Endothelium as a Source and Target of H2S to Improve Its Trophism and Function

The vascular endothelium consists of a single layer of squamous endothelial cells (ECs) lining the inner surface of blood vessels. Nowadays, it is no longer considered as a simple barrier between the blood and vessel wall, but a central hub to control blood flow homeostasis and fulfill tissue metabolic demands by furnishing oxygen and nutrients. The endothelium regulates the proper functioning of vessels and microcirculation, in terms of tone control, blood fluidity, and fine tuning of inflammatory and redox reactions within the vessel wall and in surrounding tissues. This multiplicity of effects is due to the ability of ECs to produce, process, and release key modulators. Among these, gasotransmitters such as nitric oxide (NO) and hydrogen sulfide (H2S) are very active molecules constitutively produced by endotheliocytes for the maintenance and control of vascular physiological functions, while their impairment is responsible for endothelial dysfunction and cardiovascular disorders such as hypertension, atherosclerosis, and impaired wound healing and vascularization due to diabetes, infections, and ischemia. Upregulation of H2S producing enzymes and administration of H2S donors can be considered as innovative therapeutic approaches to improve EC biology and function, to revert endothelial dysfunction or to prevent cardiovascular disease progression. This review will focus on the beneficial autocrine/paracrine properties of H2S on ECs and the state of the art on H2S potentiating drugs and tools.


Vascular Endothelium
The vascular endothelium is the tissue that lines the inside of the circulatory system (blood vessels, lymphatic vessels and heart). The cells, arranged in a single layer oriented on the longitudinal axis of the vessel, assume a flattened shape and lay side by side with each other to form a complete monolayer. Structurally, the endothelial cell (EC) apical domain is in direct contact with blood or lymph, while the basolateral domain anchors to the basal lamina, which connects EC to the underlying tissues, such as the medial or muscular layer and the adventitia, rich in fibrous tissue.
The vascular endothelium acts as a selectively permeable barrier between extravascular and intravascular compartments and provides a non-thrombogenic surface for the cardiovascular system [1]. Nowadays, the endothelium can no longer be considered a passive barrier. Indeed, its anatomical position allows it to integrate the physical and neurohumoral signals from the blood and surrounding tissues for regulating vascular tone and permeability, cell adhesion, inflammation, smooth muscle phenotype and proliferation, as well as thromboresistance and blood fluidity [2,3].
The endothelial lining represents a wide area for the exchanges between blood and tissues (about 350 m 2 in humans) [3]. Electron microscopy observations reveal the continuous nature of arterial endothelium, characterized by tight junctions among adjacent cells in order to limit macromolecule exchange, and by a complex micro-vesicular system ECs respond to the increase in blood flow through the release of NO and PGI2 by the same cells. Indeed, shear stress causes rapid activation of eNOS and increases its gene transcription; it also endorses ECs to release factors that inhibit coagulation, leukocyte migration, and smooth muscle cell proliferation, simultaneously promoting EC survival. Conversely, low shear stress and turbulent blood flow promote a pathological feature in the endothelium responsible for atherosclerosis ignition, documenting the pivotal role of ECs in finely controlling vascular functions [2,6].
A healthy functioning endothelium also provides protection against radical species of oxygen and nitrogen (ROS/RNS). It is now clear that increased levels of ROS and RNS are harmful to cells and tissues and are involved in a wide range of cardiovascular diseases having endothelial dysfunction as an underlying phenomenon. This boosted the concept of oxidation as synonymous with cell damage and senescence. The post-translational modifications involving RNS share a common ancestor-high NO concentrations mainly synthetized by inducible NOS (iNOS), upregulated in response to various endotoxin or cytokine signals. Several pathological states are linked to the deregulation of NO levels, indicating that aberrant production of NO and its products can have deleterious consequences on cells [10]. Again, one of the functions related to a healthy endothelium is the scavenging activity of reactive species through the production of antioxidant products as H 2 S or protective enzymatic pathways.
The intimate surface of a healthy endothelium is both anticoagulant and antithrombotic: ECs secrete a wide range of molecules relevant to the regulation of blood clotting and platelet functions, as PGI2 and NO. Damage to the vessels or exposure to certain cytokines and proinflammatory stimuli overturns the equilibrium towards a procoagulant and prothrombotic EC phenotype, through the exposure of basal membrane components and/or tissue factor, and reduced presence of glycosaminoglycans or tissue factor inhibitor [5].
Endothelial trophism is guaranteed by the response to vasoactive and growth factors produced by surrounding tissues or autocrinally by the same ECs. Among the various examples, we and others have contributed to characterizing the beneficial effects on vascular endothelium by NO derived from eNOS, bradykinin, substance P, vascular endothelial growth factor (VEGF), fibroblast growth factor-2 (FGF-2), prostaglandin E2, H 2 S [11][12][13][14][15]. The molecular mechanisms responsible for cell survival, proliferation, migration and functioning include eNOS/NO/cGMP/protein kinase G (PKG), PI-3K/Akt, MAPK/ERK1/2 and gene transcription of autocrine factors as FGF-2 [7].
Epigenetics is an emergent mechanism involved in the regulation of vascular biology and endothelial trophism. Through chromatin structure modification, epigenetics can modify endothelial functions with an impact on cardiovascular disease, being the regulatory functions of epigenetics also active on endothelial precursor cells and circulating angiogenic cells [16,17]. DNA methylation, variants, histone post-translational modifications, and recently discovered RNA-based mechanisms represent the major pathways involved in the molecular basis of epigenetics. VEGF-A and NOS are the key players in regulating and maintaining cardiovascular functions. Their expression can be controlled by epigenetic mechanisms. In particular, VEGF-A epigenetic control can occur mainly through changes in histone code by RNAs. VEGF-A acts through VEGFR2, which in turn is regulated by promoter DNA methylation [18]. Furthermore, accumulating evidence indicates that epigenetic pathways play an important role in eNOS gene regulation [19].
These findings suggest the importance to deeply understand the epigenetic mechanisms involved in the regulation of vascular functions both in physiology and in pathological conditions.

Endothelial Dysfunction
Endothelial dysfunction refers to a systemic condition in which the endothelium loses its physiological properties, including the control of vasodilation, fibrinolysis and platelet aggregation. Key features of the endothelial dysfunction are: (1) the reduced local production of NO due to impaired activity (uncoupling) of eNOS, and of other vasodilating mediators as H 2 S; (2) the decrease in anticoagulant factors such as heparin; (3) the increase in the secretion of reactive species, von Willebrand factor, and tissue factor; (4) the overexpression of adhesion molecules for leukocytes and platelets [20,21]. All these factors concur to compromise the physiological vascular homeostasis. Due to the impairment of the main protective transmitters NO and H 2 S, the resulting endothelial dysfunction is associated with increased ROS and RNS levels and vascular oxidative stress [22,23]. The perpetuation of this condition then leads to retraction and death of the endothelium with increased permeability and exposure of the components of the basement membrane, that further amplifies the picture of vascular inflammation [4,20,22].
Endothelial dysfunction risk factors are represented by pathological states such as hypertension, diabetes, and hyperlipidemia, and improper lifestyles such as high-fat diets, tobacco and alcohol consumption, and physical inactivity [24]. Most cardiovascular diseases share endothelial dysfunction as a hallmark: atherosclerosis, diabetes complications, thrombosis, and hypercoagulation [20]. Moreover, physiological ageing through the phenomenon of mild chronic inflammation ("inflammaging") is accompanied by endothelial dysfunction [25]. Inflammatory factors such as tumor necrosis factor-α (TNF-α), inteleukin-6 (IL-6), intercellular adhesion molecule 1 (ICAM-1), and loss of the antioxidant mechanism are among the most influential promoters of vascular impairment [21].
Mounting evidence suggests that epigenetic mechanisms may contribute to vascular complications in many pathological conditions, such as diabetes or atherosclerosis, linked to altered endothelial trophism and functions [17,26]. The inflammatory phenotype in ECs induces the transcription of many cytokines and adhesion molecules, in a nuclear factorkappa B (NF-κB)-dependent manner. Epigenetic modifications in the NF-κB promoter region produce an increased expression of p65 subunit of NF-κB, and a hyper-activation of the NF-κB pathway [27]. Another mechanism could involve histone deacetylase 2 (HDAC2), which interacts and deacetylates Nrf2. Oxidized low-density lipoproteins (ox-LDLs) are able to downregulate HDAC2 expression, resulting in increased production of eNOS-dependent reactive species [16]. Furthermore, shear stress represented by blood flow alteration modifies EC gene expression and function. Dunn et al. demonstrated that disturbed blood flow stimulates DNA methyltransferase-1 (DNMT-1) expression in endothelial cells with aberrant DNA methylation at the promoter of flow-inducible genes, contributing to atherosclerosis [28]. Finally, recent data showed the role of noncoding RNA in regulating the expression of endothelial adhesion molecule [29].
Acute or chronic infections both by bacteria and viruses have cardiovascular consequences for their direct or indirect effects on vascular endothelium, through bacterial products or cytokines released by tissue and immune system cells [30,31], and through epigenetic regulation [32]. The recent pandemic due to SARS-CoV-2 supports this concept [33,34].
Furthermore, our studies and those of others have revealed that endothelial dysfunction is associated to impaired EC survival and physiological angiogenic outcomes with subsequent rearrangement of the microcirculation that contributes to the emergence of various pathological conditions and healing disorders [7,35]. ECs play a key role in the adaptation of tissues to damage, revealing their plasticity. A change in endothelial functions following ischemia can induce the transition to a mesenchymal phenotype characterized by functional, metabolic and gene expression signatures. Indeed, the mesenchymal phenotype, with increased cell migration and clonal expansion, participates in regenerating a functioning vascular network [36].
Considering endothelial function as a "barometer for cardiovascular risk", it is crucial to identify the molecular determinants underlying endothelial integrity and functionality. Seeing endothelium as an exchange regulator between the vascular wall and surrounding tissues, it is expected that dysfunctional ECs can determine damage to other tissues [37]. Indeed, a detailed assessment of the cellular and molecular mechanisms at the base of vascular function, and, particularly, of endothelial dysfunction, will help the diagnosis and treatment choice for a broad array of human disorders, including cardiovascular and neurodegenerative diseases [21,24,[38][39][40].

Biochemistry of H 2 S Production
H 2 S is a gas physiologically produced by tissue and vascular cells. The enzymes responsible for the synthesis of H 2 S are cystathionine β-synthase (CBS), cystathionine-γlyase (CSE), and 3-mercaptopyruvate sulfurtransferase (3-MST). The first two enzymes use L-cysteine as a substrate and are dependent on pyridoxal-5 -phosphate. 3-MST, on the other hand, works in association with cysteine aminotransferase which, starting from L-cysteine and α-ketoglutarate, produces 3-mercaptopyruvate. These enzymes are differentially expressed in the various tissues. In particular, the expression of CSE and 3-MST predominates in the cardiovascular system [8,41]. CSE is only present in the cytoplasm, while CBS and 3-MST both have a cytosolic and mitochondrial form, with the latter predominating. CSE is the principal enzyme responsible for H 2 S biosynthesis, located in vascular smooth muscle cells and found mainly in ECs [42,43]. This can explain why the concentration of H 2 S in the vascular tissues is around 100 times greater than in other tissues [44], suggesting a crucial role in vascular homeostasis, endothelial function, and trophism [45]. In addition, H 2 S can be also generated via a reduction in thiols and thiol-containing molecules, in a nonenzymatic manner [8]. Compared to CBS, knockout mice for CSE have no severe phenotype and normal lifespan. Their phenotype is mainly cardiovascular, with hypertension and endothelial dysfunction [46].
Information has become available about the regulation of the expression and activity of these enzymes ( Figure 1). It has been reported that NADPHox4 derived ROS (via heme-regulated inhibitor kinase/eIF2/activating transcription factor 4 (ATF4) signaling) enhance the expression of CSE [47]. Blood flow has been reported to exert divergent effect of H 2 S producing enzymes, depending on the type of endothelium and flow stress. While laminar flow was initially demonstrated to enhance the expression of CSE and 3-MST in ECs [47,48], recent data are more complex. In particular, laminar flow (high shear stress) has been demonstrated to inhibit CSE expression via KLF2 regulated miRNA-27b [49], while turbulent flow (low shear stress) seems to upregulate CSE [50,51]. In rat aortic ECs, calcium-sensing receptors increase CSE expression in a phospho-calmodulin kinases II-dependent manner to inhibit platelet activation [52]. Transcription factors specifically controlling CSE expression are among the others NF-κB in lipopolysaccharide (LPS)-stimulated macrophages [53], specificity protein 1 (Sp1) in smooth muscle cells [54] and Elk1 in beta pancreatic cells [55]. Nuclear factor of activated T cell (NFAT) binding sites have been identified in the CSE promoter. Intermittent hypoxia exposure reduces Ca 2+ -dependent activation of calcineurin/NFAT to lower CSE expression and impair vasodilation, while NFAT activation lowers CSE expression at the cell and microvascular levels [56].
OxLDL in one of the major stimuli to cause the endothelial damage that leads to atherosclerosis. One of the epigenetic mechanisms underlying CSE reduced expression at both mRNA and protein levels by oxLDL is increased histone deacetylase 6 (HDAC6) [57]. Recently this finding has been extended to blood pressure control, by assessing the role of tubastain A in AngII induce hypertension, and documenting that upregulation of CSE and H 2 S through HDAC6 inhibition can be a valid therapeutic strategy [58].
Additional information on CSE gene and enzyme control is provided in the chapters below and is summarized in Figure 1.

Molecular Signaling Activated by H 2 S into ECs
H 2 S in ECs performs a protective action on vessels in an autocrine/paracrine manner. It plays a role in the regulation of vasodilation, angiogenesis, inflammation, oxidative stress and apoptosis [59]. Three are the main mechanisms through which H 2 S exerts it biological effect: (i) reactive oxygen species/nitrogen species scavenging; (ii) interaction with metal centers; (iii) persulfidation (called also S-sulfhydration).
H 2 S acts through a post-translational modification-the S-sulfhydration of cysteine residues, which modifies the structure and activity of the target proteins [60]. The mechanism is persulfidation on reactive cysteine residues (-SH) of target proteins to form a persulfide group (-SSH). An example is the persulfidation of the ATP-dependent K channel (K ATP ) in ECs and smooth muscle cells, responsible for fast hyperpolarization and vasorelaxation [61]. Indeed, evidence has been provided in support of H 2 S function as an EDHF, exerting more remarkable vasorelaxation in the peripheral resistance arteries [62]. H 2 S also reduces ROS levels through their direct inactivation and by enhancing antioxidant defense mechanisms. One of the mechanisms underlying oxidative protection is given by the H 2 S regulation of the Keap1/Nrf2 pathway. Normally, the transcription factor Nrf2 is inhibited by its binding to Keap1 in the cytoplasm. In conditions of oxidative stress, H 2 S promotes the translocation of Nrf2 into the nucleus by means of the S-sulfhydration of the Keap1 inhibitor, causing the dissociation of the Keap1/Nrf2 complex. Nrf2 in the nucleus activates the antioxidant responsive element. Consequently, the transcription of many antioxidant genes such as superoxide dismutase, catalase, glutathione peroxidase and glutathione-S-transferase is induced, requiring hours or days to produce a biological effect [63].
Among other targets, H 2 S decreases inflammation by inhibiting transcription factors such as NF-κB through persulfidation [64], thus decreasing the expression of proinflammatory mediators. Most persulfidation reactions lead to target inhibition as phosphatase and tensin homolog (PTEN) [65], except for MEK1 activity, which in HUVEC leads to increased activity with DNA damage repair and senescence impairment [66]. Recently, Prx6 has been identified as a further target of sulfhydration on Cys47, which controls its decamerization and peroxidase activity [49], while the "S-sulphydrome" was identified among the many target proteins, and β3 integrin was identified as the key element of endothelial mechanotransduction [67].
The interaction of H 2 S with the NO/NOS pathway involves different modalities: as inhibition of PDE in smooth muscle cells, PI-3K/Akt-dependent phosphorylation of eNOS in Ser1177 [68] and stabilization of eNOS in the dimeric state through enzyme persulfidation [69]. Additionally, heme reduction in sGC enzyme with facilitated response to NO [70] and activation of protein kinase G Iα (PKG Iα) through disulfide bond formation [71] have been reported to potentiate the NO/cGMP pathway. H 2 S also stimulates endothelial proliferation and migration, aiding the process of angiogenesis and wound repair. H 2 S acts at several levels on the mechanisms responsible for angiogenesis, including the control of VEGF expression, through upregulation of the transcription factor hypoxia inducible factor-1α (HIF-1α) or direct modulation of the PI-3K and Akt pathways in ECs (signaling pathways also activated by VEGF) [41,72,73]. In angiogenic ECs, H 2 S has also been reported to activate signal transducer and activator of transcription 3 (STAT3) [74], mammalian target of rapamycin (mTOR) and the VEGFR2 pathway [75]. It has been reported that CBS silencing in ECs reduces VEGF signaling through VEGFR2 and neuropilin-1 downregulation [76].
A schematic summary of the molecular mechanisms activated or inhibited by H 2 S in ECs is reported in Figure 1. As a note, the effects of H 2 S are context and tissue-dependent, sometimes producing divergent functional effects. It is plausible that this depends on endothelial heterogeneity, tissue microenvironment, and physio-pathological conditions where there is the influence of epigenetic mechanisms.

Hypertension
Altered levels of H 2 S have been reported in both experimental models and clinical studies on patients with severe hypertension, where lower plasma H 2 S levels are described along with reduced content of CBS and CSE (see [77,78] for recent reviews). A human cohort study demonstrated a reduced H 2 S plasma level in hypertensive patients, suggesting H 2 S as a potential therapeutic target and diagnostic marker [79].
The intimate relationship between H 2 S-associated endothelial dysfunction and hypertension comes from the observation that CSE KO mice develop hypertension particularly with impaired endothelium-dependent relaxation in resistance mesenteric arteries [46].
The levels of H 2 S-producing enzymes are reduced in the vessel wall of spontaneous or drug induced hypertensive animals [80][81][82]. Exposure of cultured endothelial cells to AngII and, similarly, to H 2 O 2 , downregulated the expression and activity of CSE with induction of endoplasmic reticulum stress [83]. In a mouse model of Ang II-induced hypertension, H 2 S reversed the aortic endothelial dysfunction and reduced NO bioavailability, while blockade of endogenous H 2 S exacerbated these alterations [84]. Other studies demonstrated that administration of H 2 S donors decreases blood pressure and reverses vascular remodeling through the suppression of smooth muscle cell proliferation and collagen deposition in the vessel wall [61,[85][86][87][88][89]. H 2 S treatment noticeably reestablishes eNOS function and NO bioavailability in N ω -nitro-l-arginine methyl ester (L-NAME)-induced hypertensive rats [90]. From a mechanistic point of view, H 2 S improves endothelial function through the inhibition of oxidative stress, suppression of renin angiotensin system, downregulation of bone morphogenic protein 4/cyclooxygenase-2 (BMP4/COX-2) pathway, or activation of the PPARδ/PI-3K/Akt/AMPK/eNOS cascade, thus contributing to the antihypertensive mechanism of H 2 S in renovascular hypertensive rats [91][92][93]. In SHR, administration of H 2 S significantly decreases blood pressure and abrogates endothelial dysfunction through inactivation of NLRP3 inflammasome and oxidative stress [94]. In another disease model, lead-induced hypertension in rats, H 2 S treatment normalizes blood pressure and ameliorates endothelial dysfunction with an inhibition of oxidative stress [95].
Recently, it has been proposed that H 2 S can regulate EC pathological behavior through epigenetic mechanisms. H 2 S induces miR-129, which inhibits DNA methyltransferase-3 (DNMT3) and IL-17, found to be overexpressed in hypertension [96].
Summarizing all results, it can be concluded that the CSE/H 2 S signaling pathway may represent a potential therapeutic target for hypertension.

Diabetes
The relation between diabetes-induced endothelial dysfunction and H 2 S impairment is now well established (for review see [77,97]). H 2 S levels have been observed to be reduced in rats with diabetes induced by streptozotocin (STZ) and in subjects with type 2 diabetes mellitus [98][99][100][101][102]. In line with these findings, a high fat diet downregulates and dietary restriction induces (via ATF4) CSE expression [103,104]. Hyperglycemia lowers H 2 S levels due to the high H 2 S catabolism favored by the extremely oxidizing environment or the reduced gasotransmitter production due to a lower expression of the generating enzymes or their inhibition such as oxidative inactivation of the 3-MST at endothelial level [105]. The lack of H 2 S bioavailability supports the accumulation of intracellular ROS, which are not completely scavenged by H 2 S due to its consumption in high-glucosetreated ECs [106]. The consequent oxidative status favors mitochondrial dysfunction and mitophagy, cell damage, and apoptosis [107][108][109][110].
Endothelial dysfunction in diabetes correlates with angiogenesis impairment. CSE expression and H 2 S levels are strongly diminished in wound granulation tissues of obese diabetic mice [111], thus explaining the angiogenesis impairment described in wounds and critical limb ischemia in diabetes [112][113][114]. The availability of H 2 S donors or CSE upregulators could be an innovative therapeutic strategy to promote endothelial function and proper neovascularization of wounds. However, despite the protective effect of H 2 S on endothelial function and wellness, the stimulation of angiogenesis in atherosclerosis plaques by high CSE expression could have a negative outcome, resulting in plaque vulnerability and rupture [115]. The choice of the proper strategy and best control of H 2 S at tissue level is still a critical point to be resolved at the experimental and clinical level.

Antioxidant and Anti-Inflammatory Properties
Several studies document that H 2 S limits vascular permeability, directly or indirectly through antioxidant and anti-inflammatory actions. The multiplicity of the mechanism downstream H 2 S production is schematically reported in Figure 1. Vascular hyperpermeability was inhibited in mice undergoing cardiac arrest and blood-brain barrier disruption following H 2 S inhalation [116]. This protective effect was linked to reduced expression of VEGF and metalloproteinase-9 and increased angiopoietin-1. Another study documented scavenging of ROS and activation of Akt [117]. Data, however, are not all in the same direction, documenting that the final effect is context dependent.
The protective effect of H 2 S in conditions such as hypertension, atherosclerosis, and vascular diabetic complications may be related to multiple actions by the gasotransmitter: H 2 S inhibition of ROS production, blunting of ROS by direct scavenging, upregulation of glutathione, and antioxidant enzymes [101,118]. H 2 S reduces ROS levels in ECs exposed to high glucose, preventing their apoptosis and damage [108,119]. Gene transfer of CSE or administration of exogenous H 2 S in diabetes models reduced ROS levels and improved endothelial dependent vasorelaxation, while CSE KO was responsible for a greater impairment of endothelial function [106]. Many studies support the inhibitory effect of H 2 S on endothelial inflammation [120]. The autocrine/paracrine action of endothelial-derived H 2 S has been documented by endothelial specific deletion of CSE, which predisposes to vascular inflammation and atherosclerosis [50]. In ECs exposed to high glucose, a suppression of NF-κB activity and reduction in ICAM-1 levels were found upon NaHS pretreatment [121]. Moreover, stimulation of ECs with high glucose significantly promotes ET-1 secretion, which was reduced by administration of H 2 S [122].
Recently, inhibition of necroptosis together with ROS downregulation have been described in ECs exposed to hyperglycemia [123]. Inhibition of adhesion molecules such as ICAM-1 in ECs has been described in response to NaHS through NF-κB inhibition [124], while CSE inhibition increased leukocyte adherence to the endothelium [125]. The antiinflammatory activity of H 2 S is not only related to the impairment of adhesion molecules as vascular cell adhesion molecule (VCAM) and ICAM, but also to the inhibition of inflammatory mediator production, such as IL1-β, TNF-α, IL-6 and monocyte chemoattractant protein-1 by ECs and monocytes/macrophages [126,127]. IL-1β, in turn, was found to be increased in atherosclerotic plaques, and induces the phosphorylation of Ser377 and inactivation of CSE [127].
In cultured ECs, the stability of eNOS regulated by miR-455-3p and NO production is induced by H 2 S. Moreover, H 2 S levels and miR-455-3p are incremented in human atherosclerosis plaques, implying that H 2 S could be involved in the miR-455-3p/eNOS/NO pathway controlling atherosclerosis development [128].
Further studies demonstrated that H 2 S reduces the severity of atherosclerosis in a mouse model of disturbed blood-flow, through the upregulation of ACE-2 and increase in Ang(1-7) levels [131]. At cellular level, in LPS-activated ECs, H 2 S promotes the upregulation of the beneficial side of the renin-angiotensin system [131], documenting a multitargeting effect of H 2 S.
An original mechanism of action of H 2 S in controlling endothelial dysfunction in atherosclerosis was recently proposed [50]. In both cultured ECs and in mice, endogenous CSE-derived H 2 S leads to sulfhydration and dimerization of the RNA-binding protein human antigen R (HuR), described to be inhibited in atherosclerosis [50]. The administration of SG1002, a slow polysulfide donor, in ECs isolated from CSE knockout mice, re-established HuR sulfhydration with subsequent inflammatory marker (CD62E) downregulation. Moreover, SG1002, administered to ApoE −/− CSE knockout mice exposed to partial carotid ligation, limited plaque formation, demonstrating an H 2 S-induced antiatherogenic effect [50].
Based on the above results, it appears that H 2 S donors may be a potential promise for the treatment of endothelial inflammation related disorders [132].

Proangiogenic Effect
Several studies report the effects of H 2 S, derived from endogenous biosynthesis or released by exogenous donors, on the process of angiogenesis and in the wound heal-ing context, mainly at low micromolar concentration range, mimicking the physiological concentration of the gasotransmitter [15,68,105]. Additionally, CSE overexpression promotes in vitro angiogenesis [68,133], while CSE silencing, KO or pharmacological inhibition blocks in vitro and in vivo neovascularization responses [133,134]. In addition to the activation of the autocrine eNOS pathway [7], the exposure of ECs to VEGF produces an increase in CSE-dependent H 2 S [134].
Recently, a role of 3-MTS participation in angiogenesis occurrence has been demonstrated in vitro [135]. A connection between 3-MTS-derived H 2 S and EC metabolism has been demonstrated: 3-MTS downregulation decreased mitochondrial respiration and ATP production, increased glucose uptake, and perturbed the whole EC metabolome [135].
Pro-angiogenic effects of H 2 S are evident as increased EC proliferation, migration, and tube formation in vitro. Exogenous H 2 S has been shown also to promote in vivo angiogenesis in models of chicken chorioallantoic membrane and to induce neovascularization in mouse subcutaneous Matrigel plugs [15,134]. In a model of cutaneous burn injury and wound healing, topical administration of a H 2 S-saturated physiological solution has been demonstrated even to significantly increase the wound closure [134].
Therefore, various studies have investigated the cellular signaling pathways involved in the pro-angiogenic effect of H 2 S to discover its molecular targets (Figure 1). Hydrogen sulfide has been shown to activate multiple signaling pathways with a key role in the contribution of EC migration during angiogenesis. Exposure of ECs to H 2 S donors induced increased phosphorylation of Akt, ERK1/2, and p38 MAPK, resulting in their activation [15,134].
There is also evidence about the effect of H 2 S on the activity of eNOS, promoting its phosphorylation on Ser1177 and consequent NO production inside ECs [133,136,137]. Ultimately, H 2 S and NO emerged as being mutually dependent in inducing angiogenesis of ECs and vasorelaxation [68].
In addition, a reverse mechanism appeared to also be effective in controlling EC viability: NO, produced by eNOS, is able to induce CSE activation, resulting in further production of H 2 S in ECs [138].
However, the direct molecular target of H 2 S on angiogenesis remains to be elucidated. From studies of mass spectrometry and additional investigations, it emerged that a disulfide bond between Cys1045 and Cys1024 in the intracellular kinase core of VEGFR-2 serves as a molecular switch for H 2 S to regulate the function of VEGFR-2 [139]. In particular, data revealed that HS − (in aqueous solution, H 2 S is a mixture of H 2 S and HS − ) breaks an inhibitory disulfide bond, bringing VEGFR-2 in an active conformation, probably promoting the activation of downstream signaling [139,140].
An alternative theory about H 2 S interaction with its target molecules is associated with S-sulfhydration, a post-translational modification of cysteine residues, induced by H 2 S on target proteins, involved in signaling pathways [141]. An example of this mechanism is S-sulfhydration of eNOS on Cys443 by NaHS, resulting in increased activity and stability of eNOS and promotion of its phosphorylation, with higher NO bioavailability in ECs to promote their survival and trophism [69].

Wound Healing Promotion
H 2 S has been reported to accelerate the healing of gastric ulcers and skin burn wounds [15,134,142]. Topical application of H 2 S improved recovery from burns in wildtype rats, while genetic ablation of CSE delayed healing in mice [134]. H 2 S improves angiogenesis and wound healing in db/db mice by promoting transcription of VEGF, epidermal growth factor (EGF), HIF-1α and eNOS, by upregulating VEGF and platelet-derived growth factor (PDGF) proteins and receptor phosphorylation [143,144]. H 2 S accelerates wound healing in STZ-induced diabetic mice with the formation of granulation tissue and increased levels of anti-inflammatory factors and VEGF [59]. Additionally, attenuation of inflammation has been attributed to H 2 S, thus improving diabetic wound healing in ob/ob mice [111]. Accordingly, H 2 S facilitates wound closure through the inhibition of neutrophil extracellular traps (NET) release-coupled neutrophil death (NETosis) in db/db mice [114]. Interestingly, H 2 S improved wound healing via restoration of endothelial progenitor cell functions and activation of angiopoietin-1 in db/db mice [113].
Recent epigenetic data document that treatment of ECs with H 2 S or upregulation of CSE rescued migration impairment due to high glucose, through a pathway involving miR126-3p upregulation and DNA methyl trasferase-1 downregulation [145].

Therapeutic Strategies to Improve H 2 S Concentration at Endothelial Level
The use of H 2 S donor compounds or gene therapy to increase the expression of enzymes responsible for the endogenous synthesis of H 2 S has the aim of restoring endothelial function and preventing the onset of pathologies associated with endothelial damage. Several efforts have been made to synthetize effective H 2 S donors showing different H 2 S releasing kinetics and site of action. The main objective has been to control blood pressure and correct endothelial dysfunction, vascular inflammation and redox state and to improve neovascularization and healing of wounds ( Figure 2). Furthermore, H 2 S donor drugs have been evaluated as hypoglycemic agents in type 2 diabetes. H 2 S has been shown not only to protect cells from damage induced by hyperglycemia, but to prevent the onset of type 2 diabetes, preserving the functionality of β-pancreatic cells and regulating the sensitivity of target organs to insulin [146]. Due to the divergent responses induced by different concentrations of H 2 S, it is important for H 2 S donor drugs to maintain the plasma concentration of H 2 S at physiological levels, in the nanomolar order. Therefore, an ideal H 2 S donor should possess two qualities: slow and gradual production and intracellular release of H 2 S. Indeed, NaHS, widely used for experimental purposes as H 2 S donor, is unsuitable for clinical use, due to its fast kinetics of H 2 S release, difficulties to titer the dosage, and its toxic effects [147,148].
A slow-releasing H 2 S donor was developed, GYY4137, demonstrating its vasodilating property in aortic, renal, and cardiac arteries in an L-NAME-induced hypertension model [85]. The antiatherogenic and endothelium-dependent vasodilating effects of GYY4137 were reported in ApoE −/− mice, through decreasing vascular inflammation (lower ICAM-1, IL-6 and TNF-α expression) and oxidative stress [149]. An additional antithrombotic action was demonstrated in mice [150]. Among the others, GYY4137 was proposed in post-ischemia remodeling. The beneficial effects on cardiac functions were correlated to greater vessel density in the infarcted area [151].
The mitochondria-targeted H 2 S donors AP123 and AP39 have been demonstrated to prevent hyperglycemia induced oxidative stress and metabolic alteration in microvascular ECs, suggesting their use in vascular complications of diabetes [119]. The slow releasing H 2 S donor AP39 remarkably reduced systemic blood pressure, heart rate and arterial stiffness in L-NAME treated rats [88].
The positive interaction and synergistic action between NO and H 2 S [152] lead to the development of H 2 S-NO hybrid donor as ZYZ-803, recently reported to promote angiogenesis with a crosstalk between STAT-3 and CAMKII [153]. The authors reported an increased blood flow and vascular density in the hind limbs of mice exposed to femoral artery ligation.
The orally available prodrug SG1002 is an inorganic mixture (sodium polysulthionate) which in vivo was demonstrated to increase both H 2 S and NO levels [154]. Its protective effects have been demonstrated in animal models of atherosclerosis and acute limb ischemia, and patients with heart failure [50,132,155,156].
Among the established drugs, ACE inhibitors bearing a SH group, such as captopril, can promote blood pressure reduction though the sulfhydryl moiety beside the primary pharmacological target. Zofenopril demonstrates vasorelaxant and proangiogenic properties in addition to its ACE inhibitory activity. Indeed, we have contributed to demonstrating that its active moiety, zofenoprilat, can be considered an H 2 S donor and an upregulator of CSE expression at the EC level [80,[157][158][159].
A further therapeutic option is represented by H 2 S-releasing derivatives of a number of drugs, such as non-steroidal anti-inflammatory drugs [160]. Alongside its antithrombotic properties, H 2 S releasing-aspirin was recently demonstrated to exert pro-proliferative and anti-apoptotic actions on cultured ECs together with anti-inflammatory and anti-oxidative features [161].
The use of orally active compounds able to endogenously produce H 2 S, such as Nacetylcysteine (NAC) and taurine, has been proposed, but clinical trials unfortunately were not followed up with published data (Table 1). NAC, a well-tolerated compound, clinically employed to enhance cellular levels of glutathione, is rapidly cleaved in vivo to yield cysteine. On the other hand, in vivo and ex vivo studies demonstrate that the sulfur amino acid taurine markedly and dose-dependently increased the expression of both CSE and CBS, with a higher effect on CSE upregulation [162]. A reduction in blood pressure in patients with prehypertension has been described [163].  [154]).
In the complex, only very few clinical studies on H 2 S donors or enhancers administered in endothelial dysfunction-related disorders are currently registered in NIH Clin-icalTrials.gov as listed in Table 1, but no results have been posted or are available on PubMed.
In a manner similar to NAC, cysteine/cysteine-rich undenatured whey protein supplement improved pressure ulcer recovery in a small group of diabetic patients [164].
The development of natural compounds, present in the diet, as a H 2 S source is interesting, such as polysulfides. Diallyl trisulfide (DATS), diallyl disulfide (DADS) and diallyl sulfide (DAS) are the active principles of the Alliaceae family, such as garlic, which is recognized worldwide as a popular remedy of hypertension. These polysulfides have been demonstrated to exert vasodilating properties in relation to H 2 S release [165], behaving as anti-hypertensives in L-NAME-treated rats [166]. In animal experiments, DATS improved cardiac function in aortic constricted mice, via an upregulation of VEGF, reduced angiostatin and increased myocardial vascular density [167]. Systemic administration of DATS or local transplantation of DATS-treated or CSE-overexpressing bone marrow cells improved capillary density, cell survival and blood perfusion in ischemic hindlimb of db/db mice [168]. Administration of DATS improved neovascularization in STZ-induced diabetic mice through increased NO availability [169].
Erucin [4-(methylthio) butyl isothiocyanate] is a natural isothiocyanate particularly abundant in Eruca sativa Mill. (rocket salad), an edible cruciferous plant belonging to the family of Brassicaceae. Isothiocyanates (ITCs) in general represent a source of different beneficial biological effects on human health, and most are investigated in relation to their chemo-preventive and anti-cancer properties [170][171][172]. Numerous studies demonstrated a general anti-inflammatory and antioxidant activity [173], together with protective properties for the cardiovascular system, where ITCs exhibit vasorelaxing and antihypertensive activity and a protective effect against endothelial dysfunction [174][175][176]. Several biological effects of ITCs may be associated with their ability to release H 2 S inside cells in a slow and long-lasting manner, leading to the definition of "smart H 2 S-donors" [175,177]. H 2 S release from ITCs occurs in a specific manner, depending on the presence of thiols, and it is particularly facilitated in the cell cytosol, where high concentrations of organic thiols, glutathione (in 1-10 mM range), and cysteine (in 30-200 µM range) are present [175,176]. Natural isothiocyanates, including erucin, may therefore represent a possible exogenous source of H 2 S, which, if gradually released, could mimic the physiological concentrations of the endogenous gasotransmitters. On the other hand, they can be the base for the design of synthetic H 2 S donor hybrids with antioxidant property and interesting pharmacological development [178].
Considering the requirement of dressings able to protect ulcers with high exudate levels and to promote wound healing (i.e., in diabetic patients), medicated dressings have been designed and developed. In particular, a functional sodium alginate dressing with H 2 S-releasing properties (SA/JK-1) was fabricated incorporating a pH-dependent donor, JK-1 molecule, into a sodium alginate sponge [179]. The resulting construct provided a moist healing protection able to continuously release H 2 S under acidic pH and absorbing exudate at the wound interface. In vitro, the construct was demonstrated to be biocompatible and effective in improving fibroblast migration and proliferation. When tested in animal model of full thickness dermal defect, SA/JK-1 promoted granulation tissue formation, angiogenesis, collagen deposition, and re-epithelization [179]. Overlapping results were demonstrated by the same group with hyaluronic acid hydrogels doped with H 2 S which was shown to induce M2 macrophage polarization [180]. Another example is represented by silk fibroin porous scaffold loaded with GYY4137, reported to facilitate in vitro bone cell trophism and angiogenesis [181]. These data demonstrate that H 2 S-medicated wound dressing/biomaterial may represent promising strategies for non-healing wounds or bone healing and regeneration. Extensive animal and clinical studies are, however, necessary for assessing their safety and validation.

Concluding Remarks
H 2 S is nowadays considered an important transmitter able to maintain vascular homeostasis. Most of its activities are due to autocrine/paracrine actions by ECs, with a fine control of its plasma concentrations. The availability of H 2 S depends on the activity of endothelial and other cells that express the key enzymes involved in the gasotransmitter release, the reactivity of H 2 S and its inactivation by redox systems and the efficacy of elimination reactions. On top of these, the production of H 2 S by the gut microbiota and intestinal epithelium is important to consider, due to the increasing recognition of circulating molecules coming from this source and finely controlling the cardiovascular system performance [96,182].
There is still demand for the availability of safe and effective synthetic H 2 S donors or enhancers, and natural products or nutraceuticals are helping to fulfil this demand. Indeed, few clinical trials based on H 2 S exogenous sources have been interrupted or have not published their results for unknown reasons.
Since endothelial dysfunction and inflammation continue to be the main causes of morbidity and mortality all over the world, knowledge of the molecular and biochemical mechanisms underlying cardiovascular pathologies and their complications is still required, as well as the definition of new treatment options to prevent endothelial dysfunction or revert cardiovascular disorders [183,184]. The recent pandemic evidenced this unresolved medical need [185].
What it is expected from novel molecules in order to be druggable is the exhibition of H 2 S levels near the physiological ones, and many compounds are actually druggable H 2 S-donors, but it seems that no clinical trials are currently running against endothelial dysfunction (Table 1). Sulfur compounds with natural origin represent helpful pharmaceutical/nutraceutical tools to be used in therapy or as a template for the ideation of advanced H 2 S-donor molecules with improved pharmacodynamic and/or pharmacokinetic properties [132,178].
Although experimental data clearly document a protective effect of H 2 S donors against endothelial dysfunction, further clinical studies are needed. To the best of our knowledge, there are no clearly active clinical trials on patients affected by pathologies due to endothelial dysfunction and treated with H 2 S donors. Even a modest improvement in endothelial function and viability would be a therapeutic success due to the lack of drugs against this diffuse condition predisposing to cardiovascular pathologies.

Conflicts of Interest:
The authors declare no conflict of interest.