME)induced hypertensive rats (Ji et al., 2014). In agreement with this, the improvement of endothelial function by H 2 S is largely attributed to inhibition of oxidative stress, suppression of renin angiotensin system (RAS), downregulation of BMP4/COX-2 pathway, or activation of the PPARd/PI3K/Akt/AMPK/eNOS pathway, thus contributing to the antihypertensive mechanism of H 2 S in renovascular hypertensive rats (Xue et al., 2015;Xiao and Dong, 2016;Xiao et al., 2018). In SHR, exogenous H 2 S administration significantly reduces blood pressure and abrogated damaged endothelial dysfunction via inactivation of NLRP3 inflammasome and oxidative stress . H 2 S treatment blunts increases in systolic blood pressure and ameliorates endothelial dysfunction by inhibiting oxidative stress in lead-induced hypertensive rats (Possomato-Vieira and Goncalves-Rizzi, 2018). These results demonstrate that the H 2 S pathway may provide potential therapeutic target for treating different hypertension models.
Importantly, supplementation with S-zofenopril ameliorates vascular endothelial dysfunction by potentiating the H 2 S pathway in spontaneously hypertensive models (Bucci et al., 2014). Moreover, exercise training counteracts hypertension, ameliorates vascular remodeling, and endothelial dysfunction via restoring bioavailability of H 2 S and NO in hypertensive rats (Gu et al., 2013). HDAC6 inhibitor tubastatin A alleviates Ang II-induced high blood pressure and vasoconstriction by preventing the protein degradation of CSE (Chi et al., 2019). Overall, these studies suggest that upregulation of H 2 S may be considered as a promising strategy for preventing the progression of hypertension and its associated endothelial dysfunction (Figure 4). However, further in-depth research is still required to understand the precise underlying mechanisms, and this will be helpful to develop better therapeutic employment of H 2 S in the treatment of hypertension.

EVIDENCE FOR H 2 S/NO CROSSTALK IN ENDOTHELIAL INFLAMMATION
Due to the importance of H 2 S and NO in cardiovascular disease, the interactive regulatory functions of H 2 S and NO in endothelial dysfunction-associated cardiovascular disease may be a very attractive subject. In other words, the biological interactions of H 2 S with NO could influence each other's fate in the endothelium as described previously (Wang et al., 2015b;Nagpure and Bian, 2016;Wu and Hu, 2018) (Figure 2).
Studies on H 2 S/NO interaction in inflammation response, especially in endothelial cell inflammation, have been less extensive. Administration of LPS increases H 2 S synthesis, upregulates CSE and iNOS expressions, and promotes myeloperoxidase activity in the liver, whereas the effects are inhibited by the NO donor, nitroflurbiprofen (Anuar et al., 2006). These results suggest that downregulation of H 2 S biosynthesis is responsible for the augmented antiinflammatory activity of nitroflurbiprofen in the liver (Anuar et al., 2006). In turn, pretreatment with H 2 S is able to inhibit LPS-induced iNOS expression and NO production via heme oxygenase 1 (HO-1) expression in macrophages (Oh et al., 2006). In accordance with this finding, H 2 S donor inhibits the release of the pro-inflammatory mediators and NO production, potentially via NF-kB inactivation in LPS-treated macrophages (Whiteman et al., 2010). The pulmonary CSE expression and H 2 S levels are downregulated in a model of inflammatory lung disease (Chen et al., 2009;Whiteman and Winyard, 2011). NaHS, a donor for H 2 S, significantly attenuates pulmonary iNOS activation in ovalbumin-treated rats (Chen et al., 2009). Moreover, H 2 S is found to act as an anti-inflammatory agent contributing to gastrointestinal mucosal defense through NO-dependent pathway (Jensen et al., 2017). In cardiovascular system, vasodilation is impaired and endothelial H 2 S content is decreased in vessels from obese mice, this may be attributed to the increased iNOS activity in proinflammatory macrophages (Candela et al., 2017). This finding suggests that macrophagesderived iNOS promotes microvascular endothelial dysfunction through reducing the bioavailability of H 2 S in the blood vessel (Candela et al., 2017). GYY4137, a novel slow-releasing H 2 S compound, is reported to attenuate vascular inflammation and improve endothelial function via activating aortic eNOS phosphorylation in ApoE -/mice . A number of studies have showed that H 2 S strengthens endothelial NO production via activating eNOS phosphorylation (Predmore et al., 2011;Xiao et al., 2018), which results in ameliorating the development of hypertension (Xiao et al., 2018). Taken together, the above studies imply that a complex interaction between H 2 S and NO might serve as an important regulator for endothelial inflammation and associated endothelial dysfunction. However, the potential mechanisms of the interactions between H 2 S and NO in endothelial inflammation remain unclear. As research in this area progresses and more data are available, it will help us to better understand the underlying mechanisms.

ROLE OF H 2 S IN INTESTINAL MICROBIOTA AND CIRCADIAN RHYTHMS
It should be emphasized that intestinal microbiota is also an emerging factor for human health and disease, including cardiovascular diseases (Tang et al., 2017). In recent years, both human and animal experiments have established that alterations in the composition, function, and metabolites of intestinal flora might induce gut microflora dysbiosis, contributing to the pathogenesis of cardiovascular disorders (Tang et al., 2019). Circadian rhythmicity is a characteristic of mammalian metabolism that orchestrates metabolic processes in living organisms based on day/ night light cycles (Liu and Chang, 2017). Disturbance of circadian rhythmicity is associated with increased risk for metabolic obesity, diabetes, and cardiovascular dysfunction (Crnko et al., 2018). Similarly, the intestinal microbiota exhibits their own circadian rhythmicity in terms of composition and functions (Tahara et al., 2017). Circadian disorganization may affect the intestinal microbiota which may result in metabolic syndrome and cardiovascular diseases (Voigt et al., 2016). Accumulating evidence has indicated that circadian rhythm disruption in intestinal microbiota is involved in various human diseases, including cardiovascular diseases . Thereafter, interfering with the composition, function, and metabolites of the intestinal flora or recovery of the normal circadian rhythm in the intestinal flora may provide valuable insights into potential therapeutic strategies for cardiovascular diseases.
Notably, the cysteine degradation by the microbiota is taken as a dominant pathway for H 2 S generation (Basic et al., 2017). Intestinal microbiota is a potential target of H 2 S, and H 2 S acts on gastrointestinal epithelium to modify the gut microbiota (Wallace et al., 2018). It has been reviewed that H 2 S is a double-edge sword for the intestinal epithelium with beneficial effect at low concentration (nanomolar to low micromolar), but deleterious effects at higher concentrations (high micromolar to millimolar) (Blachier et al., 2019). Considering the critical importance of intestinal microbiota and H 2 S in maintaining cardiovascular homeostasis, it is believed that intestinal microbiota-derived H 2 S integrates microbial and circadian cues FIGURE 4 | Schematic illustration of underlying mechanisms in which H 2 S protects against hypertension. H 2 S lowers high blood pressure via vasodilatation by activation of vascular KATP channels and inhibition of Ca 2+ influx. The PPARd/PI3K/Akt/AMPK signaling pathway participates in H 2 S-induced NO production. These above events cause vasodilation. H 2 S inhibits ROS production via Nrf-2/HO-1 related redox sensitive signaling pathways. In addition, H 2 S treatment blunts increases in systolic blood pressure by inhibiting inflammation-related signaling pathways. H 2 S, hydrogen sulfide; NO, nitric oxide; Akt, protein kinase B; eNOS, endothelial NO synthase; PKG, protein kinase G; PI3K, phosphoinositide 3-kinase; PPARd, peroxisome proliferators-activated receptor d; AMPK, adenosine 5'-monophosphate (AMP)-activated protein kinase; Nrf2, nuclear factor erythroid 2-related factor 2; HO-1, heme oxygenase 1; NF-kB, nuclear factor-kappa B; BMP4, bone morphogenetic protein 4; COX-2, cyclooxygenase-2; NLRP3, nucleotide-binding oligomerization domain, leucine rich repeat, and pyrin domain-containing protein 3.
for regulation of diurnal metabolic rhythms, thereby influencing the endothelial dysfunction in cardiovascular system. However, it is still largely unknown with respect to the roles of H 2 S in intestinal microbiota-mediated endothelial dysfunction. The relationship between H 2 S and intestinal microbiota in cardiovascular regulation may be a very interesting topic. As the gut microbiota leads to much more H 2 S production from cysteine than endogenous metabolism, it is likely that H 2 S from the bacterial or intestinal epithelium may be a critical determinant for cardiovascular health or disease. However, additional investigation is warranted to identify the exact roles of H 2 S in intestinal microbiota and circadian rhythms. Our current understanding of the relationship between H 2 S and intestinal microbiota in endothelial inflammation-related cardiovascular disorders is expanding continuously. The interaction between their signaling pathways is increasingly recognized as the future direction for the research in the gasotransmitters field.

CONCLUDING REMARKS AND FUTURE PERSPECTIVES
In addition to NO deficiency and intestinal flora, other factors including oxidative stress (Nakahira et al., 2011), endoplasmic reticulum stress (Battson et al., 2017;Luchetti et al., 2017), mitochondrial dysfunction (Gao et al., 2018), hypoxia (Feng et al., 2017a), homocysteine (Esse and Barroso, 2019), and immune activation (Pan et al., 2017) are also closely related with endothelial inflammation and dysfunction in cardiovascular diseases. With in-depth research, our knowledge on the underlying mechanisms of H 2 S-mediated suppression of endothelial cell inflammation is expanding and it is now apparent that interactions between H 2 S and endothelial inflammation-regulated pathways may be proposed as a promising approach for cardiovascular disease therapy. A better understanding of such interactions will be favorable to develop novel therapeutic strategies for endothelial dysfunctionrelated cardiovascular diseases.
Due to a myriad of biological functions of H 2 S, there has been a growing interest regarding the enormously therapeutic potential of H 2 S in various diseases including cardiovascular diseases. However, our current knowledge on cardiovascular protective effects of H 2 S is mainly from animal or cell experiments using H 2 S donors or inhibitors of H 2 S-producing enzymes. Whether the promising effects of these chemicals in animal studies can be transferable to clinical studies warrants further studies. As such, it should be mentioned that clinical trial results will also pave the way to a better understanding of the effectiveness of H 2 S in human diseases. In one completed clinical trial in healthy volunteers and subjects with impaired renal function received known concentrations of sodium sulfide (clinicaltrials.gov, NCT00879645). Despite that only some results have been announced so far, but the treatment could be considered safe because no serious adverse effects are occurred in the involved patients. However, some caution can be warranted as another clinical trial regarding the potential of H 2 S in coronary artery bypass graft patients was terminated without undisclosed reasons (NCT00858936) and one had been withdrawn before enrollment (NCT01007461). Some other completed trials are completed to test the role of H 2 S in inflammatory diseases such as ulcerative colitis (NCT01282905) or septic shock and stroke (NCT01088490). However, to date, no results are posted due to unknown reasons. Therefore, further results and information from those ongoing and future trials will help to elucidate the physiological and pathophysiological importance of H 2 S in various diseases.
Until now, endothelial inflammation and dysfunction remain mortal factors for cardiovascular diseases. It is anticipated that a full understanding of the modulatory mechanisms of the link between endothelial inflammation and destructive H 2 S bioavailability might promote the translation of H 2 S biology to clinical management of endothelial dysfunction-related cardiovascular diseases. To achieve this, more original work remains to be experimentally evaluated in the future.