Stabilities of Three Key Biological Trisulfides with Implications for Their Roles in the Release of Hydrogen Sulfide and Bioaccumulation of Sulfane Sulfur

Trisulfides and higher polysulfides are important in the body due to their function as key reservoirs of sulfane sulfur and their rapid reactions to release persulfides. Recent work has shown that persulfides act as powerful antioxidants and release hydrogen sulfide, an emerging gasotransmitter with numerous therapeutic effects. Despite the important role of polysulfides, there is a lack of understanding of their stabilities in aqueous systems. To investigate the reactivity of trisulfides and polysulfides, three key biologically important trisulfides were synthesized from cysteine, glutathione, and N-acetylcysteine, and the tetrasulfide of N-acetylcysteine was synthesized as a representative polysulfide. The stabilities of sulfides were monitored in buffered D2O using 1H NMR spectroscopy under a range of conditions including high temperatures and acidic and alkaline environments. The tri- and tetrasulfides degraded rapidly in the presence of primary and tertiary amines to the corresponding disulfide and elemental sulfur. The half-lives of N-acetylcysteine tri- and tetrasulfides in the presence of butylamine were 53 and 1.5 min, respectively. These results were important because they suggest that tri- and tetrasulfide linkages are short-lived species in vivo due to the abundance of amines in the body. Under basic conditions, cysteine and glutathione trisulfides were unstable due to the deprotonation of the ammonium group, exposing an amine; however, N-acetylcysteine trisulfide was stable at all pH values tested. Hydrogen sulfide release of each polysulfide in the presence of cysteine was quantified using a hydrogen sulfide-sensitive electrode and 1H NMR spectroscopy.

. 1 H and 13 C NMR spectra of NAC tetrasulfide in D2O. S-7 Figure S6. NAC trisulfide showed no degradation after 8 days as observed by 1 H NMR spectroscopy in water at pH values of a) 5.8, b) 7.4, and c) 9.0. Figure S7. 31 P NMR spectrum of the reaction between triphenyl phosphine (5.41 ppm) and sulfur generated triphenyl phosphine sulfide (43.35 ppm).
S-8 Figure S8. 1 H NMR spectrum after 4 days shows NAC disulfide converted to tri, tetra, and polysulfides in the presence of butyl amine and elemental sulfur in D2O. Figure S9. High resolution mass spectroscopy spectrum of the reaction between NAC trisulfide, cysteine trisulfide, and butylamine shows the formation of the mixed disulfide of NAC and cysteine. Calculated: 281.0266. Found: 281.0273.

S-9
Figure S10. Graphs of the degradation of NAC trisulfide in the presence of butylamine (left) and triethylamine (right) fit to pseudo first order reaction kinetics. Due to the equilibrium, only early data points were used to provide an estimated rate constant. The estimated rate constants for the degradation of NAC trisulfide in the presence of butylamine and triethylamine were 0.74 h -1 and 0.31 h -1 , respectively. Error bars are ± instrument error. Figure S11. a) The general structure of cysteine and cysteine polysulfides are shown. b) Relevant regions of 1 H NMR spectra of cysteine (bottom), cystine (middle), and cysteine di-triand tetrasulfide mixture mixture (top) with the tetrasulfide labelled as "*". Figure S12. The stabilities of cysteine trisulfide at pH values of 5.8 (red), 7.0 (blue), and 9.0 (orange) were measured by 1 H NMR spectroscopy (top). Graphs of the degradation of cysteine trisulfide at pH 7.0 (bottom left) and pH 9.0 (bottom right). The rate constants for the degradation of cysteine trisulfide at pH 7.0 and 9.0 are 0.041 d -1 and 0.061 d -1 , respectively. Error bars are ± instrument error. Figure S13. Graphs of the degradation of cysteine trisulfide in the presence of butylamine (left) and triethylamine (right) fit to pseudo first order reaction kinetics. Due to the equilibrium, only early data points were used to yield an estimated rate constant. The estimated rate constants for the degradation of cysteine trisulfide in the presence of butylamine and triethylamine are 0.30 h -1 and 0.24 h -1 , respectively. Error bars are ± instrument error.

S-11
Figure S14. a) Stability of glutathione trisulfide at pH values of 5.8 (red), pH 7.4 (green), and pH 9.0 (blue) were measured by 1 H NMR spectroscopy. b) Graphs of the degradation of glutathione trisulfide at pH 5.8 (top left), pH 7.4 (top right), and pH 9.0 (bottom left). Due to the equilibrium, only early data points were used. The estimated rate constants for the degradation of glutathione trisulfide at pH 5.8, 7.0, and 9.0 are 0.11 d -1 , 0.77 d -1 , and 0.88 d -1 , respectively. Error bars are ± instrument error. Figure S15. Graphs of the degradation of glutathione trisulfide in the presence of butylamine (left) and triethylamine (right) fit to pseudo first order reaction kinetics. Due to the equilibrium, only early data points were used. The estimated rate constants for the degradation of glutathione trisulfide in the presence of butylamine and triethylamine are 0.53 h -1 and 0.96 h -1 , respectively. Error bars are ± instrument error. Figure S16. a) The stability of NAC tetrasulfide at pH values of 5.8 (blue), pH 7.4 (red), and pH 9.0 (orange) were measured by 1 H NMR spectroscopy. b) Graphs of the degradation of NAC tetrasulfide at pH 5.8 (top left), pH 7.4 (top right), and pH 9.0 (bottom left). The estimated rate constants for the degradation of NAC tetrasulfide at pH 5.8, 7.0, and 9.0 are 0.013 d -1 , 0.093 d -1 , and 0.085 d -1 , respectively. Error bars are ± instrument error.
S-14 Figure S17. Graphs of the degradation of NAC tetrasulfide in the presence of butylamine (left) and triethylamine (right) fit to pseudo first order reaction kinetics. Due to the equilibrium, only early data points were used to yield an estimated rate constant. The estimated rate constants for the degradation of NAC tetrasulfide in the presence of butylamine and triethylamine are 0.46 min -1 and 0.44 min -1 , respectively. Error bars are ± instrument error.    Figure S22. Graph of the degradation of NAC trisulfide under biological conditions in the presence of butylamine fit to pseudo first order reaction kinetics. Due to the equilibrium, only early data points were used. The estimated rate constant was measured to be 0.90 h -1 . Error bars are ± instrument error.
S-17 Figure S23. a) Concentration of glutathione trisulfide in the presence of glutathione, and b) concentration of cysteine trisulfide in the presence of cysteine measured by 1 H NMR spectroscopy. Each trisulfide degraded very quickly, reaching equilibrium in >15 minutes.