Reactive cysteine persulfides and S-polythiolation regulate oxidative stress and redox signaling

Abstract

Using methodology developed herein, it is found that reactive persulfides and polysulfides are formed endogenously from both small molecule species and proteins in high amounts in mammalian cells and tissues. These reactive sulfur species were biosynthesized by two major sulfurtransferases: cystathionine β-synthase and cystathionine γ-lyase. Quantitation of these species indicates that high concentrations of glutathione persulfide (perhydropersulfide >100 μM) and other cysteine persulfide and polysulfide derivatives in peptides/proteins were endogenously produced and maintained in the plasma, cells, and tissues of mammals (rodent and human). It is expected that persulfides are especially nucleophilic and reducing. This view was found to be the case, because they quickly react with H2O2 and a recently described biologically generated electrophile 8-nitroguanosine 3',5'-cyclic monophosphate. These results indicate that persulfides are potentially important signaling/effector species, and because H2S can be generated from persulfide degradation, much of the reported biological activity associated with H2S may actually be that of persulfides. That is, H2S may act primarily as a marker for the biologically active of persulfide species.

Keywords: electrophilic signaling; hydrogen sulfide; polysulfidomics; thiol redox.

Fig. 1.

Enzymatic generation of CysSSH and polysulfides catalyzed by CSE and CBS. Yields of Cys polysulfides by conversion of CysSSCys by (A) CSE and (B) CBS. (A) CysSSCys (1.25 mM) was incubated with 50 μg/mL CSE in 30 mM Hepes buffer (pH 7.5) containing 50 μM pyridoxial phosphate at 37 °C. (B) CysSSCys (0.5 mM) was incubated with 5 μg/mL CBS in 30 mM Hepes buffer (pH 7.5) containing 50 μM pyridoxial phosphate and 0.1 mM S-adenosyl methionine at 37 °C. H₂S-like species represent not only H₂S/HS⁻ but also other unidentified products that may directly or indirectly generate bis-S-bimane through reaction with Br-bimane and originate from Cys persulfide derivatives. Details are in SI Appendix.

Fig. 2.

Cellular formation of reactive sulfur species, mainly Cys and GSH polysulfides, identified by cellular fluorescence imaging and polysulfur metabolomics. (A) SSP2-induced fluorescence imaging of A549 cultured cells. (Scale bars: 50 μm.) (B) Intracellular levels of Cys and GSH polysulfides in A549 cells overexpressing CBS or CSE as assessed by LC-MS/MS. Data represent means ± SDs (n = 3). *P < 0.05 versus control; **P < 0.01 versus control.

Fig. 3.

Mechanism of intracellular maintenance of various Cys and GSH polysulfides. (A) Schematic representation of the Cys polysulfur cycle involving CBS, CSE, and GSR for intracellular maintenance of various Cys and GSH polysulfides. (B) Regeneration of GSSH and GSSSH from GSSSG catalyzed by GSR. GSSSG (0.1 mM) was reacted with GSR (1 U/mL) and NADPH (0.2 mM) in 20 mM Tris⋅HCl buffer (pH 7.4) at room temperature for 5 min followed by incubation with 5 mM Br-bimane at 37 °C for 15 min. Shown are the (Left) LC-MS elution profile and (Right) quantitative data for the reaction. Data represent means ± SDs (n = 3).

Fig. 4.

Polysulfur metabolomics in vivo. Effects of dietary sulfur sources on levels of various Cys-related polysulfides in mouse heart as determined by LC-MS/MS. Mice were maintained with the diets containing different amounts of cystine and methionine for 4 wk. Data are means ± SDs (n = 3). *P < 0.05 versus control diet [+, cystine; methionine at 100% (wt/wt)].

Fig. 5.

Unique proteomic analysis based on the Tag-Switch-Tag strategy for protein-bound reactive sulfur labeling. (A) Protein-Cys polysulfurs identified as (Upper) biotin-labeled bands in Western blotting and (Lower) quantitative data by densitometric analysis. Data are means ± SDs (n = 3). *P < 0.05 versus control. (B) Schematic diagram for identification of protein-bound Cys-CN-biotin. (C) LC-MS/MS identification of Cys-CN-biotin (m/z 474.1). Lysates of A549 cells overexpressing CSE reacted with CN-biotin [CN-biotin (+)] or without CN-biotin [CN-biotin (−)] were analyzed as (Left) whole-cell lysates (Lysate) or (Right) after gel electrophoresis (Gel). Representative results analyzing the gel at the position indicated by the white arrowhead in A are shown. (D) Protein-Cys polysulfurs identified as spots on membranes transferred from 2D gels. Some labeled spots in the 2D image were subjected to polysulfur proteomics, which identified several polythiolated proteins (SI Appendix, Table S5).

Fig. 6.

The antioxidant effect of GSSH. Potent antioxidant ability of GSSH as determined by scavenging of H₂O₂. (A) Direct scavenging of H₂O₂ by GSSH in vitro. H₂O₂ (0.1 mM) was reacted with GSSG or GSSSG (0.2 mM each) in the absence or presence of GSR (1 U/mL) and NADPH (0.4 mM) at 37 °C for 30 min. (B) Effects of CSE overexpression on H₂O₂-mediated cell death. A549 cells with or without CSE overexpression were treated with H₂O₂ (1,000 or 2,000 μM) for 24 h. Cell viability was determined and expressed as a percentage of control without H₂O₂ treatment. Data represent means ± SDs (n = 4). *P < 0.05 versus control.

Fig. 7.

The interaction of GSSH with 8-nitro-cGMP. (A) 8-SH-cGMP formation by chemical polythiolation, with formation of polysulfide cGMP as an intermediate molecular species (SI Appendix, Fig. S10), which led to effective 8-SH-cGMP formation from 8-nitro-cGMP. 8-Nitro-cGMP (1 mM) was reacted with NaHS (0.1 mM) in the absence or presence of GSH (0.1 mM), P-NONOate (0.1 mM), or both in 10 mM Tris⋅HCl buffer (pH 7.4) at 37 °C. Addition of 10 mM DTT enhanced 8-SH-cGMP formation. (B) In vivo endogenous formation of 8-SH-cGMP in organs in normal mice. Both 8-SH-cGMP and cGMP were quantified by using LC-MS/MS of the methanol extract of each organ. (C) Endogenous 8-SH-cGMP formation dependent on CBS expression in C6 cells in culture as identified by LC-MS/MS. C6 cells were treated with LPS, IFN-γ, TNF-α, and IL-1β to induce endogenous NO generation, leading to the formation of 8-nitro-cGMP. (D) Effects of CBS knockdown on production of (Left) 8-SH-cGMP and (Right) CysSSH in C6 cells treated with LPS plus cytokines as in C for 36 h assessed by LC-MS/MS. Error bars indicate means ± SDs (n = 3).