Gasping for Sulfide: A Critical Appraisal of Hydrogen Sulfide in Lung Disease and Accelerated Aging

Evolutionary aspect of H₂S

Hydrogen sulfide (H₂ S) gas has been speculated to play a critical and paradoxical role in evolutionary adaptation, contributing to the origin of life and ecological diversification as well as mass extinction (156, 223). Geological evidence indicates that in the Archean Eon (∼3.8 billion years ago), before the “great oxidation event,” the Earth was characterized by oceanic anoxia with chemolithotroph sulfur-reducing bacteria (44, 121, 195). These bacteria use sulfates as electron acceptors for metabolic sustenance, creating H₂S and hence contributing to the generation of a highly sulfidic environment (44, 121, 195). H₂S is therefore hypothesized to have contributed to the formation of the first eukaryotes, suggesting that sulfides catalyzed the formation of the “primordial soup” (155, 156). During the mitochondrial endosymbiotic event, α-proteobacteria established residence in the cytoplasm of sulfur-reducing Archaea hosts, allowing them to adapt to euxinic ocean conditions and utilize H₂S as a sustainable energy source (53, 55). This phenomenon gave origin to mitochondria in classical aerobic adenosine triphosphate (ATP)-producing eukaryotes (53, 55). Following the endosymbiotic incorporation, mitochondria retained their ability to reduce sulfates and produce H₂S, imparting eukaryotic cells the capability to endogenously synthesize and utilize H₂S for energy production and cell signaling (205).

H₂S as an important cellular mediator

H₂S is a well-characterized lipophilic molecule [extensively reviewed in Ref. (223)] and important endogenous gaseous mediator along with nitrous oxide and carbon monoxide (72). It is notorious for the distinct smell of “rotten eggs” and is synthesized in mammalian cells endogenously from the amino acids cysteine and homocysteine by the three enzymes cystathionine-γ-lyase (CTH, EC 4.4.1.1), cystathionine-β-synthase (CBS, EC 4.2.1.22), and 3-mercaptopyruvate sulfurtransferase (MPST, EC 2.8.1.2), although a small portion is also synthesized from nonenzymatic reactions (30, 203). The synthesis is both cytosolic and mitochondrial (Fig. 1); however, during cellular stress, the responsible enzymes translocate predominantly to the mitochondria, where the synthesis concentrates; CSE and CBS may also translocate to the nucleus (68). As a gaseous mediator, H₂S is involved in a plethora of cellular/extracellular signaling processes and has roles in the pathogenesis of various diseases (22). H₂S is thought to exert its signaling activity predominantly through S-persulfidation (often referred to as S-sulfhydration) of cysteine residues with proteins or by reacting with reactive oxygen species (ROS) and reactive nitrogen species or with metalloproteins (22). More recently, H₂S has been proposed to possess pleiotropic effects including “antioxidant,” cytoprotection, pro/antiapoptotic, and anti-inflammatory properties, among others (26, 74, 104, 129, 194, 254). Thus, an imbalance of H₂S bioavailability could be a contributory factor to several chronic conditions, which for lung diseases include asthma, chronic obstructive pulmonary disease (COPD), and pulmonary fibrosis.

OPEN IN VIEWER

FIG. 1.

Endogenous biosynthesis of H₂S in mammalian cells. H₂S is synthesized both in the cytosol and in the mitochondria, predominantly through enzymatic pathways and in minor proportion from nonenzymatic reactions. CBS and CTH are cytosolic enzymes involved in the reverse transulfurylation pathway involved in the conversion of l-methionine to l-cysteine. CBS is responsible for the condensation of l-homocysteine and l-serine to form cystathionine and for the catabolism of cysteine resulting in the release of free H₂S. CTH reacts with cysteine and cystathionine to release H₂S but is also involved in the sulfhydration of cysteine to thiocysteine and further catabolizing the latter to release H₂S. Upon cellular stress, CTH is also known to translocate to the mitochondria. Thiocysteine, along with other “polysulfides” and thiosulfates, can also generate H₂S via nonenzymatic reactions by reduction of GSH to GSSG. Cysteine also feeds substrates for the mitochondrial synthesis of H2S, as is converted to 3-MP by CAT, which subsequently diffuses in the mitochondria. MPST, in the presence of dithiols, receives sulfide from 3-MP to produce an MPST persulfide (3MST-SS) and pyruvate. The sulfide is then transferred to a thiol or other sulfur acceptors, resulting in the formation of a sulfane-bound sulfur. Free H₂S would consequently be formed through the reduction of the sulfane sulfur with a reductant such as thioredoxin. 3-MP can also be alternatively supplied via peroxisomal pathways, whereby d-cysteine is converted to 3-MP by DAO, although this pathway may be limited to kidney and cerebral tissues. 3-MP, 3-mercaptopyruvate; CAT, cysteine transaminase; CBS, cystathionine-β-synthase; CTH, cystathionine-γ-lyase; DAO, d-amino acid oxidase; GSH, glutathione; GSSG, reduced glutathione; MPST, 3-mercaptopyruvate sulfurtransferase. Color images are available online.

H₂S in lung development and physiology

The lungs, like many other organs, constitutively express H₂S synthase enzymes (Table 1) and hence are capable of generating H₂S locally, either spontaneously or upon induction (e.g., by cytokines). There is evidence that H₂S plays important roles in maintaining physiological functions and is also fundamental for organ development. Indeed, CTH expression is implicated in embryogenesis and aging of certain organs, and therefore, the production of H₂S is thought to influence mammalian developmental stages (223, 266). Indeed, supplementation with the slow-release H₂S-generating compound GYY4137 (100 μM) increased endothelial tube number and length in a three-dimensional in vitro model using human pulmonary arterial endothelial cells; H₂S has been proposed to be involved in lung vascularization and alveolarization (141). Systemic application of GYY4137 (50 mg/kg) in an in vivo model of normobaric hyperoxia to induce lung development arrest in mouse pups showed improved alveolarization after 10 days of treatment (142). Similarly, in vitro treatment of primary alveolar type II cells with 100 μM sodium hydrosulfide (NaSH), a H₂S-generating salt, promoted wound healing in response to hyperoxia-induced lung damage, through the activation of the Akt pathway (142). Further evidence also indicates that CTH is indeed developmentally regulated and that its dynamic expression along with CBS is crucial for alveolarization in the developing fetus (141).

H₂S also has critical roles in various lung physiological processes. It is thought to be responsible for the central control of rhythmic respiration via the brain stem (91, 124). The application of 10–400 μM NaSH to brain respiratory centers in neonatal rats induced a diphasic regulation of respiration, whereby in the first stage, respiratory rate was suppressed followed by a second excitatory stage (91). The mechanism by which H₂S regulates respiratory frequency was through the opening of K_ATP channels and activation of adenylate cyclase (91, 158). H₂S appears to modulate electrolyte absorption in the airway epithelium, resulting in improved mucociliary clearance as evidenced by reduced epithelial sodium channel-mediated absorption, potentially via ERK1/2 signaling and decreases in Na⁺/K⁺ ATPase activity (168). Endogenous H₂S also appears to be important in protection against viral infections (99, 126). However, many of the “physiological” interpretations in these studies should be viewed with caution, as conclusions are based on the use of substantial concentrations of exogenous H₂S-generating molecules, often up to 0.4–5 mM (56, 67, 70, 132, 170, 215, 220), which do not translate to the physiological kinetics of H₂S generation and importantly do not reflect relevant physiological concentrations, being orders of magnitude higher than endogenously attainable.

Growing evidence suggests that perturbations in H₂S synthesis and/or its bioavailability are associated with the development of a broad range of lung and many other diseases. While the major focus is on the contribution of abnormal levels of endogenous H₂S in pathological processes, only a few studies report observations on altered expression of endogenous H₂S synthase enzymes in disease (Table 2). Indeed, reports on enzyme activities in disease are lacking; therefore, it remains challenging to establish whether the decrease in H₂S bioavailability is a result of dysregulated enzyme expression profiles, reduced catalytic activity, or consumption of H₂S, for example, by detrimental oxidants. Moreover, it is not possible to exclude that depletion of H₂S pools could also occur through other metabolic pathways that might be differentially regulated in disease and/or through the consumption by detrimental processes, such as proinflammatory oxidants. To highlight the importance of how variations in H₂S bioavailability dictates the difference between a physiological and pathological state, associations between H₂S dysfunction with the development of several respiratory conditions are reviewed below. However, there are several pitfalls in establishing accurately what levels are defined as “physiological” or “pathological.” Vast discrepancies have been reported with respect to “physiological levels” of H₂S in biofluids of healthy individuals. Studies report levels in serum or plasma between 0.3 and 380 μM in healthy controls and up to 600 μM in disease (e.g., in asthma and COPD) (90, 95, 111, 119, 177, 184, 185), yet blood does not emit the characteristic sulfurous odor. In a study of asthmatic individuals, serum H₂S levels were reportedly around 75 μM in the healthy controls, with half that in severe asthmatic patients (31–41 μM) (243). In contrast, others reported nanomolar concentrations in plasma as low as 0.55 μM, and picomolar concentrations in alveolar air (71, 197).

Environmental exposure to inhaled H₂S

H₂S is widely present in the environment and can originate from several sources, which can be either organic (e.g., as products of bacteria decomposition of organic matter) or inorganic (e.g., those originating from natural gas reservoirs, industrial establishments, sewage, sulfur springs, and volcanoes) (223).

Inhalation through the respiratory tract represents the major route of environmental exposure to H₂S in humans. Following inhalation, H₂S will dissolve in the blood and is partly dissociated into hydrosulfide ions. The majority of remaining free circulating H₂S are oxidized to sulfate, methylated by methyltransferases, or interact with thiols and metalloproteins (19).

H₂S odor is detectable at around 0.003–0.02 ppm with concentrations >50 ppm irritating the respiratory tract (175). At higher concentrations (e.g., 600 ppm), H₂S inhalation can lead to pulmonary edema, unconsciousness, and eventually death with concentrations >600 ppm (175). The toxic effects of H₂S are thought to be predominantly due to the inhibition of the mitochondrial respiratory chain, especially cytochrome c oxidase (41). It is well established that beneficial properties of H₂S are strictly concentration-dependent as with every drug and that at low micromolar concentrations H₂S can exert cytoprotective effects, whereas high levels of H₂S lead to detrimental and toxic effects (6, 151). Most toxicological studies are restricted to the acute effect of gas inhalation, either experimentally or associated with occupational exposure (e.g., in industrial paper treatment, pig farming, natural gas, petroleum and sewage exposure, and rayon manufacture) or communities living in proximity to geothermal H₂S (80, 241). The impact of environmental exposure to H₂S on lung disease is not clearly defined. Its pathogenic effects are supported by several community-based and occupational exposure studies that report an increased incidence of asthma in association with increased H₂S emissions (27, 31, 102, 180, 187). Conversely, other studies failed to consistently find any negative association between environmental H₂S exposure and the development of respiratory illnesses and even suggest that moderate exposure could have beneficial effects (13, 14, 77). Interestingly, reports of reduced endogenous H₂S levels in chronic lung diseases such as asthma, combined with the beneficial effects achieved with administration of H₂S at the nanomolar scale, have been consistently reported to have potential diagnostic or therapeutic roles in these conditions (39, 210, 243).

Acute lung injury

A variety of stimuli are known to cause acute lung injury (ALI), such as mechanical ventilation, pneumonia, sepsis, thermal injury, acid aspiration, ischemia, pancreatitis, and blunt or blast trauma (105, 230). Aging is correlated with increased morbidity and mortality rates in ALI, in part not only due to the dysfunctional regenerative capacity of senescent alveolar epithelial cells but also due to a greater neutrophilic inflammation and increased air–blood barrier permeability observed in elderly patients (118, 248). H₂S gas inhalation can exert a protective effect from the exaggerated inflammatory signals observed in ALI. Interestingly, H₂S appeared to induce a suspended animation-like state in small mammals, which protected the organism from hypoxia and free radical damage (11). This is hypothetically attributed to the evolutionary aspects of H₂S, whereby competing with oxygen upon binding on cytochrome c, it adaptively inhibited oxidative phosphorylation under hypoxic conditions (42, 108, 205). More recently, we showed mitochondrial H₂S kept cytochrome c in a reduced state, fueling the electron transport chain and promoting protein persulfidation, (219). ALI is a hypoxic condition that requires critical care and requirement for respiratory support (230). However, this can aggravate lung injury and cause further organ damage; the ability of H₂S to induce a hibernation-like state (hypometabolism) and resistance to hypoxia may be a promising protective strategy to reduce damage inflicted by mechanical ventilation. However, it should be noted that data thus far are only limited to small animals (11). Other studies on ventilation-induced lung injury showed that the application of inhaled H₂S protected against organ damage by preventing edema, apoptosis, and inflammation; furthermore, pulmonary neutrophilia and HMOX-1 expression were inhibited (60). In a rodent model of blunt chest trauma, a hibernation-like state was achieved with inhaled H₂S, allowing the depreciation of metabolic expenditure and partial attenuation of proinflammatory responses. However, H₂S had no impact on neutrophil counts and did not confer any protective effect against blunt trauma lung injury, with no amelioration of the alveolar barrier dysfunction (188). In a rodent model of lipopolysaccharide (LPS)-induced ALI, 80 ppm of inhaled H₂S substantially reduced lung edema and damage due to its anti-inflammatory effects, inhibiting neutrophil infiltration and proinflammatory cytokine release (61). A similar study found that in addition to the anti-inflammatory effects, H₂S inhalation ameliorated ALI by attenuating oxidative stress through the downregulation of heat shock proteins, MAP kinases, and ROS signaling pathways (265). Liu et al. found an association between H₂S and upregulation of glucose-regulated protein 78 (GRP78) and phosphorylated eukaryotic initiation factor 2 α (p-eIF2α), suggesting that H₂S may confer protection to alveolar epithelial cells by promoting the upregulation of endoplasmic reticulum (ER) stress response proteins during the initial stages of ALI (135). Conversely, Francis et al. showed that inhaled H₂S did not confer any protection against ventilation-induced injury, but rather promoted pulmonary edema formation by increasing the expression of adhesion molecules and chemoattractants. In contrast, intravenous administration of sulfide exerted some protective effects by modulating nuclear factor erythroid-related factor 2 (Nrf-2)-mediated antioxidant pathways (66). Similarly, an early study showed that intravenous administration of 14 μmol/kg NaSH induced a potent proinflammatory response, with consequent increase in plasma tumor necrosis factor (TNF)-α and lung myeloperoxidase activity causing pulmonary damage (127).

Fu et al. reported that endogenous H₂S could also protect the lung from ischemia–reperfusion injury in rats by reducing the proportion of malondialdehyde and the activity of superoxide dismutase (SOD) and catalase (69). In a limb ischemia–reperfusion injury model, a downregulation in aquaporin 1/5 (AQP1/5) expression with accompanying inflammation was observed. Supplementation with exogenous H₂S restored normal water transport and decreased inflammation, whereas H₂S synthesis inhibition worsened conditions, suggesting that H₂S played an important role in suppressing pulmonary edema in ALI (171). In a mouse model of burn injury, Ahmad and Szabo demonstrated that supplementation with AP39 (a mitochondria-targeted H₂S-generating molecule) attenuated the increase in myeloperoxidase and malondialdehyde expression in the lung in response to thermal injury (4). In other animal models of smoke and burn lung injury, inhalation of H₂S gas reduced tissue injury and inflammatory responses (57, 58). To further corroborate the role of H₂S in alveolar development, a model of neonatal O₂-induced lung injury was used to show that 37.75 mg/kg/day intraperitoneal injections of GYY4137 protected rat pups from impaired alveolar growth and pulmonary hypertension by promoting capillary network formation, improving mitochondrial functions, reducing oxidative damage, and arresting pulmonary artery smooth muscle cell proliferation (216).

Viral respiratory infections

H₂S also appears to play a role in antiviral protection, although the precise mechanisms involved have yet to be determined. Respiratory syncytial virus (RSV) infection decreased CTH expression and endogenous H₂S synthesis and increased sulfide quinone reductase expression, which resulted in H₂S deficiency in airway epithelial cells (126). Studies with in vivo and in vitro models of RSV infection showed a downregulation of antioxidant enzymes and Nrf2, a transcription factor that regulates their expression and those of H₂S-synthesising enzymes (100). Using Nrf2^−/− mice, it was demonstrated that following RSV and human metapneumovirus infections, they not only developed enhanced airway pathology and higher viral titers but also showed reduced lung levels of antioxidant enzymes and H₂S synthases compared with wild-type mice (100). Hence, these results indicated that in viral infections a dysregulation in Nrf2-dependent induction of H₂S signaling could account for depleted H₂S levels and consequent impaired antiviral responses (100). The well-known anti-inflammatory properties of H₂S were also observed with exogenous H₂S supplementation in RSV infection, showing a significant reduction in NF-κB and IRF-3 binding to RANTES and interleukin-8 (IL-8) endogenous promoters, leading to inhibition of proinflammatory gene expression (126). Conversely, inhibiting CTH aggravated viral titers and exacerbated inflammatory responses to RSV (126), whereas infected Cth^−/− mice had worse clinical features, with increased weight loss, viral replication, airway inflammation, and airway hyperreactivity (AHR) compared with wild-type mice (99). Following intranasal administration of 50 mg/kg GYY4317 to Cth^−/− mice, reduced viral load, neutrophil/macrophage infiltration, inflammasome activation, overall suppression of proinflammatory cytokine release, and airway resistance were observed (99). A more recent study corroborated the implication of H₂S metabolism imbalances in the clinical severity of RSV infection. The authors demonstrated that cigarette smoke (CS) exposure in the same Cth^−/− mice substantially increased viral replication, lung inflammation characterized by severe neutrophilia, airway obstruction, and AHR. These results indicate a more pronounced dysfunction in H₂S synthesis or depletion of bioavailable H₂S upon CS exposure, offering a potential explanation as to why passive smoke may be a risk factor for RSV infection-associated pulmonary complications (101). A thiol-activated H₂S-releasing molecule was also tested in RSV infection and showed potent antiviral activity in vitro, but only moderate efficacy in vivo. It was speculated that the in vivo activity was not substantial due to the depletion of glutathione (GSH) pools during infection, which is essential for thiol-activated donors as they are reported to rely on cysteine residues and thiols such as GSH for H₂S generation, rather than hydrolysis (18). These findings further stress the importance of bioavailable H₂S in endogenous antiviral defenses while also suggesting that exogenous H₂S-generating molecules could be a potential therapeutic strategy for paramyxovirus infections, having important implications in the prevention of the development of severe viral-induced asthma (83). Interestingly, evidence indicates that supplementation of human embryonic lung fibroblasts with 60 μM allitridin (diallyl trisulfide), a naturally occurring organosulfur compound and apparent source of H₂S, inhibited cytomegalovirus replication, which causes opportunistic infections in the respiratory tract of immunocompromised individuals (217, 256), although control experiments such as hydrolysis/end products were not conducted. H₂S offered the same antiviral protection from other RNA viruses such as influenza A and B virus, Ebola virus (EBOV), Rift Valley fever virus (RVFV), Crimean-Congo hemorrhagic fever virus (CCHFV), and the far-east tick borne flavivirus, with GYY4137 treatment (although 10 mM), downregulating the expression of viral mRNA and proteins, including neuraminidase, hemagglutinin and nucleoprotein, and viral titers (17). Very high concentrations (10 mM) of GYY4137 and NaSH consistently attenuated NF-κB and interferon proinflammatory responses to virus (17), suggesting that “H₂S” may confer some antiviral activity at sufficiently high levels. Although some reduction in viral RNA expression was observed in RVFV and CCHFV, no effect was observed for EBOV and RSSV, suggesting that high concentrations of H₂S could act on specific viral targets. In support of this, increased viral replication was observed with pharmacological inhibition or genetic deletion of CTH (99, 126); excitingly, this indicates that CTH-derived H₂S represents a novel endogenous antiviral defense pathway that we are pursuing. It has also been suggested that since a downregulation of M2 protein was also observed with GYY4317 (10 mM) treatment, and given the role of M2 in nuclear transport of viral RNA and proteins and assisting with virus budding, “GY4137” decreased viral titers by interfering with the assembly and release mechanisms in the viral replication process (17). However, it should be noted that H₂S “release” from GYY4137 (10 mM in vitro or 200 mg/kg i.p. in vivo) was not assessed in these virus studies (17, 95, 121), and in the absence of control experiments, non-H₂S-mediated effects particularly from these high concentrations/doses cannot be ruled out at present. These include the effects of the parent compound, hydrolysis products, and/or intermediates formed during/after hydrolysis and H₂S generation, equimolar morpholine, or carbon monoxide produced from near equimolar dichloromethane present in the GYY4137 crystal lattice structure (7, 234).

In December 2019, a novel coronavirus (severe acute respiratory syndrome coronavirus 2, SARS-CoV-2) caused a global outbreak of respiratory illness ranging from paucisymptomatic manifestations to severe fatal pneumonia, resulting in more than 788,500 deaths worldwide as of August 21, 2020 (125, 237). Interestingly, severity and mortality from SARS-CoV-2-induced pneumonia was significantly greater in patients with serum H₂S levels lower than 150 μM (176). Furthermore, serum H₂S levels negatively correlated with interleukin-6 (IL-6) and CRP, whereas it positively correlated with lymphocyte counts, suggesting that serum H₂S levels could represent a potential marker for predicting outcome and severity in SARS-CoV-2 pneumonia (176). Given these very recent observations, and the apparent antiviral effects of H₂S-generating molecules against single-stranded RNA viruses and its protective role in lung damage, modulating cellular H₂S generation could represent a promising therapeutic strategy for this novel coronavirus disease (COVID-19) (246). H₂S may be able to counter SARS-CoV-2 infection by inhibiting viral entry, replication, or viral-induced hyperinflammation and lung damage (246).

Asthma

Chronic and life-course persistent asthma has been linked to a decrease in leukocyte telomere length and therefore to accelerated aging (21). Since nitric oxide presents itself as a useful biomarker for asthma, the notion of using gases as noninvasive biomarkers has also extended to investigating H₂S concentrations in body fluids as a discriminant endpoint for characterizing various asthma phenotypes (2). A few studies have examined levels of H₂S in asthma. Wang et al. examined the evident prevalence of childhood allergic compared with the adult-onset asthma and questioned age-dependent mechanisms that are accountable for this propensity using a murine ovalbumin (OVA) sensitization and challenge model (221). The authors demonstrated that the allergic asthma phenotype was exacerbated in young mice and displayed a lower expression of CTH compared with adults and concluded that a higher incidence of allergic asthma at young ages is associated with a lower production of H₂S (221). In another rodent model of asthma, decreased baseline expression of endogenous H₂S and its synthesizing enzymes were also reported (130). In a similar OVA-induced asthma rodent model, the administration of NaSH reduced inflammatory cell infiltration and AHR, with consequent decrease in mast cell activation and degranulation in the lungs (39, 181). Additional in vivo studies have demonstrated that intranasal administration of both slow- and rapid-release H₂S-releasing molecules (GYY4137 and NaSH, respectively) effectively reduced interleukin-1β (IL-1β) and TNF-α secretion in bronchoalveolar lavage fluid and prevented parenchymal edema in a murine model of LPS-induced acute lung inflammation (113). Consistent with this, in the OVA model, airway inflammation and AHR were comparable to responses observed with genetic ablation of CTH. In contrast, supplementation of NaSH reversed the OVA-induced inflammation and AHR and rescued Cth ^−/− mice from the severe phenotype (255). H₂S levels also appeared to be negatively correlated with human asthma severity and increased body mass index (BMI) combined with refractory AHR (185, 231). To corroborate the BMI finding, studies in murine models with high-fat diets also demonstrated that hyperlipidic intake contributes to downregulation of H₂S-synthesizing enzymes and consequent decreases in lung H₂S levels, associated with sustained chronic inflammation (163). Interestingly, vitamin D3 increased H₂S concentrations in mouse brains and kidneys, suggesting a possible correlation between vitamin D and H₂S depletion and steroid resistance in obesity (239). Airway remodeling is a well-known consequence of inflammation in asthma, characterized by increases in airway smooth muscle (ASM) mass due to hyperproliferation of ASM cells that contribute to inflammatory responses and release of IL-8. The administration of H₂S-generating compounds such as GYY4317 and NaSH in an ASM ex vivo model inhibited cellular proliferation via MAPK and ERK-1/2 inhibition and suppressed IL-8 release; in contrast, opposing effects were observed with inhibition of H₂S-synthesizing enzymes (165). Mouse models that recapitulate different phenotypes of asthma, particularly severe steroid-resistant asthma (a major issue in asthma management), have recently been developed and can be used to define the role and benefits of therapeutic intervention with H₂S (85, 115, 116).

Several small clinical studies have clearly shown perturbed bioavailability and/or biosynthesis of H₂S in mild-to-moderate and severe asthma. Exhaled H₂S levels were lower in patients with severe asthma compared with healthy controls and lowered further during exacerbation before increasing with recovery (89, 257). Other reports showed that in children and adults with asthma, serum and plasma H₂S levels were decreased compared with healthy individuals, and negatively correlated with reduced pulmonary function parameters (210, 243). Interestingly, H₂S bioavailability was lower in both acute and severe asthma compared with healthy controls and stable asthmatics, and negatively correlated with increased percentage of macrophage infiltration in lungs, decreased lung function, and moderate-to-severe episodes of exacerbations (243). A negative correlation was also found between exhaled H₂S and sputum eosinophil counts, which suggest a useful role of exhaled H₂S as a biomarker for eosinophilic asthma (257). Consistent with this, Suzuki et al. reported significantly lower H₂S serum levels in exacerbating compared with healthy or stable asthmatic patients (202). H₂S levels were substantially lower in uncontrolled asthmatic patients, showing an association between lower sputum H₂S concentrations and increased neutrophils, suggesting that H₂S could be a good indicator and predictor of asthma severity in difficult to manage patients and a biomarker for asthma management (202). Conversely, Saito et al. showed that serum and sputum H₂S were threefold higher in patients with asthma compared with healthy controls, and serum H₂S levels were 10-fold higher (up to ∼600 μM) than those observed in sputum without any distinction between severe and nonsevere phenotypes (185). This increase may reflect the induction of endogenous H₂S synthesis as a compensatory anti-inflammatory mechanism, although in this latter study blood levels far exceeded by an order of magnitude previously published levels (185). Furthermore, a negative correlation was observed between increasing H₂S levels and lung function. The authors also reported that serum H₂S levels were 10-fold higher than those observed in sputum and hence suggested that sputum might be a better matrix for quantifying H₂S levels specific to asthma, as reflective of its generation by ASM and less influenced by systemic nonrespiratory diseases (185).

Allergic rhinitis

Although allergic rhinitis (AR) is not classed as a lung disease per se, it is linked to the development of other pulmonary diseases, such as COPD, and therefore relevant to the scope of this review. The incidence of rhinitis in geriatric patients is reported with increasing frequency, as aging and correlated alterations in immune functions potentially influence nasal physiology and contribute to the pathogenesis of AR (167). Literature on the involvement of H₂S in AR is scant but consistent with other lung-related reports, with contrasting studies between rodents and humans. Shaoqing et al. showed in a guinea pig model of OVA-induced AR that CTH mRNA expression in the nasal mucosa of sensitized animals was significantly lower, with a consequent decrease in serum H₂S levels (189). This was further exacerbated upon administration with CTH inhibitor PAG (propargylglycine) and rescued by the H₂S-generating molecule NaSH (189). Using the same model, the authors showed that a decrease in H₂S levels inversely correlated with the expression of heme oxygenase-I (HMOX-1) in the nasal mucosa, suggesting impairment of redox balance and antioxidant responses. Moreover, the administration of carbon monoxide exacerbated the allergic phenotype with a further reduction in serum H₂S levels and CTH expression, although enzyme activity was not measured (251). Conversely, studies in humans showed that patients with mild or moderate-to-severe persistent AR had a higher expression of CTH and CBS in the nasal mucosa (with profuse immunoreactivity for CTH and CBS in the epithelium and submucosal glands) and elevated plasma H₂S levels compared with healthy controls (where immunoreactivity for CTH was localized exclusively in the vascular endothelium), although without any significant correlation with disease severity (162). The authors suggested that the localization of H₂S-synthesizing enzymes could account for the development of AR symptoms, such as rhinorrhea and sneezing (162), since the predominant distribution of CBS in the nasal epithelium and submucosal glands and H₂S may be involved in regulating nasal secretions by modulating ion channels (162). Similar results were observed in patients with chronic rhinosinusitis with and without polyps, with elevated H₂S levels in the mucosal sinuses and translocation of H₂S enzymes to the superficial epithelium and submucosal glands. Furthermore, higher H₂S concentrations in AR positively correlated with increases in cytokines, in particular interleukin-5 (IL-5), TNF-α, and interferon-γ (IFN-γ) (94). However, it remains unclear whether H₂S levels were driving the inflammatory (or allergic) response or were elevated as a compensatory protective mechanism.

Chronic obstructive pulmonary disease

There is increasing evidence linking COPD to aging. Indeed, as observed in senile emphysema, lung functions deteriorate along with structural changes and increased pulmonary inflammation, strongly suggesting a relationship between aging and disease development (98). Furthermore, environmental factors, such as CS, contribute to accelerate the aging processes, representing a major risk factor for COPD (23). CS induces oxidative stress and inflammation in the lungs at least in part by inhibiting H₂S synthesis and decreasing its bioavailability (201). Indeed, pharmacological inhibition of H₂S synthesis exacerbates CS-induced inflammation and lung damage (38). Phenotype prediction through monitoring exhaled H₂S levels is also feasible in COPD. Although the relationship between exhaled H₂S and COPD severity needs to be further clarified, distinct inflammatory signatures could be predicted by monitoring endogenous H₂S levels. Studies with a murine model of smoke-induced emphysema following chronic CS exposure for 12–24 weeks demonstrated decreases in the expression of CTH and CBS in the lungs (82). Similarly, in another rodent model of chronic CS exposure, pharmacological inhibition of CTH with propargylglycine (PAG) exacerbated epithelial damage and emphysema as well as tracheal hyperreactivity in response to acetylcholine or potassium chloride in organ baths; the administration of exogenous H₂S reversed ASM contraction demonstrating an anti-inflammatory and bronchodilatatory effect (38). Additionally, sulfide salts such as NaSH also reduced tobacco smoke-induced lung inflammation, reversed emphysema, and improved lung function in mice (38, 82, 220). More recently, a possible regulatory mechanism of H₂S in airway remodeling in COPD has been proposed (79). CS-induced increase in airway α-smooth muscle actin (α-SMA) expression and collagen deposition was suppressed by supplementation with NaSH in both mice and human bronchial epithelial cells when administered at 40 mg/kg and 400 μM, respectively (79). NaSH reversed airway remodeling by promoting the upregulation and activation of SIRT1 signaling pathway, which prevented epithelial–mesenchymal transition and oxidative damage through modifying transforming growth factor (TGF)-β1-mediated SMAD3 transactivation (79). Further studies with mouse models that are reflective of human CS exposure that develop the major features of the human disease will be valuable in elucidating the roles of H₂S including in exacerbations and maximizing therapeutic interventions (20, 84, 106, 198).

There is also evidence that CS induces ER stress and ER-mediated apoptosis combined with a suppression of H₂S biosynthesis, which suggests a crucial role of the ER in COPD development (196, 227, 261). These effects were reversed by supplementation with exogenous H₂S and were aggravated by pharmacological inhibition of H₂S synthesis (46, 132). Indeed, in vivo intraperitoneal administration of 14 μmol/kg NaSH reversed CS-induced overexpression of GRP78 and caspase-12, with similar outcomes in vitro in human bronchial epithelial cells (132). These studies suggest that endogenous H₂S may protect against smoking-induced lung injury and highlight again the therapeutic potential of pharmacological H₂S supplementation.

In a small study of 77 subjects with COPD, lower concentrations of exhaled H₂S were found in COPD patients with significant eosinophilia, worse lung function, and frequent exacerbations (Table 3) (258). However, the authors did not find any correlation between exhaled H₂S and levels of inflammatory mediators IL-8, TNF-α, and leukotriene B4. It is important to note that the study was limited by its cross-sectional nature without controls and by the assessment of a limited number of cytokines and chemokines, which do not cover a broad picture of the systemic inflammation (258). A study by Saito et al. demonstrated that sputum-to-serum H₂S ratio positively correlated with sputum neutrophils, IL-6 and IL-8 concentrations in both sputum and serum. They also showed that the ratio was increased during an exacerbation, although again the micromolar levels measured in both tissues are questionable, particularly since H₂S in sera substantially exceeded the concentrations previously reported in blood (184). Investigations of plasma H₂S levels in COPD smokers revealed that concentrations were substantially lower compared with nonsmokers and healthy individuals, suggesting that CS may contribute to disease severity by depleting endogenous H₂S pools (243). This is in line with other studies showing the association of CS with bioavailability of H₂S, in which altered metabolic profiles and decreased synthesis of H₂S have been identified in smokers and patients with CS-associated COPD (184, 201). Sun et al. demonstrated that supplementation with GYY4317 (0–500 μM) not only conferred protection against CS-induced oxidative damage and inflammation but also showed that generated H₂S restored steroid-induced TNF-α inhibition in alveolar macrophages (201).

H₂S also exerts anti-inflammatory and antiproliferative effects on ASM cells (166). H₂S-generating molecules inhibited fetal calf serum-induced proliferation in ASM and suppressed IL-6 and IL-8 release, although the effects were less in cells from COPD patients compared with smokers and nonsmokers without COPD (166). To further demonstrate that H₂S may be involved with COPD activity and severity, Chen et al. reported a lower serum H₂S concentration in cohorts with acute exacerbation chronic obstructive pulmonary disease (AECOPD) compared with stable COPD patients, correlating with predicted forced expiratory volume 1 (FEV₁); however, even lower levels were observed in smokers within both the AECOPD and healthy patient pools compared with nonsmokers (37). Furthermore, the authors reported that H₂S levels negatively correlated with proportion of sputum neutrophils and positively correlated with lymphocytes and macrophages, suggesting an important role of H₂S in modulating neutrophilia and consequent airway obstruction (37). Similar results were observed in another study, where serum H₂S levels were lower in asthmatic and COPD patients with acute exacerbations and negatively correlated with neutrophils in sputum (103). Smokers and COPD patients have also been reported to have a lower lung expression of CTH at a translational level and CBS at a transcriptional level (Table 2), further corroborating a connection between the disruption of endogenous H₂S synthesis and the pathogenesis of COPD (201). The impairment of H₂S synthesis was also associated with a decrease in antioxidant defenses, through decreases in GSH and SOD expression (201).

Interstitial lung diseases

Idiopathic pulmonary fibrosis (IPF) is an aging-associated progressive and irreversible lung condition, whereby telomere shortening, mitochondrial dysfunctions, and accelerated cell senescence have been reported in patients' lungs (161). Perturbed H₂S synthesis and/or depletion has also been reported in pulmonary fibrosis (PF) (5, 200, 201). A small-scale clinical study on 30 patients with IPF showed that plasma H₂S levels were significantly lower than healthy controls and inversely correlated with serum concentrations of hyaluronan, type III pro-collagen and laminin (markers of extracellular matrix remodeling) (5). Similarly, another study reported lower plasma H₂S concentrations in 27 IPF patients compared with controls, which negatively correlated with lung function; additionally, H₂S levels appeared to further decrease upon disease progression (200). Although clinical studies to date are small, they are supported by animal models. Bleomycin-administered rats displayed fibrotic lung lesions and dysfunctional CTH activity and H₂S synthesis, which were at least partly responsible for PF development (63). Promising therapeutic applications are represented by studies in which administration of NaSH ameliorated PF in rodent bleomycin models, suggesting that the beneficial effects could be mediated by H₂S through its antioxidant properties and by attenuating lipid peroxidation damage (29, 63). Interestingly, homozygous Cbs^−/− mice manifested fibrotic injuries and air space enlargement in the lungs, with increased expression of HMOX-1, collagen I, TGF-β1, and α-SMA, but surprisingly an absence of inflammatory cell infiltrates and unremarkable alteration in proinflammatory gene expression (81). It is important to note that homozygous ablation of Cbs manifests with a neonatal lethal phenotype due to hepatic dysfunctions, with Cbs ^−/− mutants surviving at most for up to 5 weeks after birth with severe growth retardation (225). However, Hamelet et al. observed that mutant survival could be increased to 4 months of age, by supplementing mice with a high choline chloride diet (81). Similarly, H₂S was also implicated in PF for its effect in decreasing ERK phosphorylation and inhibiting TGF-β responses, hence suppressing lung fibroblast migration and proliferation, and myofibroblast differentiation, limiting collagen deposition and alveolar damage (62). Accelerated lung aging is associated with PF, and H₂S has been implicated in RNA splicing events controlling cellular senescence (35, 122). Indeed, depletion of H₂S and H₂S-mediated signaling is observed in aging processes across species, further corroborating the likely importance of H₂S in controlling these pathways in PF (164, 266). Cigarette smoking, which depletes lung H₂S-synthesizing ability and H₂S levels (82, 201), is a significant risk factor for IPF (15), and some studies have demonstrated that CS results in depleted pools of endogenous H₂S, further stressing the correlation between decreased H₂S availability and development of lung inflammation and PF (38, 201). Inflammatory processes that characterize IPF include a marked elevation of secreted cytokines such as IL-6, IL-8, and other inflammatory cytokines (160). In a rat model of passive CS-induced fibrosis, NaSH showed protective effects against PF by attenuating lung inflammation and oxidative stress. H₂S reduced serum IL-6, IL-1β, and TNF-α levels by suppressing NF-κB activation and inhibiting the phosphorylation of ERK1/2, p38 MAPKs, and JNK, and upregulated Nrf-2-dependent expression of the antioxidant genes HMOX1 and TRX-1 (264). Collectively, these observations render the depletion of H₂S levels in the inflammatory processes in PF relevant in disease progression, which could be potentially rescued by H₂S supplementation, although this must be determined experimentally through therapeutic rather than prophylactic intervention.

Cystic fibrosis

To date, only limited studies have investigated the role of H₂S in cystic fibrosis (CF), offering some evidence that H₂S imbalances are implicated in CF (whereby levels in sputum correlated with the reduced state and pH acidification of mucus) (43). This suggests that H₂S could be a biomarker of redox potential in the lungs that is likely to reflect physiological adaptation to the metabolically complex microbiota that characterize CF lungs (25). Although Cowley et al. did not measure exhaled H₂S levels, they established a negative correlation between sulfide and decreased likelihood of hospitalization and oxygen therapy requirement, suggesting that H₂S could be a positive prognostic factor in CF patients (43). It remains to be determined whether H₂S is a proinflammatory mediator driving the disease and potentiating tissue damage or whether it exerts a protective response to lung injury. In a study by Kamboures et al., sulfides in breath of CF patients were measured, and the authors reported a higher level of carbonyl sulfate, which has some potential to be rapidly hydrolyzed to H₂S, which correlated with poor lung function. Thus, they concluded that breath sulfides may be a marker of respiratory bacterial colonization in CF (109). Interestingly, H₂S modulates Na⁺ transport across the pulmonary epithelium. H₂S induced a reversible inhibition of Na⁺ transport and absorption by reducing Na⁺/K⁺ ATPase currents, through the inhibition of basolateral calcium-dependent K⁺ channels. Since in CF, Na⁺ absorption in the airway epithelia can promote mucus thickening, H₂S supplementation could represent a viable pharmacological approach for modulating transepithelial Na⁺ transport imbalances (9). Arguably, however, such effects were observed following the supplementation of substantial concentrations of NaSH (between 300 μM and 1 mM for electrophysiological experiments), which are orders of magnitude higher than physiological H₂S levels making it difficult to accurately extrapolate biological conclusions (9). It is important to note that oral and inhaled N-acetylcysteine (NAC) has been considered for the management of CF due to its mucolytic properties, although several trials failed to find robust evidence of therapeutic benefits (52, 78, 143, 174, 211). Interestingly, studies have found that in addition to changing rheological properties of mucus, NAC exerts antioxidant and anti-inflammatory effects, which may be achieved through its deacetylation that increases cellular cysteine pools, potentially triggering intracellular H₂S and sulfane sulfur production in mitochondria by MPST and sulfide:quinone oxidoreductase (33, 59). This further supports the potential therapeutic use of H₂S supplementation in CF. Nevertheless, more research is required to establish correlations between H₂S levels and lung function in CF.

Lung transplantation

Recent evidence demonstrates that supplementation of organ preservation fluid with H₂S-generating molecules improved post-transplant graft survival and functions, protecting tissues from ischemia–reperfusion injury in models of syngeneic and allogeneic renal transplant (96, 107, 137, 138). These findings suggest that the application of exogenous H₂S in the form of synthetic delivery molecules to preservation fluid could also be extended to lung transplantation. The beneficial roles of H₂S in lung transplantation are mostly related to its protective effects in ischemia–reperfusion injury, such as increasing perfusion flow rate and lung compliance and decreasing oxidative damage and lung injury (69). It has been demonstrated that 150 ppm of inhaled H₂S gas improved graft function in a rodent model of ex vivo lung reperfusion, suggesting a potential and complementary role for H₂S in conventional preservation practices. Following transplant, animals had improved lung oxygenation, ventilation, and lowered pulmonary arterial pressure, which correlated with better transplantation outcomes. Furthermore, H₂S lowered ROS levels during reperfusion and maintained metabolic functions by preserving mitochondrial cytochrome c oxidase activity (75). Although elevated concentrations of H₂S are typically associated with inhibition of cytochrome c oxidase, the authors observed an increase in its activity upon H₂S gas inhalation, speculating that under hypoxic conditions low concentrations of H₂S could act as an electron donor, increasing ATP and preventing apoptosis (75). In addition to protecting lungs from ischemia–reperfusion injury, showing a decrease in neutrophil infiltration and alveolar edema after pulmonary artery ligation, H₂S also prolonged allograft survival in a swine model of MHC-mismatched orthotopic lung transplant and reduced the upregulation of proinflammatory cytokines (183).

Lung cancer

Although the relationship between H₂S and cancer is still unclear and under debate, it has been suggested to be protective in cancer through proapoptotic activities, which is also true in lung cancers. Baskar et al. investigated the clastogenic effects of H₂S in human primary lung fibroblasts. They showed that NaSH supplementation in a range of concentrations between 10 and 75 μM induced p53 and formation of micronuclei and G₁ cell cycle arrest, upregulated the expression of apoptotic markers, and triggered mitochondrial cytochrome c release into the cytoplasm and the translocation of Bax from the cytosol to mitochondria. These findings implicate H₂S in altering the cell cycle and inducing cell death, and possibly being an important proapoptotic molecule against cancer cells (12). In a study investigating the therapeutic effect of H₂S-releasing nonsteroidal anti-inflammatory drugs (NSAIDs) on multiple adenomatous cancer cell lines, a potent inhibitory effect on cell growth in A459 lung carcinoma cells compared with the corresponding unconjugated NSAIDs was observed (34). The authors suggested that the effects may be COX-independent and speculated that cancer growth inhibition may be inherent to other potential regulatory mechanisms involving cell cycle phase transition and apoptosis. However, they did not demonstrate cellular mechanisms, and hence, it is difficult to reliably conclude that the effects were H₂S-dependent. Furthermore, these authors envisaged that the potentiation of the effects of H₂S-releasing NSAIDs over the unconjugated donor may be due to a change in H₂S generation, (34) although there is no evidence, or chemical explanation, to support this. In contrast, there are several reports showing that H₂S may be protumorigenic. Upregulation of CBS has been observed in colorectal cancer cells, suggesting that elevated H₂S stimulates cancer cell growth and inhibitors could be useful therapeutic agents (87, 203, 204). Notably, the expression of all H₂S synthases (CBS, CTH, and MPST) is highly upregulated in lung cancer tissues with consequent increments in endogenous H₂S, whereas inhibition of these promoted tumor suppression (206). Another study showed that HS-NBD, a fluorescent 3-hydroxyflavone H₂S detection probe, promoted Nrf2/HMOX-1 antioxidant pathway activation and inhibited the growth of A459 cells (73), presumably through removal of endogenous H₂S. Furthermore, contrary to the study of Baskar et al., endogenous H₂S can contribute to lung carcinoma growth by aiding mitochondrial DNA repair via the activation of EXOG gene through S-persulfidation; thus, the DNA repairing activity of H₂S would in turn counter the efficacy of chemotherapeutic drugs that induce DNA damage (206). The same study showed that CTH and CBS inhibition by amino-oxyacetic acid (AOAA) sensitized lung adenocarcinoma cells to antineoplastic drugs by inhibiting mitochondrial DNA repair mechanisms. Other mechanisms by which H₂S may promote lung cancer could include its involvement in generating vitamin B₁₂-induced-S-adenosylmethionine, which acts as a first substrate of METTL1/3 complex, contributing to the induction of m⁶A methylation and hence promotes tumor proliferation and lung adenocarcinoma progression (131, 212). Based on all these studies, we conclude that pharmacological inhibition of H₂S may be beneficial against tumor growth and sensitization toward antineoplastic agents.

Pharmacological activity of H₂S

Outside lung disease, H₂S is known as a broadly anti-inflammatory agent that acts by reducing extravasation of leukocytes by modulating their adhesion to the vascular endothelium (54, 254). Furthermore, it induces neutrophil apoptosis and promotes a shift in macrophage function toward a pro-resolution phenotype and suppression of TNF-α secretion (51, 128, 233). Other anti-inflammatory mechanisms are evidenced by the efficacy of H₂S in inhibiting cyclooxygenase 2 (COX-2) and phosphodiesterase activities, as well as inhibiting the proinflammatory transcription factor NF-κB (128, 233), presumably through binding to heme and zinc centers of these proteins. Also, H₂S signaling has been implicated in a plethora of acute and chronic inflammatory processes, including sepsis, trauma, edema, gastrointestinal inflammation, pancreatitis, burn injuries, arthritis, neurogenic inflammation, and notably for this review, lung inflammation (235). There are several approaches toward the pharmacological modulation of H₂S signaling, which can be achieved either via supplementation of H₂S or through the inhibition of its biosynthesis. Common “donors” and inhibitors are summarized in Table 4.

H₂S-generating molecules

It is becoming increasingly evident that several lung diseases are associated with reduced bioavailability and depletion of endogenous H₂S pools or dysfunctional H₂S biosynthesis. Thus, the most promising therapeutic approach for modulating H₂S dysfunction is to supplement using exogenous H₂S-generating molecules. These and the concentrations used in lung conditions are summarized in Table 5. Several experimental studies in vitro and in vivo have demonstrated the therapeutic potential of administering exogenous H₂S sources in lung diseases in resolving inflammation, suppressing airway remodeling, and improving lung functions (28, 29, 39, 49, 113, 120, 132, 165, 181, 255). However, it is important to note that frequently, H₂S-generating molecules are used at supraphysiological concentrations, and hence, it is difficult to reliably establish the potency of H₂S. In addition, the choice of H₂S-generating molecules is often suboptimal with respect to sulfide release kinetics, which could account for common biphasic effects. The uncontrolled release of H₂S combined with the use of micromolar concentrations could also account for variability in pharmacological effects and discrepancies reported in the literature. Moreover, it is not always clear whether biological effects are induced by the released H₂S or from sulfide-releasing functional groups of the molecules, further stressing the importance of exercising caution in using adequate concentrations of such molecules and appropriate controls (159). With respect to H₂S-generating molecules, some may have more advantages. While sulfide salts such as NaSH and Na₂S are widely used, these are considered inadequate due to the nonphysiological kinetics of H₂S generation, especially during systemic delivery where H₂S is instantly generated as a bolus with a short half-life (178, 233). Furthermore, only one third of sodium sulfides dissociate to free H₂S in aqueous solution at physiological pH, whereas the remaining two thirds exist as hydrosulfide anions (175). However, it has also been demonstrated that the half-life of sulfide salts in solution is between 0.5 and 5 min, and most dissociated H₂S is passively and exponentially lost via volatilization rather than oxidation (45). The instantaneous pH-dependent dissociation of these salts in aqueous solution, followed by a rapid decline in H₂S, could explain the bell-shaped curve that characterizes some biphasic pharmacological responses observed when using H₂S sodium salts often at high concentrations (233). Moreover, the instant generation of sulfide from inorganic salts is in stark contrast with the physiological synthesis kinetics at which H₂S is endogenously produced enzymatically, potentially exposing cells to initially toxic concentration in the first instants when the salts dissipate and local pH changes, especially when administering a second bolus, which exacerbates the spike in concentrations (159). Additionally, controlled release of H₂S is further hindered by impurities in sodium sulfides, such as extensive contamination with polysulfides or oxidizing metal ions (150), which are frequently left uncontrolled. GYY4137 is often considered a slow-releasing H₂S molecule, and the kinetics of sulfide release are highly dependent on pH (203), with ∼0.1 μM of H₂S per hour released at physiological pH at 25°C from 1 mM of the drug (133, 179, 234). However, many authors fail to specify the crucial pH conditions used when administering NaSH and GYY4317, and hence, it is difficult to accurately extrapolate and control the rate of H₂S dissociation and frequently effective concentrations/doses of the resulting H₂S generated are not measured (223). Furthermore, it is unclear whether GYY4317 demonstrates biological activity independent of H₂S release since its effects are observable shortly after administration when most of the compound remains un-hydrolyzed in the parent form, further stressing the importance of implementing adequate experimental controls (7, 234). Dithiolethiones (i.e., ADT), unlike GYY4137, release H₂S via pH-independent hydrolysis, which is typically catalyzed upon reactions with biological material, such as thiols and esterases, resulting in slow, sustained, and controlled H₂S generation in the micromolar range when compounds are used at millimolar concentrations (112, 133, 169). Moreover, anethole dithiolethione (ADT-OH) is another class of H₂S-releasing molecules amenable to esterification reactions with other therapeutic compounds to generate novel H₂S drugs, including H₂S-donating NSAIDS (153). The ease of esterification allows “donors” such as ADT-OH to be developed into molecules that target H₂S delivery to organelles such as mitochondria when coupled to triphenylphosphonium (TPP+) such as AP39 (213). Prodrugs are another chemical tool for a controlled and tunable H₂S delivery, which make use of molecules that liberate H₂S by different mechanisms such as reactions with light, pH, esterases, and thiols (249, 252). Molecules are modified at the ester group to increase susceptibility to esterases, and subsequently undergo lactonization reactions to release H₂S (263). Furthermore, these molecules can be conjugated to NSAIDs for generating hybrid prodrugs (263). Additionally, photolabile molecules could also be used to release H₂S in a controlled manner upon light irradiation (244). Controlled release can also be achieved with slow-releasing molecules such as thiol-activated “donors” such as gem-dithiol-based molecules. These generate H₂S independent of hydrolysis and are reported to rely on thiols such as cysteine residues and GSH to achieve slower and more consistent generation (262). Similarly, naturally occurring sulfides such as molecules in garlic extracts (e.g., disulfide) have also been proposed to generate H₂S only in the presence of thiols (16), which may preclude their use in conditions with extensive oxidative stress such as COPD and asthma where thiols are depleted or oxidized (173). Improving formulations and utilizing enhanced drug delivery systems may further improve efficacy (50). Interestingly, emerging data suggest that even though the H₂S release is dependent on hydrolysis, liberated H₂S from generating molecules may interact with cysteine residues on albumin, which could act as a circulating carrier and slow-release “donor” itself (186, 229).

H₂S inhibition

Therapeutic antagonism of H₂S is a highly debatable strategy and potentially limited to very few indications such as lung cancer (87) since H₂S synthesis from endogenous enzymes is also crucial for cellular sulfur amino acid metabolism and transsulfuration pathways. However, blocking H₂S production can be useful for better understanding its effects and the pathomolecular mechanisms that connect total sulfide metabolism to diseases (232). Thus, the role of H₂S synthesis inhibitors in lung disease is mostly confined to demonstrating that the disruption of H₂S metabolism has negative impacts on disease phenotype such as exacerbation of inflammatory responses, lung injury, and susceptibility to oxidative stress. This is achieved either by inhibiting the activity of H₂S synthases using small molecules (such as PAG, AOAA, and β-cyano-l-alanine [BCA]) or by their genetic ablation (10). It is important to consider that currently available inhibitors lack optimal selectivity, potency, and permeability and exhibit undesired interference with the function of other PLP-dependent enzymes, resulting in off-target effects, extensively reviewed in Ref. (232). With respect to the widely used PAG, only the l-isoform inhibits CTH, whereas the d-isoform is metabolized to a toxic compound in vivo (159). Additionally, PAG and BCA are typically effective in the millimolar range due to their poor permeability and potency (203). Furthermore, PAG, BCA, and AOAA, and others, exhibit a stronger selectivity in inhibiting CTH versus CBS, and there are currently no selective inhibitors of CBS available (10). To circumvent the poor selectivity of small-molecule inhibitors, the optimal approach for selective inhibition is through genetic silencing (159). However, H₂S scavengers or traps could also be used to sequester H₂S and thus decrease its bioavailability without affecting the functions of the synthesizing enzymes (159, 245), although these are currently lacking at present. This is particularly important when considering that these enzymes exert predominantly other biological functions. Genetically deleting CBS in animal models is neonatally lethal, whereas ablation of CTH and MPST may manifest as hypertensive or hyperanxious phenotypes (149, 225, 247). H₂S-sequestering molecules have included hemoglobin, hydroxocobalamin, and zinc (Zn²⁺); however, these also interact with other biomolecules and hence have a variety of other pharmacological effects (159).

Synergistic pharmacological effects in combination therapies in lung disease

The anti-inflammatory effects exhibited in vitro by incubation of CS-exposed monocytic lines, and alveolar macrophages with NaSH and GYY4137 were further potentiated with dexamethasone when given in combination (201), although no controls (e.g., hydrolyzed/spent GYY4137) were presented. This suggests that H₂S may play a role in regulating steroid sensitivity (201). Li et al. investigated the relationship between corticosteroids and H₂S levels and provided evidence that steroids partly exert their anti-inflammatory effects by upregulating the expression of CBS and CTH, thus increasing H₂S bioavailability (130). Theophylline is a phosphodiesterase inhibitor, which restores glucocorticoid responsiveness in refractory COPD smokers by increasing HDAC2 activity (65). Therefore, a clinical trial investigating the impact of combined treatment with theophylline and corticosteroid on H₂S levels in COPD patients suggested that when compared with patients treated with fluticasone only, the addition of theophylline restored steroid responsiveness and improved FEV₁, while also increasing sputum H₂S levels. This indicates that combination treatment of theophylline and steroids may increase H₂S levels to suppress disease (117). Thus, it can be speculated that increasing H₂S bioavailability would ameliorate responsiveness to corticosteroids. In the context of steroid refractory lung inflammation (85, 106), where H₂S could be used as a biomarker to characterize steroid-resistant phenotypes, it is plausible that its supplementation could restore corticosteroid responsiveness in severe asthma and COPD patients.

Mitochondrial H₂S donors in lung disease: too early?

With mitochondrial dysfunction, oxidative stress, immunometabolic alterations, and mtDNA damage being more frequently recognized to contribute to acute and chronic lung conditions (40, 64, 182, 259), it is plausible to contemplate a targeted approach to restore mitochondrial H₂S levels, especially considering that H₂S metabolism and mitochondrial functions are tightly connected (205). Furthermore, the pleiotropic role of H₂S in ameliorating multiple cellular functions including inflammatory responses and oxidative stress suggests that pharmacological interventions in H₂S metabolism represent a promising therapeutic strategy (26, 74, 104, 129, 194, 254). Mitochondria orchestrate inflammatory signaling via inflammasome-mediated responses, whereas H₂S has been shown to dampen inflammasome activity (32, 134). Targeted delivery is advantageous, as H₂S is delivered in a physiologically relevant compartment, but it also requires decreased nominal concentrations to achieve therapeutic effects as molecules accumulate in mitochondria resulting in higher local concentrations, compared with a nontargeted approach and hence reduce the risk of cytotoxic effects of H₂S (76, 207, 213). Considering that many studies use substantial millimolar concentrations to achieve measurable pharmacological effects, a possible solution to overcome potency and related toxicity issues could be targeting mitochondria, which require concentrations order of magnitudes lower than in the nontargeted approach. Targeted delivery to the mitochondria relies on specific physicochemical properties of drugs such as high lipophilicity and positively charged carrier molecules (whether small molecules or peptides), which poses several issues that could potentially antagonize the therapeutic effects of H₂S donors. Indeed, cationic molecules may exert depolarizing effects on mitochondrial membrane potential and hence affect ATP production and contribute to ROS production (86). However, the potent beneficial effects of H₂S achieved with slow release at nanomolar concentrations should be sufficient to outweigh and balance any of the negative side effects induced by the carrier moiety, as demonstrated by the improvement of bioenergetic functions and stabilization of mitochondrial membrane potential when using the mitochondria-targeting H₂S donors AP39 and AP123 (76, 207). A downside of relying on mitochondrial membrane potential to achieve targeted delivery is that in the diseased state, mitochondrial polarity may be altered preventing accumulation of cationic compounds within these organelles, as exemplified in COPD patients (238). Molecular structures that enable targeted delivery are limited to cationic small molecules (such as triphenylphosphonium), Szeto-Schiller peptides, and nanocarriers such as cationic liposomes (146, 148, 208). Although these small molecules showed promising outcomes in preclinical studies, to date, their use in humans have been limited to mitochondrial diseases. Nevertheless, these molecules display favorable safety profiles, and even though their efficacy is limited, several promising clinical studies are still ongoing. Our group specializes in developing targeted molecules and have demonstrated in several diseases that a targeted approach was more effective at ameliorating outcomes, especially with respect to inflammatory and age-related diseases (3, 110, 122, 226, 228), and we are currently evaluating these in acute and chronic lung conditions. Mitochondria-targeted approach of H₂S supplementation offers hope for novel therapeutic approaches in respiratory diseases characterized by inflammation, metabolic dysfunction, and accelerated aging, although these remain to be characterized in lung disease.