Hydrogen Sulfide: Redox Metabolism and Signaling

Abstract

The recognition of hydrogen sulfide (H₂S) as an endogenously produced gas with signaling potential has stimulated research on a multitude of physiological effects mediated in the cardiovascular, immune, gastrointestinal, genitourinary, endocrine, and central nervous systems. The heightened activity in the area of H₂S biology led to convening of the first international conference on H₂S in Shanghai in the summer of 2009 and to two Forum issues published in 2010 by Antioxidants & Redox Signaling on the physiological effects of H₂S. Yet, fundamental questions regarding the biogenesis and regulation of H₂S, the bioenergetics of its catabolism, its tissue concentrations, and elucidation of its molecular targets remain. Some of these issues are the subject of the current Forum on H₂S. Antioxid. Redox Signal. 15, 339–341.

Reactions in the sulfur metabolic pathway involving the amino acids cysteine and homocysteine are potential sources of hydrogen sulfide (H₂S) (4, 28). These include the two enzymes in the transsulfuration pathway, cystathionine β-synthase (CBS), and γ-cystathionase (CSE), which successively convert serine and homocysteine to cystathionine and cystathionine, to cysteine. Lax substrate specificity creates the potential for a multitude of alternative reactions to be catalyzed by both enzymes leading to H₂S generation from either cysteine or homocysteine or both (6, 7, 27). A third potential route for H₂S generation is via the combined actions of cysteine aminotransferase and mercaptopyruvate sulfurtransferase (26), enzymes involved in cysteine catabolism. However, the product of coupling these two enzymes is persulfide rather than H₂S, which can be liberated in the presence of reductant (15).

An efficient mitochondrial pathway for H₂S oxidation exists (10) and is important for holding steady-state tissue levels of H₂S at very low values. This has led to the hypothesis that H₂S might function as an oxygen sensor since, under conditions of hypoxia, the catabolic removal of H₂S would be impeded, leading to augmented H₂S levels and activation of its signaling responses (23). Similarities between the effects of H₂S and hypoxia and enhancement and abrogation of the hypoxic responses by H₂S precursors and H₂S synthesis inhibitors, respectively, support the proposal that H₂S might be involved in oxygen sensing (23).

The discovery of H₂S as a physiological mediator was made by Abe and Kimura, who first reported its neuormodulatory effects (1) and soon thereafter, its vasorelaxant effect on smooth muscle (11). Today, a plethora of physiological effects are associated with H₂S with some of the major links being to the cardiovascular and central nervous systems and to inflammation. H₂S modulates a range of cardiovascular effects, including cardioprotection against ischemic and myocardial reperfusion injury and vascular contraction or relaxation (9, 23, 25, 29). The endothelial-derived relaxing factor activity of H₂S in mice lacking the CSE gene is controversial, with two groups reporting conflicting results. Thus, Wang and coworkers reported marked age-dependent hypertension (30), whereas Ishii and coworkers reported a normotensive phenotype in the CSE knockout mice (14). H₂S stimulates opening of K_ATP channels, suggesting a mechanism for its vasorelaxant effects in addition to its reported involvement in the Nrf2 and ERK and PI3K/Akt signaling pathways (5, 13, 31). In brain, H₂S has varied effects ranging from enhancing hippocampal long-term potentiation (1), to triggering calcium waves and increasing intracellular calcium levels in astrocytes (22), to increased neuroprotection by stimulating glutathione synthesis (19, 20). A possible mechanism for neuromodulation by H₂S involves induction of cyclic AMP synthesis and activating N-methyl D-aspartate-receptor-mediated excitatory postsynaptic currents (18). Both pro- and anti-inflammatory effects have been ascribed to H₂S and efforts are underway to develop the therapeutic potential of H₂S-releasing drugs to counter inflammatory diseases.

Large discrepancies have been reported in the literature on tissue concentrations of H₂S and H₂S production rates in the presence of exogenous substrate. These stem in part from the technical difficulties associated with handling H₂S, which is redox sensitive, and the considerably larger tissue stores of sulfane sulfur and acid-labile sulfur that can be released if adequate precautions are not taken during sample preparation. Additionally, the complexity in utilization of cysteine and/or homocysteine for H₂S generation by CBS and CSE leads to a significant underestimation of H₂S production rates and bias against CBS when only cysteine is applied as substrate, which is commonly the case in the field. Finally, the use of compounds such as hydroxylamine and aminooxyacetate that inhibit many pyridoxal phosphate enzymes and are not specific for CBS or CSE produces results that are not readily interpretable. Two articles in this Forum address these quantitative issues surrounding H₂S determination. In one, Levitt and coworkers report that steady-state tissue levels of H₂S are negligibly low with the exception of aorta, which exhibits 20–100 times higher H₂S concentrations (21). In the other article, the author's group uses quantitative Western blot analyses to determine the absolute protein levels of CBS and CSE in liver and kidney and reports the tissue H₂S production capacity in the presence of saturating concentrations of cysteine and homocysteine used alone or in combination as substrates (16). These studies reveal the differential importance of CBS versus CSE in various tissues. From simulations of the liver data adjusted for the difference in protein levels of CBS and CSE, it is estimated that CSE accounts for ∼96% of H₂S production at physiologically relevant concentrations of substrate.

Bouillaud and coworkers discuss sulfide bioenergetics in the context of the mitochondrial catabolic pathway and the similarities between cyanide and H₂S inhibition of cytochrome c oxidase (3). They also discuss potential mechanisms for neutralizing sulfide particularly as it pertains to the biology of colonocytes, cells that are routinely exposed to high sulfide concentrations, and to the invertebrates living in sulfide-rich habitats. The first enzyme in the H₂S oxidation pathway, sulfide quinone reductase, has a very high affinity for sulfide, and oxidation activity can be detected at intracellular sulfide concentrations ≥10–20 nM, thus protecting cytochrome c oxidase from inhibition. These data also argue against significant steady-state H₂S levels in tissues under normoxic conditions. However, when oxygen concentrations are limiting, H₂S levels are expected to rise leading to inhibition of cytochrome c oxidase concomitant with enhanced mitochondrial reactive oxygen species generation, which is in fact observed under hypoxic conditions.

Tiranti and coworkers discuss the effects of chronic sulfide exposure as seen in the inherited disorder, ethylmalonic encephalophathy, on degradation of cytochrome c oxidase (8). Ethylmalonic encephalopathy results from mutations in Ethe1, the gene encoding the second enzyme in the mitochondrial sulfide oxidation pathway, sulfur dioxygenase. Using tissue-specific conditional Ethe1 knockouts, the authors demonstrate that the cytochrome c oxidase deficiency is limited to the tissues targeted for ablation. Hence, H₂S acts locally in tissue with disrupted Ethe1, leading to heme a inhibition and enhanced degradation of cytochrome c oxidase subunits.

López-Garriga and co-workers (24) discuss hemeproteins such as cyctochrome c oxidase, myoglobin, and hemoglobin as targets for H₂S. Inhibition of cytochrome c oxidase is responsible for decreasing metabolic activity and inducing a hibernation-like state (2). The basis for the differential reactivity of H₂S with hemeproteins that results in formation of stable hexacoordinate low spin Fe^III-SH₂ complex versus iron reduction and conversion to an unligated Fe^II heme and the influence of the iron oxidation state are discussed.

The remaining two articles focus on the neurophysiological effects of H₂S. One, a comprehensive review by Jin-Song and colleagues (12), covers the current state of our understanding of the neurophysiology and neuropathology of H₂S, discussing the underlying cellular mechanisms and the neuroprotective effects of this gaseous mediator. The original research article by Ichinose and coworkers examines the protective effects of H₂S on a murine model of Parkinson's disease induced by the neurotoxin, 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (17). They demonstrate that the neuroprotective effects of H₂S are correlated with induction of antioxidant genes, including glutamate cysteine ligase and heme oxygenase-1.

Abbreviations Used

CBS

cystathionine β-synthase

CSE

cystathionase

H₂S

hydrogen sulfide

Acknowledgment

This work was supported in part by a grant from the National Institutes of Health (HL58984 and DK64959).