Rapid Report pubs.acs.org/crt
Reaction of Dimethyl Trisulfide with Hemoglobin Xinmei Dong, Lóránd Kiss,† Ilona Petrikovics, and David E. Thompson* Department of Chemistry, Sam Houston State University, Huntsville, Texas 77341, United States S Supporting Information *
ABSTRACT: Dimethyl trisulfide (DMTS) is a promising antidotal candidate for cyanide intoxication. DMTS acts as a sulfur donor in the conversion of cyanide to the less-toxic thiocyanate. The alternate reaction pathways of DMTS in the blood are not well understood. We report changes in the hemoglobin absorption spectrum upon reaction with DMTS. These changes closely match those induced by the known methemoglobin former, sodium nitrite. The kinetics of methemoglobin formation with DMTS is slower than with sodium nitrite. These results support the hypothesis that a potentially significant side-reaction of the therapeutically administered DMTS is the oxidization of hemoglobin to methemoglobin.
C
antidote was initiated because of the potential of DMTS to serve as a sulfur donor for the CN detoxification pathway. In vitro, DMTS is an efficient sulfur donor. DMTS converts CN to SCN 40 times more rapidly than thiosulfate in the presence of rhodanese at the experimental pH of 7.4.9 To enhance its solubility for intramuscular injection, DMTS was formulated in micelles and polysorbate 80.12,13 When tested in a mouse model, these formulations of DMTS showed improved efficacy against CN poisoning relative to the Cyanokit and Nithiodote therapies.9 In concert with the initial formulation and efficacy studies, analytical methods were developed for monitoring DMTS in pharmacokinetic studies. These included stir bar sorptive extraction GC−MS,14 an HPLC method for detecting DMTS from blood, and a GC−MS method for detecting DMTS from brain.15 While developing the HPLC method, DMTS recoveries were higher from aged than from fresh blood, and blood samples were observed to darken following DMTS addition. These observations led us to investigate the causes of these phenomena. The characteristic red color of blood is due to oxyhemoglobin (HbO2). HbO2 is formed when the iron(II) ion in Hb reversibly binds oxygen. When HbO2 is oxidized, the resulting iron(III) Hb is called metHb. MetHb does not bind oxygen effectively. Because the blood was observed to darken following DMTS addition, and because the oxidation of Hb to metHb is a cause of darkening in blood, it was hypothesized that DMTS might be oxidizing Hb to metHb. This reaction could explain the lower recoveries of DMTS from fresh blood in which the Hb/metHb ratio was higher than in aged blood. To test the hypothesis that DMTS might be oxidizing Hb, experiments with both blood and isolated Hb were completed. Systematically increasing amounts of DMTS were added to aliquots of heparinized and defibrinated sheep blood (Carolina Biological Supply Company) to replicate the DMTS-induced darkening of blood (Figure 1a). Subsequent blood samples were
yanide (CN) is naturally produced by volcanoes, bacteria, fungi, and plants such as cassava; encountered in the smoke from fires; and anthropogenically produced in large quantities for application in industry.1 CN can be ingested, inhaled, or dermally absorbed. CN is toxic because it inhibits the activity of cytochrome c oxidase, which thus impairs the cell’s oxygen utilization and adenosine triphosphate production. 2 CN intoxication arises unintentionally from smoke inhalation, ingestion of improperly prepared cyanogenic foods, and industrial accidents and intentionally from the use of CN as a suicidal, homicidal, chemical warfare, and genocidal agent.3,4 The major route of endogenous CN detoxification involves converting CN to the less toxic thiocyanate (SCN) with the help of sulfur transferases such as rhodanese and mercaptopyruvate sulfurtransferase.5 A sulfur atom from an endogenous sulfur donor (e.g., thiosulfate) is transferred to the enzyme to form a persulfide intermediate. The enzyme then transfers the sulfur atom irreversibly to CN to yield SCN.5 Antidotes provide sustained antagonism when the endogenous CN detoxification pathway is overwhelmed due to the limited availability of endogenous sulfur donors. The current CN antidotes in the United States are hydroxocobalamin (Cyanokit)6 and the combination therapy of sodium thiosulfate and sodium nitrite (NaNO2) (Nithiodote).7 Hydroxocobalamin successfully competes with cytochrome c oxidase for CN by forming stable cobalt CN complexes.8 Sodium thiosulfate is a sulfur donor that replenishes and supplements the endogenous thiosulfate supply. Nitrite generates methemoglobin (metHb), which binds CN with high affinity to form the relatively stable cyanomethemoglobin.4 The fact that Nithiodote and Cyanokit are formulated for intravenous administration makes them impractical responses in mass-casualty scenarios.9 Novel antidotes for intramuscular delivery are being developed to meet the need for an effective mass-casualty response. One of these is dimethyl trisulfide (DMTS). DMTS contains three consecutive divalent sulfur atoms and is found in foods such as garlic.10 In solution, DMTS undergoes spontaneous interconversion to form a stable equilibrium with dimethyl disulfide and dimethyl tetrasulfide.11 The investigation of DMTS as a CN © 2017 American Chemical Society
Received: July 3, 2017 Published: August 15, 2017 1661
DOI: 10.1021/acs.chemrestox.7b00181 Chem. Res. Toxicol. 2017, 30, 1661−1663
Chemical Research in Toxicology
Rapid Report
(Figure 1c inset). The characteristic Q bands of HbO2 at 540 and 576 nm are present in the negative control spectrum. The weaker Q bands from metHb at 500, 540, 576, and 630 nm are present in the positive control spectrum.17 In the DMTS treated sample, the Q-band pattern is intermediate to those of the negative and positive controls, as would be predicted for a solution containing a mixture of HbO2 and metHb. The second set of experiments addressed the question of whether DMTS could induce the oxidation of Hb in isolation of the other components of blood. Bovine metHb was dissolved in phosphate buffered saline solution (PBS), reduced by sodium dithionite (DT), oxygenated, and spiked with DMTS. The initial concentrations of HbO2 and DMTS were, respectively, 0.012 and 0.79 mM. UV-vis spectra tracked the conversion of HbO2 to metHb by following the Soret peak shift over a period of 113 min. To account for air-borne oxidants, the Soret peak shift was also followed for an unspiked control solution of HbO2 in PBS. Each spectrum (Sn) was modeled as a linear combination of an HbO2 (SoxyHb) and a metHb (SmetHb) spectrum (1):
Figure 1. (a) Sheep blood spiked with increasing concentrations (0, 5, 10, 50, 100, 500, 1000 μg/mL) of DMTS in vitro. (b, c) UV/vis absorption spectra of DMTS-spiked (red, dot) and NaNO2-spiked positive control (blue, dash) and unspiked negative control (black, solid). Inset: Magnified Q-band region. The Hb concentration in panel c was two orders of magnitude lower than that in panel b.
Sn = m1 × SoxyHb + m2 × SmetHb
(1)
Linear regression was used to estimate the constants m1 and m2. The fraction of metHb in each of the DMTS spiked and control samples was calculated as m2/(m1 + m2) and plotted as a function of time to obtain concentration isotherms in the presence and absence of DMTS (Figure 2a). The addition of DMTS accelerates the oxidation of Hb to metHb and drives it to a much higher level of completion than is observed in the control.
obtained by bleeding CD IGS rats and CD-1 mice (Charles River Laboratories, IACUC permission number 15−09−14−1015− 3−01). The blood obtained from CD IGS rats was lysed, diluted, and centrifuged. Three samples were prepared from the supernate: a DMTS spiked sample (7.9 mM DMTS), a negative unspiked control, and a positive control spiked with the known metHb former NaNO2 (7.3 mM NaNO2). The Hb concentration in the sample and controls was 2.3 mM. After a 10 min reaction period, these three samples were diluted 100-fold, and UV−vis absorption spectra (Figure 1b) were collected. The Soret band of the Hb porphyrin ring gave rise to the most prominent peak in the Figure 1b spectra.16 The Soret peak in the negative control was centered at 413 nm, which indicated that the predominant species prior to spiking was, as expected, HbO2.17 The blue shift of the Soret peak induced by DMTS matched that induced by NaNO2 in the positive control, which provided strong support for the hypothesis that DMTS converted Hb to metHb. In the next experiment (Figure 1c), the reaction rates were slowed by diluting the Hb 100-fold prior to spiking with DMTS or NaNO2. Figure 1c shows the UV−vis spectra obtained from three samples of lysed, diluted CD-1 mouse blood: a DMTS spiked sample (8.1 mM DMTS), a negative (unspiked) control, and a positive control spiked with the known metHb former NaNO2 (2.8 mM NaNO2). The Hb concentration in the sample and the two controls was 0.023 mM. The Soret peak at 413 nm in the control spectrum (Figure 1c, black solid) shows that the predominant species prior to spiking is HbO2. The spectra from the DMTS spiked sample (red, dot) and the positive control (blue, dash) both showed a blue-shift in the Soret peak that was consistent with the formation of metHb; however, the DMTS induced blue-shift was smaller than that induced by NaNO2, even though the DMTS concentration exceeded the NaNO2 concentration. This suggested that DMTS oxidizes Hb more slowly than NaNO2. The characteristic Hb Q-band peaks between 500 and 700 nm are responsible for the red to purple color of the porphyrin and are used to distinguish different forms of Hb.16 While higher Hb concentrations are normally used to study the weaker Q bands, they can still be seen in these spectra when the scale is magnified
Figure 2. (a) Reaction of DMTS with Hb in PBS. MetHb fraction in DMTS treated Hb (square), and the untreated Hb control (circle) as a function of time. (b) Reversibility of DMTS induced metHb formation. Absorption spectra of metHb formed by treating an HbO2 with DMTS before (solid) and after (dash) addition of DT.
The conversion of HbO2 to metHb was expected to be reversible. When the reducing agent DT was added to a metHb solution (formed by addition of DMTS), the Soret peak shifted to 430 nm, the metHb Q bands were lost, and a new Q-band appeared at 554 nm. These spectral changes are consistent with the formation of deoxyHb,17 which is the oxygen-free form of reduced Hb. The absence of a peak near 620 nm in the reduced spectrum of Figure 2b suggests that sulfmetHb and sulfHb were not significant products in the reaction with DMTS. In summary, the in vitro addition of DMTS to both diluted blood and isolated Hb solutions was investigated. Consistent effects were seen with sheep, rat, and mouse blood. Addition of DMTS caused a visual darkening of blood from each source. The UV-vis absorption spectra of DMTS-treated samples of diluted 1662
DOI: 10.1021/acs.chemrestox.7b00181 Chem. Res. Toxicol. 2017, 30, 1661−1663
Chemical Research in Toxicology
■
rat blood matched the positive control spectra for metHb obtained by spiking with NaNO2. The Soret shifts and changes in Q-band structure upon the addition of DMTS were consistent with those of metHb. When the Hb concentration was lowered 100-fold in samples of diluted mouse blood to slow the reaction kinetics, DMTS was observed to oxidize Hb to metHb more slowly than NaNO2. In PBS solutions of isolated bovine Hb, DMTS oxidized Hb to metHb in the absence of other components of blood. This metHb was successfully reconverted to Hb by the addition of the reducing agent DT. The primary therapeutic role of DMTS is as a sulfur donor that reacts with CN to form SCN. The present work provides strong initial support for the hypothesis that the reaction of DMTS with Hb to form metHb in blood is an important secondary reaction that contributes both to the half-life of DMTS in blood and also to the antagonism of CN through the formation of small amounts of the CN scavenger, metHb. This reaction may be a direct reaction, or it may be mediated by the interconversion products of DMTS. Since metHb has been found to catalyze the polymerization of H2S to hydropolysulfides, and since these are posited to play a role in post-translational signaling modifications,18 the present report of polysulfide-induced metHb formation may also be relevant to H2S signaling studies. Ongoing investigations in our laboratories are focusing on the in vivo formation of metHb by DMTS.
■
REFERENCES
(1) Thompson, D. E., and Petrikovics, I. (2015) Cyanide physicochemical properties, synthesis, uses, and applications, in Toxicology of Cyanides and Cyanogens: Experimental, Applied, and Clinical Aspects; John Wiley & Sons, Inc., New Jersey. (2) Potter, V. R. (1941) Studies on the mechanism of hydrogen transport in animal tissues: III. Cyanide inhibition of cytochrome C reduction. J. Biol. Chem. 137, 13−20. (3) Hall, A. H., Isom, G. E., and Rockwood, G. A. (2015) Neurological syndrome ingestion of cassava, in Toxicology of Cyanides and Cyanogens: Experimental, Applied, and Clinical Aspects; John Wiley & Sons, Inc., New Jersey. (4) Way, J. L. (1984) Cyanide intoxication and its mechanism of antagonism. Annu. Rev. Pharmacol. Toxicol. 24, 451−481. (5) Westley, J., Adler, H., Westley, L., and Nishida, C. (1983) The sulfurtransferases. Fundam. Appl. Toxicol. 3, 377−382. (6) (2015) in Product Development Under the Animal Rule Guidance for Industry, p. 4, USFDA, CDER, CBER, Maryland. (7) Chen, K. K., and Rose, C. L. (1952) Nitrite and thiosulfate therapy in cyanide poisoning. J. Am. Med. Assoc. 149, 113−119. (8) Borron, S. W., Stonerook, M., and Reid, F. (2006) Efficacy of hydroxocobalamin for the treatment of acute cyanide poisoning in adult beagle dogs. Clin. Toxicol. 44, 5−15. (9) Rockwood, G. A., Thompson, D. E., and Petrikovics, I. (2016) Dimethyl trisulfide: A novel cyanide countermeasure. Toxicol. Ind. Health 32, 2009−2016. (10) Molina-Calle, M., Priego-Capote, F., and Luque de Castro, M. D. (2017) Headspace-GC-MS volatile profile of black garlic vs fresh garlic: Evolution along fermentation and behavior under heating. Food Sci. Technol. 80, 98−105. (11) Steudel, R. (2002) The chemistry of organic polysulfanes R−Sn− R (n > 2). Chem. Rev. 102, 3905−3946. (12) Kovacs, K., Jayanna, P. K., Duke, A., Winner, B., Negrito, M., Angalakurthi, S., Yu, J. C. C., Füredi, P., Ludányi, K., Sipos, P., Rockwood, G. A., and Petrikovics, I. (2016) A lipid base formulation for intramuscular administration of a novel sulfur donor for cyanide antagonism. Curr. Drug Delivery 13, 1351−1357. (13) Kovacs, K., Duke, A. C., Shifflet, M., Winner, B., Lee, S. A., Rockwood, G. A., and Petrikovics, I. (2016) Parenteral dosage form development and testing of dimethyl trisulfide, as an antidote candidate to combat cyanide intoxication. Pharm. Dev. Technol. 0, 1−6. (14) Manandhar, E., Maslamani, N., Petrikovics, I., Rockwood, G. A., and Logue, B. A. (2016) Determination of dimethyl trisulfide in rabbit blood using stir bar sorptive extraction gas chromatography-mass spectrometry. J. Chromatogr. A 1461, 10−17. (15) Kiss, L., Holmes, S., Chou, C., Dong, X., Ross, J., Brown, D., Mendenhall, B., Coronado, V., De Silva, D., Rockwood, G. A., Petrikovics, I., and Thompson, D. E. (2017) Method development for detecting the novel cyanide antidote dimethyl trisulfide from blood and brain, and its interaction with blood. J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 1044, 149−157. (16) Huang, X., Nakanishi, K., and Berova, N. (2000) Porphyrins and metalloporphyrins: versatile circular dichroic reporter groups for structural studies. Chirality 12, 237−255. (17) Ratanasopa, K., Strader, M. B., Alayash, A. I., and Bulow, L. (2015) Dissection of the radical reactions linked to fetal hemoglobin reveals enhanced pseudoperoxidase activity. Front. Physiol. 6, 39. (18) Vitvitsky, V., Yadav, P. K., An, S., Seravalli, J., Cho, U. S., and Banerjee, R. (2017) Structural and mechanistic insights into hemoglobin-catalyzed hydrogen sulfide oxidation and the fate of polysulfide products. J. Biol. Chem. 292, 5584−5592.
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.chemrestox.7b00181. Experimental procedures; additional figures (PDF)
■
Rapid Report
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Phone: + 1 (936) 294 3270. ORCID
David E. Thompson: 0000-0002-2934-5729 Present Address †
Department of Pathophysiology, University of Szeged, Szeged H-6720, Hungary. Funding
This research was supported by the CounterACT Program, NIH Office of the Director, and the NIAID, NIH/Department of Defense Interagency Agreement [AOD16026−001−00000/ A120-B.P2016−01] and by the Robert A. Welch Foundation [X-0011]. Notes
The authors declare no competing financial interest.
■ ■
ACKNOWLEDGMENTS We thank Dr. Donovan C. Haines, and Dr. Afshin Ebrahimpour for helpful conversations and experimental assistance. ABBREVIATIONS DMTS, dimethyl trisulfide; CN, cyanide; SCN, thiocyanate; Hb, hemoglobin; HbO2, oxyhemoglobin; metHb, methemoglobin; deoxyHb, deoxyhemoglobin; NaNO2, sodium nitrite; DT, dithionite; PBS, phosphate buffered saline solution 1663
DOI: 10.1021/acs.chemrestox.7b00181 Chem. Res. Toxicol. 2017, 30, 1661−1663
