Enhanced generation of hydroxyl radical and sulfur trioxide anion radical from oxidation of sodium sulfite, nickel(II) sulfite, and nickel subsulfide in the presence of nickel(II) complexes.

Electron spin resonance (ESR) spin trapping was utilized to investigate the generation of free radicals from oxidation of sodium sulfite, nickel(II) sulfite, and nickel subsulfide (Ni3S2) by ambient oxygen or H2O2 at pH 7.4. The spin trap used was 5,5-dimethyl-1-pyrroline-N-oxide (DMPO). Under ambient oxygen, a solution of sodium sulfite alone generated predominantly sulfur trioxide anion radical (.SO3-) due to the autoxidation of sulfite. Addition of nickel(II) chloride [Ni(II)] enhanced the .SO3- yield about 4-fold. Incubation of sulfite with Ni(II) in the presence of chelators such as tetraglycine, histidine, beta-alanyl-3-methyl-L-histidine (anserine), beta--L-histidine (carnosine), gamma-aminobutyryl-L-histidine (homocarnosine), glutathione, and penicillamine did not have any significant effect on that enhancement. In contrast, albumin, and especially glycylglycylhistidine (GlyGlyHis), augmented the enhancing effect of Ni(II) by factors of 1.4 and 4, respectively. Computer simulation analysis of the spin-adduct spectrum and formate scavenging experiment showed that the mixture of sodium sulfite, Ni(II), and GlyGlyHis generated both hydroxyl (.OH) radical and .SO3- radical, in the ratio of approximately 1:2. The free-radical spin adduct intensity reached its saturation level in about 5 min. The yield of the radical adducts could be slightly reduced by deferoxamine and very strongly reduced by diethylenetriaminepentaacetic acid (DTPA). Aqueous suspensions of sparingly soluble nickel(II) sulfite in the presence of air and GlyGlyHis generated surface-located .SO3- and .OH radicals. The same radicals were generated in Ni3S2 suspension in the presence of GlyGlyHis and H2O2, indicating sulfite production by oxidation of the sulfide moiety of this compound.(ABSTRACT TRUNCATED AT 250 WORDS)


Introduction
It has been shown that sulfur dioxide (SO2) is a major air pollutant (1). In aqueous media it exists in equilibrium with the sulfite ion (So<32) (2). Sulfite appears to have genotoxic effects (3). It can act as a mutagen or co-mutagen (4) and a co-carcinogen (5). The mechanism of sulfite toxicity is believed to be related to sulfite oxidation processes involving sulfur trioxide radical ion (-SO;) formation (1,(6)(7)(8).
Recent studies have shown that hydroxyl This paper was presented at the Second International Meeting on Molecular Mechanisms of Metal Toxicity and Carcinogenicity held 10-17 January 1993 in Madonna di Campiglio, Italy.
The authors are grateful to Drs. L. K. Keefer, J. M. Rice, and R. W. Nims for helpful comments and suggestions on the manuscript, and to Kathy Breeze for editorial assistance. The West Virginia University's contribution was supported under grant No. 1  (-OH) radical is also generated in the sulfite oxidation pathway (9). This radical may play an important role in sulfiteinduced genetic damage. For example, incubation of sulfite with DNA produces 8-hydroxy-2'-deoxyguanosine (8-OH-dG) (10), a marker of oxidative DNA damage (11). It has been reported that Mn(II) (12) and chromate (9) enhance sulfite oxidation either as catalysts or as oxidants. However, it has not been clear whether sulfite oxidation can also be enhanced by other transition metal ions, Ni(II) in particular. Ni(II)-enhanced sulfite oxidation would be highly interesting, since nickel in many physicochemical forms is carcinogenic to humans and animals (13)(14)(15)(16). Among nickel compounds, the crystalline sulfides, especially the subsulfide Ni3S2, appear to be the strongest carcinogens (17). Ni3S2 is potentially genotoxic, causing random polymerization of histones in vitro (18), formation of DNA-protein crosslinking bonds (18), and hydroxylation of 2'deoxyguanosine (dG) to 8-OH-dG in the presence of ambient oxygen (19). The latter reaction can be enhanced by H202 (19). Most importantly, Ni3S2 interacts with molecular oxygen and undergoes slow oxidation of both its cationic and anionic constituents to soluble Ni(II) and S042 (20,21). Oxidation of the sulfur moiety of Ni3S2 is gradual and goes through sulfite as an intermediate (21). Thus, it appears that Ni3S2 yields two products that are capable of damaging DNA.
Unlike other transition metal ions such as Fe(II), Cu(II) and VO2, Ni(II) does not cause efficient free radical generation from 02, H202 or lipid hydroperoxides.
However, recent studies (22,23) have shown that reactivity of Ni(II) with those oxygen derivatives can be modulated by chelation, e.g., with certain histidine-and cysteine-containing ligands. We could expect, therefore, that chelation of Ni(II) might also facilitate the stimulatory effect of this cation on sulfite oxidation. Hence, in the present study we investigated freeradical generation during autoxidation of Environmental Health Perspectives 91 sulfite with Ni(II) added in either nonchelated or chelated form. The results show that in the presence of certain chelators, Ni(II) significantly increased -SO radical generation and, in addition, caused -OH radical generation from sodium sulfite reacting with ambient oxygen. Aqueous suspensions of sparingly soluble nickel(II) sulfite, under ambient atmosphere, generated surface-bound free radicals only in the presence of a chelator. For Ni3S2 suspensions, both a chelator and a stronger oxidant, H202, were required for the free radical generation. Cooperation of both constituents of Ni3S2 in free-radical production may explain the exceptionally high carcinogenic potential of this compound.

ESR Measurements
ESR spin-trapping methodology (24,25) was employed for detecting short-lived free radical intermediates. All ESR measurements were performed with a Varian E3 ESR spectrometer and a flat-cell assembly. Hyperfine splittings were measured (to 0.1 G) directly from the magnetic field separation using tetraperoxochromate (K3CrO8) and 1,1 -diphenyl-2-picrylhydrazyl (DPPH) as reference standards. A Bruker ASPECT 2000 computer was used for spectral analyses. Reactants  test tubes in a total final volume of 250 pl. The reaction mixture was then transferred to a flat cell for ESR measurement. The concentrations given in the figure legends represent final concentrations. All experiments were carried out at room temperature in the presence of air unless otherwise indicated.

Effect of GlyGlyHis and Other
Chelators on Ni(ll)-mediated Free-Radical Generation in Sodium Sulfite Solution  Figure 1 c seemed to be a superposition of those from the DMPO adducts of -SO3 and -OH radicals. To test this possibility, we used the well-known reaction of K2Cr2O7 with Na2SO3 to generate the -SO radicals (9,28). As shown in Figure 3a, this reaction generated the typical DMPO/-SO adduct, with hyperfine splittings of aN = 14.7 G and aH =  We also used the Fenton reagent to generate authentic DMPO/-OH adduct, with hyperfine splittings of aN = aH = 14.9 G ( Figure 3b). An ASPECT 2000 computer was used to combine various percentages of the DMPO/-SO3-and DMPO/-OH adduct spectra. The profile in Figure 3c, which corresponded most closely with the spectrum in Figure Ic, was a combination of 65% DMPO/.SO3 and 35% DMPO/-OH. To ascertain whether -OH radicals were indeed produced in the sulfite oxidation in the presence of Ni(II) and GlyGlyHis, formate was used as an -OH radical scavenger. It was added to a mixture containing 2.5 mM sodium sulfite, 1 mM Ni(II) and 2 mM GlyGlyHis at concentrations of up to 0.8 M. The resulting spectrum is shown in Figure 4a. Formate caused the appearance of DMPO/-COO adduct, suggesting that -OH radical was indeed being produced in the mixture of sodium sulfite, Ni(II), and GlyGlyHis.
An experiment carried out under argon showed a significant decrease in the overall spectral intensity (Figure 4b). Thus, the free radical generation from the mixture of sodium sulfite, Ni(II), and GlyGlyHis required molecular oxygen. The residual free radical generation observed under argon may be due to remnants of oxygen dissolved in the reaction mixture. Figure 5a shows the ESR spectrum obtained from a mixture containing 1 mM Ni(II), 2.5 mM sulfite, 2 mM GlyGlyHis, and 2 mM deferoxamine. The latter caused only a slight decrease in the overall spectral intensity (compare Figures 5a and 1c). Moreover, upon addition of deferoxamine, the second peak of the spectrum exhibited a doublet, similar to that in Figure 3a. This indicates that the major spin adduct obtained in this case was DMPO/.SO0.
Hence, deferoxamine inhibited -OH generation to a greater degree than -SO; generation. Addition of DTPA sharply reduced the spectral intensity (Figure 5b). In the absence of GlyGlyHis, deferoxamine and DTPA decreased the enhancing effect of Ni(II) on free radical generation from autoxidation of sulfite alone (compare Figures 5c and 5d with Figure Ib).
Several other Ni(II) chelators, such as GlyGlyGlyGly, histidine, anserine, carnosine, homocarnosine, GSH, penicillamine, and albumin at 2 mM concentrations were also tested for their effect on 1 mM Ni(II)-  Figure 7b were presen tion, as indicated by the relative spectral peaks (compared with F Like that in Figure 6b, the spe Figure 7b can any significant amounts of free radicals (Figure 7d, e). Figures 6b and 7b show that the intensities of the two spectra, representing N1503 NiSO3 and Ni3S2, respectively, are compa-1 minutes rable but the linewidth of the former is very broad. Since the integration of the spectrum represents the amount of free radicals generated, the relative yield of free radical production by Ni3S2 was much sphate buffer lower than that by NiSO3. ml NiSO 3 and Discussion uin after reacafter reaction It has been known that Ni(II) does not eas-yHis. The ESR ily react with oxidants to produce free radiose described cal species (31,32). However, its reactivity toward molecular oxygen, H202, and lipid hydroperoxides can be greatly enhanced by cated and coordination with certain chelating agents, )3 suspen-including GlyGlyHis (22,23,30,32,33). )nsisted of Likewise, the results obtained in the pregned to a sent study show that while Ni(II) can 03-and increase -SOs radical generation from sulprevious fite to some extent, Ni(II) chelated with ess of the the oligopeptide GlyGlyHis is far more rmine the effective. Most importantly, in addition to /OH.
-SOj radical, chelated Ni(II) markedly enhances generation of -OH radicals in this adlcal system. Unlike GlyGlyHis, other histidyl chelators, including anserine, carnosine, homocarnosine, and histidine itself, as well e spectra as some thiol-containing chelators such as 50 mg/ml GSH and penicillamine, did not affect the 7.4) in the Ni(II)-enhanced free radical generation nd 5 mM from sodium sulfite. It is noteworthy that ed at time all these chelators have been shown to n, respec-enhance Ni(II)-mediated free radical gener-L1 adducts ation from model lipid hydroperoxides, it in solucumene hydroperoxide and t-butyl ly narrow hydroperoxide (22,23). Similarly, ligure 6b). GlyGlyGlyGly, which appeared to lack any ectrum in effect on Ni(II) action in the present study, mbination has been reported to enhance Ni(II)-pro-OH. The moted free radical generation from H2O2 AlyGlyHis (30). It appears, therefore, that the enhancrelatively ing effect of various Ni(II) chelates on free gure 7c). radical production in the case of oxidation it this sig-of different substrates depends on the uperoxide chemistry of those substrates. fine split- The present study also shows that besides assisting Ni(II) in the enhancement of free radical production by autoxidation of the sulfite anion in solution, GlyGlyHis could as well increase radical production from sparingly soluble NiSO3 particles in aqueous suspension. In contrast, Ni3 S2 suspension, alone or in the presence of GlyGlyHis, did not generate a detectable amount of free radicals under ambient oxygen. Free radicals could be detected in the Ni3S2/GlyGlyHis system only after the addition of H202. Unlike in the suspension of NiSO3 in which 'SO3 and 'OH radicals were associated mainly with the particle surface (broad ESR peaks), the same radicals generated in the Ni3S2 system were detectable in the solution (narrow peaks). The reason for this difference is not clear. It may be due to the different physicochemial properties of the two sys-Environmental Health Perspectives v Ni S y yHi s tems, e.g., different distribution of Ni(II) and sulfite between the aqueous and the solid phases and the presence of H202. Apparently, H202 functioned as a Ni3S2 oxidant to produce Ni(II) and sulfite, a mixture facilitating generation of SO-and -OH, especially in the presence of GlyGlyHis, as discussed above. It also seems possible that Ni(II), derived from Ni3S2 and chelated with GlyGlyHis, reacted with H202 to yield -OH and 02-radicals, as described elsewhere (30). Both, -OH and *02-could, in turn, react with sulfite to produce SO3-radicals (34,35). However, no DMPO/-02-adduct was detected, which argues against but does not exclude the second scenario.
It is now generally believed that SO3radical can cause many adverse effects while reacting with biological molecules. These include oxidation of methionine (36), diphosphopyridine nucleotides (37), and lipids (38); destruction of n-carotene (8) and tryptophan (36); addition to double bonds in unsaturated fatty acids (39); and modification of nucleic acids (7,40). Some of these effects are potentially genotoxic and may contribute to carcinogenesis (3)(4)(5). The genotoxicity of -OH radical, also generated by sulfite oxidation, is well recognized (41). For example, the -OH radical generated by sodium sulfite autoxidation was reported to be capable of producing promutagenic 8-OH-dG in DNA (10). The observed enhancement of -OH radical production from sulfite by Ni(II) chelated with GlyGlyHis and albumin may be involved in the mechanism(s) of carcinogenesis by nickel(II) sulfides, the strongest nickel carcinogens. Both Ni(II) and sulfite are derived from oxidative dissolution of Ni3S2 and crystalline NiS (20,21). Albumin and other proteins having metal binding sites of the GlyGlyHis type are major tissue Ni(II) carriers (42). All these reagents, together with H202 may thus act in concert to generate genotoxic radicals from nickel sulfides. H202 may originate from cellular metabolic or pathogenic processes; e.g., sulfides of nickel and cadmium induce H202 formation by polymorphonuclear leukocytes (43). In concordance with this assumption, in the presence of molecular oxygen, Ni3S2 was found to cause hydroxylation of dG to 8-OH-dG in vitro. The yield of 8-OH-dG was greatly enhanced by H202 (19). Ni3S2 and NiSO3 were also found to produce another potentially mutagenic effect, i.e., deamination of 5-methyl-2'-deoxycytidine to thymidine (44). NiSO3 was much more active in this reaction than the well-known deaminating agent, sodium bisulfite. Thus, the exceptionally high carcinogenic potential of Ni3S2 compared to other nickel derivatives (17) may be due to the ability of both the metal and sulfide constituents of the molecule to enhance generation of genotoxic free radicals.