Bicarbonate Enhances Peroxidase Activity of Cu,Zn-Superoxide Dismutase ROLE OF CARBONATE ANION RADICAL AND SCAVENGING OF CARBONATE ANION RADICAL BY METALLOPORPHYRIN ANTIOXIDANT ENZYME MIMETICS*

Much evidence exists for the increased peroxidase activity of copper, zinc superoxide dismutase (SOD1) in oxidant-induced diseases. In this study, we measured the peroxidase activity of SOD1 by monitoring the oxidation of dichlorodihydrofluorescein (DCFH) to dichlorofluorescein (DCF). Bicarbonate dramatically enhanced DCFH oxidation to DCF in a SOD1/H 2 O 2 /DCFH system. Peroxidase activity could be measured at a lower H 2 O 2 concentration ( (cid:1) 1 (cid:1) M ). We propose that DCFH oxidation to DCF is a sensitive index for measuring the peroxidase activity of SOD1 and familial amyotrophic lateral sclerosis SOD1 mutants and that the carbonate radical anion (CO 3 .) is responsible for ox- idation of DCFH to DCF in the SOD1/H 2 O 2 /bicarbonate system. Bicarbonate enhanced H 2 O 2 -dependent oxida- tion of DCFH to DCF by spinal cord extracts of transgenic mice expressing SOD1 G93A . The SOD1/H 2 O 2 /HCO 3 (cid:2) - dependent oxidation was mimicked (cid:1) DMPO-X. Mn(IV) O (cid:5) from the reaction between MnTBAP and peroxymonosulfate (KSO 5 with DMPO to form MnTBAP (not DMPO-X. Photolysis of a pentammine the


idation of DCFH to DCF in the SOD1/H 2 O 2 /bicarbonate system. Bicarbonate enhanced H 2 O 2 -dependent oxidation of DCFH to DCF by spinal cord extracts of transgenic mice expressing SOD1 G93A . The SOD1/H 2 O 2 /HCO
؊dependent oxidation was mimicked by photolysis of an inorganic cobalt carbonato complex that generates CO 3 . .
Metalloporphyrin antioxidants that are usually considered as SOD1 mimetic or peroxynitrite dismutase effectively scavenged the CO 3 . radical. Implications of this reaction as a plausible protective mechanism in inflammatory cellular damage induced by peroxynitrite are discussed.
The mechanism(s) by which SOD 1 mutants induce amyotrophic lateral sclerosis (ALS) pathology remain unknown. A toxic gain-in-function of SOD1 mutants (i.e. increased peroxidase activity, nitration or aggregation) was proposed to be responsible for the enhanced toxicity of SOD1 (1)(2)(3)(4)(5). The mechanism of pathogenesis of diseases in which SOD1 is overexpressed also remains unknown (6,7). Thus, a thorough understanding of the basic mechanism underlying these processes (SOD1-or SOD1 mutant-dependent peroxidase activity, nitration, or aggregation) is critical for discovering new therapies. The first evidence for increased peroxidase activity of FALS SOD1 mutants came from the use of electron spin resonance (ESR)-spin trapping (8). By monitoring the oxidation of the spin trap 5,5Ј-dimethyl-1-pyrroline N-oxide (DMPO) to its hydroxyl adduct (DMPO-OH), the investigators concluded that FALS SOD1 mutants exhibit an increased peroxidase activity (8).
Recently, it was reported that the bicarbonate anion (HCO 3 Ϫ ) is required for the peroxidase activity of SOD1 and FALS SOD1 mutants (9,10). Three decades ago, Hodgson and Fridovich (11,12) proposed that a "copper-bound hydroxyl radical" (SOD-Cu 2ϩ -⅐ OH) generated in the reaction between SOD1 and H 2 O 2 was able to oxidize several anionic ligands (11,12). The x-ray crystal structure of SOD1 indicates that access to the active site of copper is via a narrow channel that restricts the entry of a large molecule. However, a relatively small and physiologically relevant anion such as HCO 3 Ϫ could reach the active site of SOD1 and subsequently undergo oxidation to CO 3 . , a diffusible oxidant that could leave the active site and cause oxidation of various substrates in free solution (9,10,13). This model has provided a new perspective on the peroxidative mechanism of SOD1 and explains in part the published discrepancies on the nature and structure of oxidants determined by ESR spin trapping (14 -16). Bicarbonate-stimulated peroxidase activity of SOD1 has now been validated by several different approaches (9,10,13). For example, bicarbonate is required for SOD1 peroxidase-catalyzed hydroxylation of DMPO to DMPO-OH, oxidation and nitration of tyrosine to dityrosine and nitrotyrosine, and oxidation of ABTS to the ABTS cation radical. Azulenyl nitrone was converted to an aldehyde in the spinal cords of transgenic mice with SOD G93A mutant (17,18). Aldehyde formation was attributed to trapping of SOD1 peroxidase-generated hydroxyl or peroxyl radicals by the azulenyl nitrone (18). This was the first in vivo spin-trapping evidence of SOD1 peroxidase activity. The biological relevance of SOD1 peroxidase activity has recently been questioned because of the requirement of a high H 2 O 2 concentration needed to catalyze SOD1 peroxidasedependent oxidations (19). We have previously reported that bicarbonate-dependent SOD1 peroxidase activity could be monitored using the DCFH fluorescence method (20). In the present study, we have extended the scope of our preliminary report (20). We show that the bicarbonate-dependent SOD1 peroxidase activity at a lower H 2 O 2 concentration can be conveniently measured by monitoring oxidation of dichlorodihydrofluores-cein (DCFH) to dichlorofluorescein (DCF). The extent of DCFH oxidation was greatly amplified by aromatic amino acids and a tyrosine-containing peptide. We also show that SOD1/H 2 O 2 / HCO 3 Ϫ -dependent oxidation can be mimicked by photolysis of an inorganic colbalt carbonato complex that generates CO 3 . .
Metalloporphyrin antioxidants (e.g. Fe(III)TBAP, Mn(III) TBAP) scavenged the CO 3 . radical to yield the respective oxotetravalent iron or manganese species. These oxidants reacted with cyclic nitrone spin traps to form a characteristic ESRactive oxidation product. Thus, it may now be feasible to distinguish between the hydroxyl radical and carbonate anion radical formation in HCO 3 Ϫ -dependent SOD1 peroxidase and peroxynitrite/CO 2 systems.

EXPERIMENTAL PROCEDURES
Tryptophan, tyrosine, hydrogen peroxide, and angiotensin were purchased from Sigma, and dichlorodihydrofluorescein was obtained from Molecular Probes, Inc. (Eugene, OR). Bovine Cu,Zn-superoxide dismutase was obtained from Roche Diagnostics. The other chemicals used were from Aldrich. Pentammine carbonato complex of Co(III) was synthesized according to the published procedure (21). Briefly, 30 g of Co(NO 3 ) 2 ⅐6H 2 O in 15 ml of water was added to 45 g of ammonium carbonate dissolved in 45 ml of water, followed by the addition of 75 ml of concentrated ammonia. Air was bubbled through the solution for 24 h. The resulting solution was cooled in an ice bath, and the solid product was recrystallized by dissolving in 55 ml of water at 90°C and then slowly cooling the solution in an ice bath. Pure crystals were isolated and used in the experiments.
Measurement of Bicarbonate-mediated Peroxidase Activity-DCFH diacetate was hydrolyzed freshly for every experiment according to the published procedure (22). Briefly, DCFH was dissolved in methanol and hydrolyzed by NaOH (0.01 M) in the dark for 30 min at room temperature and stored in an ice bath during the experiment. SOD1 peroxidase activity was measured as follows. SOD1 (1 mg/ml) was mixed with a solution of DCFH (40 -80 M) in a phosphate buffer (0.1 M, pH 7.4) containing DTPA (100 M) and various amounts of sodium bicarbonate (0 -25 mM). The reaction was initiated by adding H 2 O 2 , and DCF formation was monitored at 504 nm. For experiments using low concentrations of H 2 O 2 (1-10 M), DCF formation was measured fluorometrically (excitation ϭ 495 nm, emission ϭ 520 nm).
Measurement of SOD1 Peroxidase Activity in Spinal Cord Homogenates-Transgenic mice at different ages (60 and 90 days) were sacrificed, and spinal cords isolated from brain were homogenized in 0.4 ml of phosphate buffer (0.1 M) containing DTPA (100 M). Following centrifugation of homogenates at 8000 rpm for 20 min, supernatants were removed and kept in an ice-bath for SOD1 peroxidase activity assays. A 20 l of homogenate was incubated with H 2 O 2 (3 mM) and DCFH (80 M) in a phosphate buffer (0.1 M, pH 7.4) containing DTPA (100 M). The reaction mixture was incubated at 37°C for 20 min, and the formation of DCF was measured at excitation ϭ 495 nm and emission ϭ 520 nm. The protein concentration was determined by Lowry's method using bovine serum albumin as a standard.
ESR Spin-trapping Analysis-A typical reaction mixture (ϳ100 l) for ESR experiments consisted of SOD (1 mg/ml), DMPO (25 mM), H 2 O 2 (1 mM), and bicarbonate (25 mM) in a phosphate buffer (0.1 M) containing DTPA (100 M). ESR spectra were recorded at room temperature on a Varian E-109 spectrometer operating at 9.5 GHz and with a 100-kHz field modulation equipped with a TE 102 cavity.

Bicarbonate-mediated
absorbance at 504 nm (Fig. 1A). Fig. 1A (inset) shows the time-dependent increase in the absorbance of DCF as a function of HCO 3 Ϫ concentration. Fig. 1B shows the H 2 O 2 dependence of SOD1/H 2 O 2 /HCO 3 Ϫ -mediated oxidation of DCFH to DCF. As seen in Fig. 1B, bicarbonate-mediated SOD1 peroxidase activity can be easily measured at low H 2 O 2 concentration (ϳ1 M). Admittedly, the rate of reactions catalyzed by HCO 3 Ϫdependent SOD1 peroxidase activity is relatively slow; however, the ability to "export" a potent and diffusible oxidant (i.e. CO 3 . ) from the active site of SOD1 to initiate the oxidation of amino acids in peptides is unique for this peroxidase system (9,13 (not shown). These results suggest that the oxidant responsible for oxidation of DCFH in the SOD1/H 2 O 2 /HCO 3 Ϫ system is not derived from "free" copper ions released from the active site of SOD1.
The addition of tyrosine, tryptophan, or angiotensin II (a peptide containing tyrosine) greatly increased the oxidation rate of DCFH to DCF (Fig. 2). Tryptophan (100 M) and tyrosine (1 mM) increased SOD1/H 2 O 2 /HCO 3 Ϫ -dependent oxidation of DCFH to DCF by 4-fold (Fig. 2, A and B). Fig. 2D shows that both tryptophan and tyrosine increased SOD1 G93A /H 2 O 2 /HCO 3 Ϫdependent DCFH oxidation. Results from these studies indicate that bicarbonate is essential for the SOD1 peroxidase activity and that aromatic amino acids (tyrosine and tryptophan) and peptides containing these amino acids exacerbate HCO 3 Ϫ -mediated DCFH oxidation by SOD1   dition of H 2 O 2 (1 mM) to an incubation mixture containing SOD1 (1 mg/ml), DMPO (25 mM), and DTPA (100 M) did not yield a significant amount of DMPO adduct in the absence of bicarbonate (Fig. 3, left). In the presence of 25 mM HCO 3 Ϫ , an intense signal (marked E) was obtained because of the DMPO-OH (␣ N ϭ 14.9 G, ␣ H ϭ 14.9 G). Previously, using oxygen-17-labeled water, formation of this adduct was attributed to a direct reaction between CO 3 . and DMPO, leading to an intermediate that underwent hydrolysis (15). The addition of the azide anion resulted in the formation of the DMPO-azide adduct (DMPO-N 3 Fig. 3, left) (23). The appearance of a small DMPO-N 3 Ϫ signal in the absence of HCO 3 Ϫ was attributed to the direct oxidation of N 3 Ϫ by the copper-bound hydroxyl radical at the active site of SOD1 (11,12). A bicarbonate-mediated increase in the DMPO-N 3 Ϫ signal is due to the oxidation of azide anion to the azide radical by CO 3 . (24) (27). The formation of an ethanol-derived carbon-centered radical is presumably due to the abstraction of a hydrogen atom from ethanol by CO 3 . (28,29).
Independent evidence for the intermediacy of CO 3 . was obtained from photolysis studies using the pentammine carbonato complex of Co(III) (Fig. 3, right). The UV photolysis of this cobalt complex has been shown to generate the authentic CO 3 .
radical (30) that reacted with substrates (e.g. azide, formate, and ethanol) to yield the same type of radical adducts as de-tected in the SOD1/H 2 O 2 /HCO 3 Ϫ system. In the absence of UV light, no radical adducts were detected.
Bicarbonate-mediated Oxidation of Azulenyl Nitrone to Azulenyl Aldehyde- Fig. 4A (middle) shows the bicarbonate-dependent oxidation of azulenyl nitrone in incubations containing SOD1, H 2 O 2 , and DTPA. In the absence of bicarbonate, oxidation of azulenyl nitrone (AZN) to azulenyl aldehyde (AZA) was negligible in phosphate buffers containing SOD1, H 2 O 2 , and DTPA (Fig. 4A, left). Bicarbonate alone did not facilitate the oxidation of AZN to AZA (not shown). AZN and AZA exhibit a characteristic optical spectral pattern (Fig. 4A, right). Fig. 4, B and C, shows that a bis-nitrone also undergoes the bicarbonatedependent oxidation to form a bis-aldehyde. Photolysis of solutions containing a pentammine carbonato complex of Co(III) (that generates the carbonate radical anion) and AZN yielded AZA. AZA formation was not detectable in the absence of the pentammine carbonato complex of Co(III) (not shown). Based on these results, we suggest that CO 3 . is responsible for SOD1/ H 2 O 2 /HCO 3 Ϫ -mediated oxidation of AZN to AZA and bis-nitrone to bis-aldehyde.
The Effect of Bicarbonate on SOD1 Peroxidase Activity in Mice Spinal Cord Extracts-The bicarbonate-dependent peroxidase activity of SOD1 was evaluated from spinal cord extracts of SOD1 transgenic and SOD1 G93A transgenic mice. A significant increase in the oxidation of DCFH to DCF was observed in spinal cord extracts from SOD1 G93A transgenic mice (Fig. 5). Bicarbonate-mediated SOD1-peroxidase activity was the highest in the spinal cord extracts of 90-day old mice. In all age groups, the SOD1 peroxidase activity was clearly enhanced in the presence of bicarbonate. The SOD1 G93A mice spinal cord extracts also exhibited a significant increase in bicarbonate-dependent peroxidase activity (Fig. 5). Even in the absence of added bicarbonate, the spinal cord extracts of 90-day old wild type and SOD G93A transgenic mice exhibited a slight increase in the peroxidase activity (Fig. 5). This could be due to the residual bicarbonate content in spinal cord extracts. As re- Middle, the same as above but containing bicarbonate (25 mM). The decrease in the absorbance maximum at 540 nm corresponds to a decrease in concentration of AZN; the increase in the absorbance at 460 nm corresponds to an increase in the concentration of azulenyl aldehyde. Right, the optical spectra of authentic AZN and AZA. B, the incubation mixture is the same as A but contained a bisnitrone instead of AZN in the presence of bicarbonate. The decrease in bis-nitrone concentration is observed at 425 nm. Inset, a concomitant increase in formation of bis-aldehyde. C, the effect of bicarbonate on the oxidation of bis-nitrone in incubations containing SOD1, H 2 O 2 , and DTPA in a phosphate buffer. ported by Liu et al. (18), the two transgenic lines, SOD1-G93A and SOD1, had comparable elevation of total Cu,Zn-SOD protein and activity in spinal cord at 60 and 90 days of age. This suggests that the increased peroxidase activity of the SOD1-G93A mutant form is due to the altered enzymatic function in the mutant form.
Trapping of Carbonate Anion Radical by Metalloporphyrin Antioxidants-The metalloporphyrins MnTBAP and FeTBAP were reported to inhibit the progression of disease in an ALS mouse model (31). Therefore, we decided to investigate the effect of metalloporphyrins in radical reactions catalyzed by SOD1/H 2 O 2 /HCO 3 Ϫ system. As shown in Fig. 6A, the addition of SOD1 to incubations containing H 2 O 2 , HCO 3 Ϫ , and DMPO yielded the DMPO-OH adduct. In the presence of MnTBAP, the four-line ESR spectrum of DMPO-OH was replaced by a narrow nine-line spectrum (␣ N ϭ 7.1 G, ␣ H (2) ϭ 4.2 G) that is assigned to the oxidation product, DMPO-X, in which the free electron is interacting with a nitrogen and two equivalent protons (Fig. 6A) (32). In the absence of HCO 3 Ϫ , no ESR signal was obtained. Again, in the presence of FeTBAP, the DMPO-X spectrum was obtained from incubations containing SOD1, H 2 O 2 , HCO 3 Ϫ and DMPO (Fig. 6). These findings (Fig. 6) suggest that HCO 3 Ϫ -derived radical (i.e., CO 3 . ) reacted with MnTBAP or FeTBAP and DMPO to form the DMPO-X. We propose that the putative oxidant formed from the reaction between Mn(III)TBAP and CO 3 . is Mn(IV)ϭO. Additional proof for this reaction mechanism was obtained from the UV photolysis of a cobalt carbonato complex and MnTBAP. In the presence of MnTBAP, the ESR spectrum due to the DMPO-OH was replaced by that of the oxidation product, DMPO-X (Fig. 6B). That a higher oxidant, Mn(IV)ϭO, is involved in the oxidation of DMPO to DMPO-X became evident from experiments using the oxidant KSO 5 and MnTBAP. When SO 5 Ϫ was mixed with MnTBAP, we observed a characteristic optical spectrum of Mn(IV) complex (data not shown). The addition of DMPO to a solution containing KSO 5 and MnTBAP yielded the characteristic ESR of the DMPO-X adduct (Fig. 6C). In the absence of SO 5 Ϫ or MnTBAP, no ESR signal was detected. Next, we showed that the hydroxyl radicals generated from the Fenton reaction (Fe 2ϩ and H 2 O 2 ) did not yield the DMPO-X spectrum (Fig. 6D) in the presence of MnTBAP, suggesting that the intermediate formed from the reaction of ⅐ OH or CO 3 . with MnTBAP may be different. However, FeTBAP reacted with ⅐ OH to form an iron-oxo intermediate, which oxidized DMPO to DMPO-X (Fig. 6D). Results from these experiments indicated that MnTBAP is an effective scav-enger of CO 3 . . (The rate constant between CO 3 . and MnTBAP is estimated to be at least 100 times greater than that of CO 3 . and DMPO.) Results also indicate that "free" hydroxyl radicals are not formed in the SOD1/H 2 O 2 /HCO 3 Ϫ system.

Intermediacy of a Carbonate Radical Anion-
The enhanced peroxidase activity of the SOD1/H 2 O 2 /HCO 3 Ϫ system has been attributed to the formation of a diffusible oxidant, carbonate radical anion (CO 3 . ) formed from a one-electron oxidation of HCO 3 Ϫ at the active site of SOD1 by a copper-bound hydroxyl radical (i.e. SOD-Cu 2ϩ -⅐ OH). It was proposed that CO 3 . could diffuse out the active site and cause oxidation of substrates in solution (11)(12)(13)(14). We suggest that CO 3 . is responsible for oxidation of DCFH to DCF via a hydrogen abstraction reaction as follows. Tyrosine, tryptophan, and angiotensin II enhanced the oxidation of DCFH by SOD1/H 2 O 2 in the presence of bicarbonate (Fig. 2). CO 3 . has been shown to react rapidly with tyrosine and tryptophan (k ϭ 10 7 to 10 8 M Ϫ1 s Ϫ1 ) forming the corresponding tyrosyl or tryptophanyl radical (28,29). These phenoxyl radicals will probably react with DCFH and enhance DCF formation as shown below. Dichlorodihydrofluorescein-derived radical, DCF ⅐ , has also been reported to undergo autoxidation in air forming DCF and the superoxide anion (33).
Recently, it was reported that human neuroblastoma cells transfected with SOD1 G93A mutant could oxidize DCFH to a much greater extent than cells transfected with wild-type SOD1 (34). The increased oxidation of DCFH to DCF was suggested to be due to an increased peroxidase activity of G93A enzyme (34). The present data clearly show that HCO 3 Ϫ markedly enhanced the peroxidase activity of SOD1/H 2 O 2 (Fig. 1). Thus, the intracellular HCO 3 Ϫ in SOD1 G93A transfected cells might have been responsible for the increased oxidation of DCFH.
The electron transfer reaction between CO 3 . and azide, formate, and nitrite anion will result in the formation of the respective ligand radical anion. In accord, radical adducts formed from trapping of these radicals by DMPO were detected in the SOD1/H 2 O 2 /HCO 3 Ϫ system (Fig. 3). Independent evidence for CO 3 . -mediated oxidation reactions was obtained from UV photolysis of the pentammine carbonato complex of Co(III) as shown below.
Irradiation of the pentammine carbonato complex of Co(III) in the presence of DMPO and azide, formate, or DCFH gave rise to radical adducts (Fig. 3, left) that were similar to those obtained from SOD1/H 2 O 2 /HCO 3 Ϫ system (Fig. 3, right). Bicarbonate-dependent Oxidation of Azulenyl Nitrone by SOD1/H 2 O 2 -Recently, Gurney, Becker, and co-workers (17,18) provided the first in vivo spin-trapping evidence for increased oxygen radical formation in the SOD1 G93A transgenic mouse model of FALS. When AZN was administered to the nontransgenic, wild-type transgenic, and mutant transgenic mice of different ages (30, 60, and 90 days), increased levels of AZA (the oxidative metabolite of AZN) were detected in spinal cord extracts of mutant mice than in transgenic wild-type and nontransgenic mice. The open-chain nitrones such as ␣-phenyl tert-butyl N-nitrone (PBN) were oxidized by SOD1/H 2 O 2 / HCO 3 Ϫ , forming the corresponding hydroxyl adduct; however, the PBN-hydroxyl adduct (PBN-OH) underwent further decomposition to form the N-tert-butyl hydronitroxide (not shown) (35). In addition, using the oxygen-17-labeled water, the source of the oxygen atom in PBN-OH was shown to be derived from water (not shown) (36). Based on these findings, we propose that the bicarbonate-derived radical (i.e. CO 3 . ) is responsible for oxidation of AZN to its cation radical, which is then hydrolyzed to give AZA (Scheme 1). Scavenging of CO 3 . by Metalloporphyrins: Plausible Mecha-SCHEME 1. A proposed mechanism for carbonate radical anion-mediated oxidation of nitrone to aldehyde. SCHEME 2. The proposed mechanism of oxidation of DMPO mediated by CO 3 . and metalloporphyrins. P-Mn(III) refers to the manganese(III)-substituted porphyrin. Similarly, P-Fe(III) refers to the iron(III)-substituted porphyrin. SCHEME 3. Decomposition of the nitrosoperoxycarboxylate intermediate formed from the reaction between peroxynitrite and CO 2 and the proposed reaction between CO 3 . and metalloporphyrins (FeTBAP and MnTBAP).  (37). However, using this technique, we were unable to detect CO 3 . produced during HCO 3 Ϫdependent peroxidase activity of SOD1. 2 Clearly, this is due to a much slower rate of formation of CO 3 . in this system. Recently, we proposed that CO 3 . formed from one-electron oxidation of HCO 3 Ϫ at the active site of SOD1 could diffuse out of the active site and cause oxidation/hydroxylation of DMPO to an ESR active DMPO-OH adduct (13). In order to prove that the DMPO-OH adduct was not formed from trapping of hydroxyl radicals by DMPO, we used the oxygen-17-labeled water and determined that the oxygen atom in DMPO-OH adduct originated from H 2 O and not from H 2 O 2 (13). Despite these reports (9,13), investigators continue to attribute the formation of DMPO-OH in SOD1/H 2 O 2 /HCO 3 Ϫ system to trapping of free hydroxyl radicals (38). Most conventional hydroxyl radical scavengers (e.g. azide anion, ethanol, formate, etc.) react with ⅐ OH or CO 3 . and DMPO to form the same radical adduct. In this regard, the use of MnTBAP may be able to distinguish spectroscopically the intermediacy of CO 3 . and ⅐ OH radicals. As shown in this study, MnTBAP oxidizes DMPO to a characteristic DMPO-X product in the presence of CO 3 . but not of ⅐ OH.
We propose the following mechanism (Scheme 2) for oxidation of DMPO to DMPO-X. The authentic Mn(IV)ϭO species ( ϭ 425 nm) formed from the reaction between MnTBAP and potassium peroxymonosulfate (KSO 5 ) reacted with DMPO to form MnTBAP (not shown) and DMPO-X. Photolysis of a mixture containing MnTBAP and the cobalt pentammine carbonato complex produced the same characteristic optical changes associated with the formation of Mn(IV)ϭO (not shown).
In vitro evidence suggests that peroxynitrite-mediated nitration of SOD1 and ALS-linked SOD1 mutants may cause an enhanced neuronal cell death in ALS (4), although the in vivo relevance of this hypothesis is debatable (39,40). Nevertheless, it is now increasingly clear that nitration/oxidation reactions mediated by peroxynitrite are modulated by CO 2 (13, 41), a ubiquitous cellular component. As shown in Scheme 3, the CO 3 .
radical plays a key role in determining the tyrosine oxidation/ nitration product yields (42). Metalloporphyrins (i.e. manganese(III) porphyrin analogs) have been reported to react with both peroxynitrite and the nitrosoperoxocarboxylate intermediate (43,44). Previously, it has been shown that CO 3 . is involved in both nitration and oxidation of tyrosine (13). Clearly, metalloporphyrin antioxidants that can scavenge the carbonate anion radical will inhibit both oxidation and nitration of tyrosine. The cytoprotective effect of iron metalloporphyrin catalysts against cytokine and endotoxin-mediated cellular damage may be attributed to the rapid scavenging of CO 3 . by iron-porphyrins (45). The ability of CO 3 . to induce aggregation, nitration, and oxidation reactions in SOD1 suggests that there may be a common radicalmediated mechanism that is operative in aggregation as well as nitration/oxidation reactions catalyzed by SOD1. By scavenging CO 3 . , metalloporphyrin antioxidants might inhibit aggregation, oxidation, and nitration of SOD1. Metalloporphyrin antioxidants (e.g. MnTBAP and FeTBAP) effectively scavenged the carbonate anion radical to form a characteristic ESR-active oxidation product in the presence of cyclic nitrone spin traps.
The use of MnTBAP thus confirmed our original tenet that the SOD1/H 2 O 2 /HCO 3 Ϫ system generated the CO 3 . radical and not the hydroxyl radical. The reasons for increased peroxidase activity in the spinal cord extracts of SOD1 G93A are not immediately obvious. Liochev et al. (46) reported that the rates of inactivation of SOD1 A4V , SOD1 G93A , and wild-type SOD1 in the presence of H 2 O 2 are similar. Thus, additional factors that enhance SOD1/ H 2 O 2 /HCO 3 Ϫ -dependent oxidations should be considered. It is likely that bicarbonate exacerbates the intrinsic difference between wild-type and mutant SOD1 with respect to peroxidase activity. Preliminary ongoing research from our laboratory indicates that bicarbonate-mediated peroxidase activity enhances SOD1 dimerization and aggregation that is inhibited by MnTBAP and nitrone spin traps. These findings suggest that CO 3 . may be involved in SOD1 aggregation.
In conclusion, we showed that bicarbonate dramatically enhanced DCFH oxidation to DCF in the SOD1/H 2 O 2 /DCFH system. Peroxidase activity could be measured even at 1 M H 2 O 2 in this system. We suggest that DCFH oxidation to DCF is a sensitive method for measuring the peroxidase activity of SOD1 and FALS SOD1 mutants and that bicarbonate-derived radical (CO 3 . ) is responsible for oxidation of DCFH to DCF (20,47). This mechanism also accounts for oxidation/hydroxylation of antioxidant nitrones in ALS animal models. The scavenging of CO 3 . by metalloporphyrin antioxidant mimetics (i.e. MnTBAP and FeTBAP) may be an important mechanism by which these compounds afford cytoprotection in cytokine-mediated cellular damage.