(Bi)sulfite Oxidation by Copper,Zinc-Superoxide Dismutase: Sulfite-Derived, Radical-Initiated Protein Radical Formation

Background Sulfur dioxide, formed during the combustion of fossil fuels, is a major air pollutant near large cities. Its two ionized forms in aqueous solution, sulfite and (bi)sulfite, are widely used as preservatives and antioxidants to prevent food and beverage spoilage. (Bi)sulfite can be oxidized by peroxidases to form the very reactive sulfur trioxide anion radical (•SO3−). This free radical further reacts with oxygen to form the peroxymonosulfate anion radical (−O3SOO•) and sulfate anion radical (SO4• −). Objective To explore the critical role of these radical intermediates in further oxidizing biomolecules, we examined the ability of copper,zinc-superoxide dismutase (Cu,Zn-SOD) to initiate this radical chain reaction, using human serum albumin (HSA) as a model target. Methods We used electron paramagnetic resonance, optical spectroscopy, oxygen uptake, and immuno-spin trapping to study the protein oxidations driven by sulfite-derived radicals. Results We found that when Cu,Zn-SOD reacted with (bi)sulfite, •SO3− was produced, with the concomitant reduction of SOD-Cu(II) to SOD-Cu(I). Further, we demonstrated that sulfite oxidation mediated by Cu,Zn-SOD induced the formation of radical-derived 5,5-dimethyl-1-pyrroline N-oxide (DMPO) spin-trapped HSA radicals. Conclusions The present study suggests that protein oxidative damage resulting from (bi)sulfite oxidation promoted by Cu,Zn-SOD could be involved in oxidative damage and tissue injury in (bi)sulfite-exacerbated allergic reactions.

volume 118 | number 7 | July 2010 • Environmental Health Perspectives Research Sulfur dioxide is one of the major atmospheric pollutants, but its two ionized forms in aque ous solution at neutral pH, sulfite (SO 3 2-) and (bi)sulfite (HSO 3 -), are widely used as anti oxidants and preservatives in beverages and foods (Danilewicz 2003). However, the preva lence of sulfite toxicity is relatively high, and it has been associated with allergic reactions characterized by sulfite sensitivity, asthma, and anaphylactic shock (Komarnisky et al. 2003). Sensitive individuals can experience such adverse reactions when they consume sulfites, with asthmatics being particularly vulnerable to such toxicity.
Sulfite is detoxified in the liver and lung to sulfate by sulfite oxidase, a molybdenum dependent mitochondrial enzyme (Cohen and Fridovich 1971); sulfite oxidase deficiency is one of the most accepted causes of sulfite hyper sensitivity and toxicity. This enzymatically catalyzed oxidation has been shown to proceed via a twoelectron oxidation without the forma tion of any detectable radical intermediates. In contrast, recent studies suggest that the cytotox icity of (bi)sulfite is mediated by free radicals, because (bi)sulfite increases reactive oxygen spe cies formation, and antioxidants and free radical scavengers prevent its toxicity (Niknahad and O'Brien 2008). In addition, transition metals catalyze the auto xidation of (bi)sulfite via sulfur trioxide anion radical ( • SO 3 -) formation: where M may be copper (Cu 2+ ), iron (Fe 3+ ), oxivanadium anion (VO 2+ ), manganese (Mn 2+ ), nickel (Ni 2+ ), or chromate anion (CrO 4 2-) (Alipazaga et al. 2004;Berglund et al. 1993;Brandt and Elding 1998;Lima et al. 2002;Shi 1994), but this reaction requires higher concentrations of (bi)sulfite to permit effective propagation of the chain reac tion. In a recent study, Alipazaga et al. ( 2009) reported oxidative DNA damage induced by (bi)sulfite solutions in the presence of Cu(II) peptide complexes. It has also been shown that free radicals have been produced by enzy matic initiation of the oxidation of (bi)sulfite by prostaglandin H synthase (Mottley et al. 1982a) and horseradish peroxidase (HRP) (Araiso et al. 1976;Mottley et al. 1982b), with formation of • SO 3 -. This predominantly sulfurcentered radical (Chantry et al. 1962) reacts with molecular oxygen by forming the peroxy mono sulfate anion radical ( -O 3 SOO • ), which is a precursor of the sulfate anion radi cal (SO 4 •-) (Neta et al. 1988): SO 4 •is a very strong oxidant, nearly as strong as the hydroxyl radical, and it is very likely to oxidize other biomolecules by one electron oxidation.
It is possible that bisulfite may also lead to further reactive sulfur species via the peroxi dase activity of enzymes such as copper,zinc superoxide dismutase (Cu,ZnSOD), a metalloenzyme that catalyzes the dismutation of the superoxide anion into O 2 and hydrogen peroxide (H 2 O 2 ). At high pH, Cu,ZnSOD exhibits peroxidase activity, with the initial step of the peroxidase cycle being a reduction of SODCu(II) by H 2 O 2 or its deprotonated form, HO 2 -, to SODCu(I) (Bonini et al. 2009;Fuchs and Borders 1983;Hodgson and Fridovich 1973). At neutral pH, the peroxi dase activity of Cu,ZnSOD is stimulated in the presence of bicarbonate (HCO 3 -) buffer (Bonini et al. 2004;Liochev and Fridovich 2004;Zhang et al. 2000). It has been pro posed that at pH 7.4, anions structurally simi lar to HCO 3 -, such as (bi)sulfite (HSO 3 -) and (bi)selenite (HSeO 3 -), may also stimulate the peroxidase activity of Cu,ZnSOD in the presence of millimolar H 2 O 2 (Sankarapandi and Zweier 1999).
In the present study, we evaluated the role of Cu,ZnSOD in (bi)sulfite oxidation and found that, under our experimental con ditions, SOD1Cu(II) is slowly reduced to SOD1Cu(I) by (bi)sulfite. We used optical spectroscopy, electron spin resonance (ESR), and oxygen uptake experiments to demon strate that (bi)sulfite (as Na 2 SO 3 ) was a one electron donor substrate for Cu,ZnSOD, leading to the generation of reactive sulfur radicals via Equations 2-4. We also applied immunospin trapping with 5,5dimethyl1 pyrroline Noxide (DMPO) to investigate oxi dation of target proteins [e.g., human serum albumin (HSA) at plasma levels] to protein radicals ( Figure 1). We found that (bi)sulfite oxidation mediated by Cu,ZnSOD generated the formation of HSA radicals, which might be responsible for the tissue injury in allergic reactions to (bi)sulfite.
ESR spectroscopy. We obtained ESR spin trapping data at room temperature using a Bruker EMX spectrometer with 100 kHz modulation frequency and equipped with an ER 4122 SHQ cavity (Bruker BioSpin Corp., Billerica, MA). We placed samples in a 10mm flat cell (200 µL final volume) and initiated recording of the spectra within 1 min of the start of the reaction.
We recorded lowtemperature ESR data at 77 K after the indicated incubation times. Initially, we mixed SOD with (bi)sulfite at room temperature; after incubation, we trans ferred the reaction mixtures into 1mL poly ethylene syringes and froze them in liquid nitrogen. We added glycerol (10%) to the samples before freezing to prevent cracking of the frozen texture.
Oxygen uptake. For oxygen uptake meas ure ments, we added 500 µL sodium sulfite to a chamber equipped with a Clark electrode and a stirrer. We initiated the reaction (1.8 mL) by SOD, and the oxygen uptake curves were obtained at room temperature with an oxygen monitor (model 53; Yellow Springs Instrument Co., Yellow Springs, OH).
Chemical reactions. Typically, we car ried out reactions of 600 µM HSA, 500 µM Na 2 SO 3 , and 50 µM Cu,ZnSOD in the pres ence or absence of 5 mM DMPO in 100 mM phosphate buffer (Chelextreated with 25 µM DTPA) at pH 7.4 in a total volume of 30 µL. After 1 hr of incubation at 37°C, we stopped reactions with 5 mM reduced glutathione and then diluted the samples with deionized H 2 O for electrophoresis and immunospin trapping analyses.

Coomassie blue stain, Western blot, and ELISA (enzyme-linked immunosorbent assay).
We electrophoresed the reaction mixtures under reducing conditions through dupli cate 4-12% BisTris NuPage acylamide gels (Invitrogen, Carlsbad, CA). We performed Western blotting and ELISA analysis as previ ously described (Detweiler et al. 2002) with minor changes (we used fish gelatin instead of casein to prevent the nonspecific binding sites).
Optical spectroscopy. We recorded opti cal data on a Cary 100 spectrophotometer (Varian Inc., Palo Alto, CA) using a 500 µM quartz cuvette. We determined Cu,ZnSOD concentration from the broad band at 680 nm (ε = 300 M -1 cm -1 in the bovine enzyme), which results from the dd transitions of the Cu atom (Foti et al. 1997). We carried out reactions in 100 mM phosphate buffer at pH 7.4. Cu,ZnSOD (1 mM) was added first, followed by 20 mM (bi)sulfite, and each scan was recorded every 3 min for 30 min.

(Bi)sulfite oxidation by Cu,Zn-SOD detected by optical spectroscopy, ESR, and oxygen uptake.
When we added a 20fold excess of (bi)sulfite to 1 mM Cu,ZnSOD, the absorp tion band at 680 nm characteristic of the active site of SOD1Cu(II) decreased slowly, then completely disappeared as the wildtype protein was reduced to Cu(I) (Figure 2A). We recorded the optical spectra every 3 min, and within < 30 min we observed a full reduction of Cu(II) to Cu(I) by (bi)sulfite. However, (B) Effect of (bi)sulfite on the ESR spectra of the active-site Cu 2+ of SOD. Spectra were observed from 50 µM Cu,Zn-SOD in 100 mM phosphate buffer, pH 7.4, at 77 K; incubations were performed at 37°C, and the samples were frozen in liquid nitrogen. Spectra were recorded with 0.5 mM (bi)sulfite at different time intervals: spectrum a, 1 min; spectrum b, 15 min; spectrum c, 30 min; spectrum d, 60 min. Instrumental parameters were as follows: microwave frequency, 9.50 GHz; microwave power, 2 mW; modulation amplitude, 4 G; receiver gain, 5 × 10 4 ; and scan rate, 9 G/sec. Each spectrum is a single scan.

Wavelength (nm) Magnetic field (G)
volume 118 | number 7 | July 2010 • Environmental Health Perspectives lower concentrations of protein (50 µM) and (bi)sulfite (500 µM) were sufficient for low temperature ESR spectra to detect the reduc tion of Cu(II) in Cu,ZnSOD ( Figure 2B). ESR data showed that the addition of a 10fold excess of (bi)sulfite to Cu,ZnSOD followed by a 1hr incubation resulted in an approximately 40% decrease in ESR intensity compared with the untreated pro tein ( Figure 2B). The anisotropic hyperfine coupling constant (A || = 135 G) remained unchanged during the incubation time, indi cating that (bi)sulfite does not bind directly to the active site Cu(II) (Strothkamp and Lippard 1981).
To determine whether (bi)sulfite is oxi dized by Cu(II) in Cu,ZnSOD, we also per formed roomtemperature ESR spintrapping experiments. When we mixed Cu,ZnSOD (50 µM) with (bi)sulfite (500 µM) in the presence of the spin trap DMPO (100 mM), it generated an intense ESR signal ( Figure 3A, spectrum a) corresponding to the assigned • SO 3 adduct of DMPO, DMPO/ • SO 3 -(a H β = 16.0 G and a N = 14.7 G) (Mottley and Mason 1988;Mottley et al. 1982aMottley et al. , 1982b. Previous studies have shown that (bi) sulfite stimulates the peroxidase function of Cu,ZnSOD and that • SO 3 is formed when the protein is treated with 1 mM H 2 O 2 in the presence of 20 mM (bi)sulfite (Sankarapandi and Zweier 1999). According to the authors, control experiments in the absence of Cu,ZnSOD confirmed that the • SO 3 signal was not due to direct oxidation of (bi)sulfite by H 2 O 2 , which is known from the literature (Flockhart et al. 1971;Mottley et al. 1982a) to proceed non enzymatically at high concen trations of H 2 O 2 via the following reaction: [5] To determine the effect of low and nontoxic concentration of H 2 O 2 (100 µM) and to con firm that • SO 3 is generated because of the enzymatic oxidation of (bi)sulfite, we per formed control experiments in the presence and absence of H 2 O 2 . Contrary to expec tation (Sankarapandi and Zweier 1999), addition of 100 µM H 2 O 2 had almost no effect on the ESR intensity of DMPO/ • SO 3 -( Figure 3A, spectra a and b), and omission of (bi)sulfite (Na 2 SO 3 ) or Cu,ZnSOD resulted in no radical adduct formation ( Figure 3A, spectra c and d, respectively). Control experi ments confirmed that the reaction is insensi tive to catalase, implying that H 2 O 2 is not involved (data not shown).
The proposed mechanism of enzymatic oxidation of (bi)sulfite to • SO 3 by the active Cu(II) site of Cu,ZnSOD proceeds in a one electron reduction reaction of Cu(II) by (bi) sulfite, similar to the oxidation of (bi)sulfite by HRP and prostaglandin H synthase (Mottley and Mason 1988;Mottley 1982aMottley , 1982bRoman and Dunford 1973). The resulting • SO 3 is known to react further with molecu lar oxygen to form -O 3 SOO • and SO 4 •in the free radical chain mechanism previously reported (Hayon et al. 1972;Mottley and Mason 1988;Reed et al. 1986). To confirm our hypothesis, we next investigated the con sumption of oxygen by 500 µM (bi)sulfite, with the reaction initiated by 0-500 µM Cu,ZnSOD. When • SO 3 reacted with oxy gen in the absence of spin trap, we observed oxygen consumption strongly dependent on the Cu,ZnSOD concentration ( Figure 3B).
Addition of 500 µM Cu,ZnSOD resulted in approximately 30% oxygen consumption after 15 min. When we examined the effect of the spin trap DMPO using 500 µM Na 2 SO 3 and 500 µM Cu,ZnSOD as the initiator, prior or later additions of 100 mM DMPO (the same amount used for the spintrapping ESR data) almost completely prevented oxy gen uptake ( Figure 3B), that is, no radical chain reactions ended in the formation of -O 3 SOO • and SO 4 •-.
To characterize the importance of Cu redox cycling at the enzymeactive site upon the generation of • SO 3 -, we mixed 500 µM Na 2 SO 3 with selected inhibi tors and initiated the reactions by 50 µM Cu,ZnSOD in the presence of 100 mM DMPO (Figure 4). The ESR intensity of the spectra showed that addi tion of 500 µM thiocyanate, azide, or cyanide in the presence of an equimolar amount of (bi)sulfite significantly inhibited • SO 3 pro duction. These results strongly suggest that because these anions bind directly to the Cu with high affinity, the enzymatic activity of Cu,ZnSOD is inhibited, and no further oxi dation of (bi)sulfite to sulfitederived radicals is possible.

Formation of HSA-DMPO nitrone adducts induced by the Cu,Zn-SOD-(bi)sulfite system as determined by immuno-spin trapping.
The optical and ESR data showed that (bi)sulfite is oxidized by Cu,ZnSOD to • SO 3 -, which will initiate the radical chain reaction with formation of -O 3 SOO • and SO 4 •via Equations 2-4. To characterize the ability of these radicals to oxidize amino acid(s) in target proteins, we incubated HSA with the enzyme generated by Cu,Zn-SOD and (bi)sulfite. Spectrum a was detected by mixing 50 µM Cu,Zn-SOD, 100 mM DMPO, 500 µM Na 2 SO 3 , and 100 µM H 2 O 2 in 100 mM phosphate buffer, pH 7.4, and then recorded immediately at room temperature. Spectra b-d are the same as spectrum a but without H 2 O 2 (spectrum b), without Na 2 SO 3 (spectrum c), or without SOD (spectrum d). Instrumental parameters were as follows: microwave frequency, 9.81 GHz; microwave power, 20 mW; modulation amplitude, 0.5 G; receiver gain, 5 × 10 4 ; scan rate, 0.5 G/sec. Each spectrum is a single scan. (B) Oxygen uptake curves as a function of Cu,Zn-SOD concentration. Sodium (bi)sulfite (Na 2 SO 3 , 500 µM) was placed in a chamber in 100 mM phosphate buffer, pH 7.4, and the reaction was initiated with different concentrations of Cu,Zn-SOD: spectrum a, 0 µM; spectrum b, 50 µM; spectrum c, 100 µM; spectrum d, 300 µM; spectrum e, 500 µM. The uptake curves were the same as spectrum e but with 100 mM DMPO added before (upper dotted line) or 400 sec after (lower dotted line) the addition of Cu,Zn-SOD.  Figure 4. Effect of Cu,Zn-SOD inhibitors on the ESR intensity of DMPO/ • SO 3 adducts. Spectrum a was observed from Cu,Zn-SOD (50 µM), Na 2 SO 3 (500 µM), and DMPO (100 mM) in phosphate buffer (100 mM, pH 7.4) and recorded immediately at room temperature. Spectra b-d are the same as spectrum a except with 500 µM of sodium thiocyanate (spectrum b), sodium azide (spectrum c), or sodium cyanide (spectrum d) added to the phosphate buffer. Instrumental parameters were as follows: microwave frequency, 9.81 GHz; microwave power, 20 mW; modulation amplitude, 0.5 G; receiver gain, 5 × 10 4 ; scan rate, 0.5 G/sec. Magnetic field (G) 3,500 3,520 and (bi)sulfite in the presence of DMPO and analyzed the reaction products by Western blotting using an antiDMPO polyclonal antibody (Detweiler et al. 2002). We chose a concentration of DMPO that was much less than the 100 mM used for the ESR and oxy gen uptake, so as to not inhibit the chain reac tion yet be sufficient for the protein radicals to react with DMPO for detection by anti DMPO antibody. We achieved the overall high yield of protein DMPO nitrone adducts by decreasing the DMPO concentration to 5 mM in the presence of the plasma concen tration of HSA (600 µM). We mixed sam ples containing 600 µM HSA with 500 µM Na 2 SO 3 in the presence of 5 mM DMPO and initiated the reactions with 10, 20, 30, 40, and 50 µM Cu,ZnSOD. Coomassie blue staining of the gel verified the amount of HSA present in all treatments and showed the presence of a single band at 60 kDa, which corresponds to the size of albumin, together with a small amount of HSA dimer at approximately 120 kDa ( Figure 5A). We also detected a very weak band at approximately 15 kDa at a Cu,ZnSOD concentration of 50 µM, corresponding to its monomer. We performed immunochemical detection of HSA-DMPO nitrone adducts using Western blotting and ELISA in parallel with SDS PAGE. As shown in Figure 5B, samples lacking Cu,ZnSOD, DMPO, or Na 2 SO 3 contained negligible antiDMPO cross reacting material. Incubation of HSA with > 10 µM Cu,ZnSOD resulted in a signifi cant increase in HSADMPO-derived nitrone adducts as assessed by ELISA ( Figure 5C). This result, together with the oxygen uptake experiments, demonstrated that 5 mM DMPO, because it did not trap the entire pri mary • SO 3 -, allowed the radical chemistry in Equations 2-4 to proceed with the formation of the damaging radical intermediates.
HSAderived nitrone adducts also depended on the (bi)sulfite concentration ( Figure 6A). Omission of HSA, DMPO, (bi)sulfite, or Cu,ZnSOD resulted in no immuno staining above the background level. When 0.1 mM (bi)sulfite and 600 µM albumin were oxidized in the presence of 5 mM DMPO and 50 µM Cu,ZnSOD, we detected a faint band of DMPO-nitrone adducts. Western blotting performed on reactions containing 0.25-3 mM (bi)sulfite showed increased production of DMPOHSA radicalderived nitrone adducts and very weak bands of DMPOHSA dimer at the higher (bi)sulfite concentrations.
We also determined the effect of time on the formation of HSA radicalderived nitrone adducts ( Figure 6B,C). In the presence of 5 mM DMPO, 500 µM Na 2 SO 3 , and 50 µM Cu,ZnSOD, Western blotting showed that DMPOHSA radicalderived nitrone adduct production increased with reaction time, reaching saturation at about 1 hr. ELISA data paralleled those from Western blotting ( Figure 6C).

Discussion
The present data confirm that the enzymatic oxidation of (bi)sulfite by Cu,ZnSOD pro ceeds via a radical mechanism as demonstrated using optical spectroscopy, oxygen uptake, and ESR experiments. Similar results have been reported for some peroxidases (e.g., HRP, prostaglandin H synthase) (Araiso et al. 1976;Mottley et al. 1982aMottley et al. , 1982b. Once the (bi) sulfite is oxidized by Cu(II) in Cu,ZnSOD and • SO 3 is formed, it reacts very rapidly with oxygen and generates -O 3 SOO • and SO 4 •- (Hayon et al. 1972), which-as very pow erful oxidants (E -O 3 SOO • / -O 3 SOOH = 1.1 V, E SO 4 •-/ SO 4 2-= 2.43 V)-can attack target proteins (e.g., HSA in plasma) (Neta et al. 1988;Steele and Appelman 1982) (Figure 1). Previous work on the oxidation of (bi)sulfite by the HRP-H 2 O 2 system and ESR spin trapping experiments showed that there is a strong competition between the spin trap DMPO and oxygen for • SO 3 - (Ranguelova and Mason 2009). In fact, in the latter system, the formation of the oxygenderived radicals -O 3 SOO • and SO 4 •was almost prevented by high DMPO concentrations (100 mM) ( Figure 3B), and a decrease of the spintrap concentration to ≤ 3 mM was required to trap protein radicals formed by -O 3 SOO • and SO 4 •- (Mottley and Mason 1988). The very slow consumption of oxygen observed even in the presence of 100 mM DMPO is likely due to the rapid reaction of • SO 3 with oxygen at a diffusioncontrolled rate to form -O 3 SOO • , which then reacts with SO 3 2to produce SO 4 •- (Figure 1).
(Bi)sulfite is one of the few sulfating agents approved by the Food and Drug Administration as a food preservative and  (Gunnison 1981). However, sulfites have been associated with adverse allergic and asthmatic reactions experienced by sulfitehypersensitive individuals. The most frequent sulfitereaction symptoms are difficulty in breathing, food intolerance symptoms, asthma, and occasion ally anaphylactic shock. There is no specific treatment for sulfite toxicity, and in general, to our knowledge, the mechanisms of the potentially toxic reactions of (bi)sulfite are poorly understood.
One reason for the toxic potential of (bi)sulfite is a deficiency of sulfite oxidase, the molybdenumcontaining enzyme that oxidizes sulfite to sulfate (SO 4 2-), and it is noteworthy that in cases of sulfite oxidase deficiency, the concentration of sulfite in plasma is abnor mal (> 1 mM) (Acosta et al. 1989;Johnson et al. 1980). The capacity of sulfite oxidase for sulfite oxidation is extremely high, with the reaction proceeding via a onestep, two electron oxidation to sulfate with no free radical intermediates (Cohen and Fridovich 1971). However, Yokoyama et al. (1971) showed that inhaled sulfur dioxide does reach the blood plasma, where the dissolved SO 2 [(bi)sulfite] forms oxidation products other than sulfate, such as Ssulfonates (Bechtold et al. 1993); this indicates the presence of another mechanism of (bi)sulfite oxidation besides the wellknown sulfite oxidase route. Another radical mechanism of xanthinede pendent aerobic oxidation of (bi)sulfite in the presence of xanthine oxidase has been proposed by McCord and Fridovich (1968). The authors concluded that xanthine oxidase, when catalyzing the aerobic oxidation of xan thine, generated a super oxide anion, which then served to initiate the (bi)sulfite chain reaction. A previous report from our labora tory (Mottley et al. 1982b) demon strated that incubation of (bi)sulfite with HRP and H 2 O 2 is not sensitive to the presence of SOD, con firming that the peroxidasecatalyzed pathway does not involve a superoxide chain reaction.
In the present study we used Cu,ZnSOD-(bi)sulfite as a source for genera tion of oxi dants ( -O 3 SOO • and SO 4 •-) that are diffus ible and radicals themselves to show their capability to oxidize the most abundant plasma protein (albumin) to protein radicals ( Figure 1). Our Western blot experiments showed that in the presence of DMPO, the Cu,ZnSOD-(bi)sulfite system produced sulfitederived radicals that oxidized albu min to produce proteincentered radicals trapped by the nitrone spintrap DMPO and detected as DMPOHSA nitrone adducts. When DMPO or any of the other system components were eliminated, no immuno staining appeared above the background signal levels, confirming that all of the reac tants are needed for detection of radicals. The extent of immunospin trapping increased with spintrap concentrations up to 10 mM and then decreased (data not shown). These results are consistent with the oxygen uptake experiments discussed above and with the ESR data for SO 4 •- (Mottley and Mason 1988), showing that lower concentrations of the spin trap must be used so that all the primary radicals are not trapped but have a chance for further reaction. Moreover, recent studies have confirmed the ability of DMPO to trap different protein radicals from the same system by varying its concentration (Bhattacharjee et al. 2007). Production of HSA nitrone adducts was also dependent on (bi)sulfite and Cu,ZnSOD concentra tions; only 500 µM (bi)sulfite was sufficient to detect positive results on the antiDMPO Western blots, whereas (bi)sulfite concentra tion in wines, where it is used as a preserva tive, is 6 mM (Gunnison 1981).
In summary, our study showed that Cu,ZnSOD-(bi)sulfite provides an enzy matic pathway to generate the reactive inter mediates -O 3 SOO • and SO 4 •-, which oxidize HSA residues to protein radicals. We also propose that Cu,ZnSOD may contribute to oxidative damage and tissue injury in (bi) sulfite (sulfur dioxide)-exacerbated allergic reactions. Our results suggest that SOD dependent, sulfitemediated oxidation of albumin residues is likely to occur in vivo, particularly at sites where Cu,ZnSOD concentration is higher. Further studies are necessary to clarify whether alterations in Cu,ZnSOD activity affect (bi)sulfite toxicity.