- Email: [email protected]
Assays using horseradish peroxidase and phenolic substrates require superoxide dismutase for accurate determination of hydrogen peroxide production by neutrophils
~
Free RadicalBiology& Medicine,VoL 17, No. 2, pp. 161-164, 1994 Copyright© 1994Blsevia ~imce I ~ Prin~d in tlw USA.All flglmrmerved 0 8 9 1 o $ ~ $6.00 + .00
rergemn
0~l-.¢,~904)E00ZT-6
"~ BriefCommunication ASSAYS USING HORSERADISH PEROXIDASE AND PHENOLIC SUBSTRATES REQUIRE SUPEROXIDE DISMUTASE FOR ACCURATE DETERMINATION OF HYDROGEN PEROXIDE PRODUCTION BY NEUTROPHILS ANTHONY J. KEITLE, ANITRA C. CARR, and CHRISTINE C. WINTERBOURN Free Radical Research Unit, Department of Pathology, Christchurch School of Medicine, Christchurch, New Zealand
(Received 24 November 1993; Accepted 7 January 1994) Abstract--We used honeradish peroxidase and either scopoletin, homovanilllc acid, or phenol red to measure hydrogen peroxide ganentted by human neutrophils.With these assays, superoxidedismutasesj£nificantlyincreasedthe anmunt of bydrogan peroxide detected. In contrast, it had no effect when the accumul~on of hydrogen peroxide was measored with a hydmgan peroxide electrode. We propose that superoxideinterferes with horseradishp e m ~ t assays so that hydrogenperoxide is n ~ s t i m a t e d , Thus, when using these assays, superoxide dismutase must he added to neutmphils to ensure that all the hydrogen peroxide they produce is detected. Keyworde---Neutrophils, Hydrogen peroxide, Horseradish peroxidase, Free radicals
generated by neutmphils.2 Also, when measuring hydrogen peroxide, azide is added to inhibit myelopemxidase. Hence, other explanations are required to account for the effect of superoxide dismutase on the yield of hydrogen peroxide. We have measured hydrogen peroxide production by neutrophils using either horseradish peroxidase with phenolic substrates, or a hydrogen peroxide electrode. Hydrogen peroxide was detected in the presence and absence of superoxide dismutase, and was related to superoxide production. We show that when measured with a hydrogen peroxide electrode the amount of hydrogen peroxide produced by neutrophils is unaffected by superoxide dismutase, and it accounts for all the superoxide generated. We propose that superoxide interferes with horseradish peroxidase-dependent assays so that superoxide dismutase must he added to cells to ensure that all the hydrogen peroxide they produce is detected.
INTRODUCTION
Stimulated neutrophils generate superoxide which dismutates to hydrogen peroxideJ a The most commonly used methods for measuring hydrogen peroxide production by neutrophils employ horseradish peroxidase in combination with a phenofic substrate. 3-6 Hydrogen peroxide is detected because, in conjunction with horseradish peroxidase, it oxidizes the phenol to cause a change in fluorescence or absorbance. Superoxide dismutase is not normally added to these assays, but when included, it has been shown to increase the production of hydrogen peroxideJ -6 The usual explanation given for this increase is that not all the superoxide discharged by neum~phils spontaneously dismutates. It is presumed that some of it is involved in oxidation or reduction reactions, and that the addition of superoxide dismutase diverts superoxide from these reactions, leading to an artefactual increase in hydrogen peroxide production. Proposed reactions of superoxide, other than dismutation, include the reduction of redox intermediates of myeloperoxidaseTM and the formation of hydroxyl radicals. 9a° However, ff produced, hydroxyl radicals account for less than 1% of the superoxide
MATERIALS AND METHODS
Neutrophils were isolated from the blood of healthy donors by Ficoll-Hypaque centrifugefion, dextran sedimentation, and hypotonic lysis of contaminating red cells, n Horseradish peroxidase (Type I), bovine erythrocyte superoxide dismutase, bovine liver catalase,
Address c c ~ l a ~ to: Anthony J. Keltic, Free Radical Research Unit, Depmlmentof Pathology, ChristchurchSchool of Medicine, Christchurch, New Zealand. 161
162
A.J. KeTn~ et al.
phorbol myristate acetate, and cytochrome c (Type HI) were purchased from the Sigma Chemical Co. (St Louis, MO).
Hydrogen peroxide assays Scopoletin assay. Horseradish peroxidase (20 nM) and scopoletin (40/~M) were added to 105 cells in 10 mM phosphate buffer pH 7.4 containing 140 mM NaC1, 1 mM CaCI2, 0.5 mM MgC12, 1 mg ml -i of glucose, and 1 mM azide. Neutrophils were preincubated for 10 rain at 37°C and then stimulated with phorbol myristate acetate (100 ng ml-~). The final reaction volume was 1 ml. After 30 min, the reaction was stopped by placing mhes in melting ice and pelleting the cells by centrifugation. The hydrogen peroxide in the supernatant was determined by measuring the loss in fluorescence due to the oxidation of scopoletin (ex 350 nm; em 460 um).n Alternatively, hydrogen peroxide was allowed to accumulate for 30 rain before cells were pelleted, then horseradish peroxidase and scopoletin were added to the supernatant. HomovaniUic acid assay. Conditions were the same as in Scopoletin assay, except the concentration of homovanillic acid was 100/dvl. After 30 min of stimulation reactions were stopped by adding 120 #1 of 100 raM glycine-NaOH buffer pH 12 containing 25 mM EDTA and cells were pelleted. The hydrogen peroxide produced was dete~jned by measuring the increase in fluorescence in the supemat_ant due to the oxidation of homovanillic acid (ex 312 nm, em 420 nm). 5 Phenol red assay. Conditions were the same as in Scopoletin assay, except the concentratiom of horseradish peroxidase, cells, and phenol red were 170 nM, 5 x 105 ml -t, and 280 ~M, respectively. After the reactions were stopped, cells were pelleted by centrifugafion and 10 #1 of 1 M NaOH was added to 1 ml of supernatant. The hydrogen peroxide produced was determined by measuring the increase in A~m due to the oxidation of phenol red. 13
Each of the horseradish peroxidase-dependent assays was calibrated with known concentrations of hydrogen peroxide (e~o 43.6 M -1 c m - t ) J 4
Hydrogen peroxide electrode. Hydrogen peroxide production was monitored continuously with a YSI model 25 oxidase meter fitted with a YSI 2510 oxidase probe (Yellow Springs Instntment Co., Yellow Springs, O H ) as desorihed previously.~5Briefly, 106 neutrophils m1-1 were stimulated with phorbol mynstate acetate in the presence of 1 raM azide. Other conditions were as described in Scopoletiu assay, except for the omission
A
B
0.7 0.6
scopot~
Phe~ned
0.4 0.3 0.2
0.1 0.0 Fig. 1. Effects of supemxide dismutase on the ratio of hydrogen peroxide to superoxide produced by neulrophils. Cells were stimulated with phorbol myristate acetate in the presence (D) or absence (1) of 50 #g m1-1 of supe~oxide dismutase. (A) Hydrogen peroxide was determined by adding horseradish peroxidase and either scopoletin, homovanillic acid, or phenol red before stimulating the cells (n = 6). (B) Hydrogen peroxide was also measured by allowing it to accumulate for 30 rain before adding ho~_A~_h peroxidase and scopoletin to supetnatan~, and by using a hydrogen peroxide electrode (n = 3). Superoxide production, measured as superoxide dismutase-inhibitable cytochrome c reduction, was 18.0 -+ 2.0 nmol per 105 cells in 30 rain (n = 6). Reactions were cmxiad out in duplicate and dAt~ are represented as the ratio of the means of average values for cells from n individuals _+ the standard error of the mean. Conditions are given in the Materials and Methods section.
of horseradish peroxidase and scopoletin. Hydrogen peroxide accumulation was monitored for 20 rain. Superoxide production was measured as superoxide-dismutase-inhibitable cytochrome c reduction for 30 rain under the same conditions as described for the scopoletin assay. ~ Fluorescence was measured with an Aminco Bowman spectrophotofluorometer, and absorbances with a CarylE spectrophotometer. RESULTS AND DISCUSSION
Hydrogen peroxide production was measured by adding horseradish peroxidase and either scopoletin, homovanillic acid, or phenol red to neutrophils, and relating it to the amount of superoxide they generated. If all the superoxide spontaneously dismutated to hydrogen peroxide according to reaction (1), the ratio of hydrogen peroxide produced to superoxide produced would be 0.5. 0 2 + HO2 + H + --* H202 + 0 2
(1)
When cells were stimulated after addition of horseradish peroxidase and one of the phenolic subsidies, this ratio was significantly less than 0.5 (Fig. 1A). For scopoletin, only half the superoxide reacted to form hydrogen peroxide. Our results indicate that superoxide must have reacted in other ways besides spontane-
Assays for hydrogenperoxide productionby neutmphils ously dismutaflng. In agreement with previous studies, 3-6 superoxide dismutase increased the yield of hydrogen peroxide, so that the ratio of hydrogen peroxide to superoxide was close to theoretical. Because the amount of hydrogen peroxide detected varied with the phenolic substrate used, we tested the possibility that superoxide was reacting with components of the horseradish peroxidase assay. Cells were stimulated and hydrogen peroxide was allowed to accumulate for 30 rain before adding horseradish peroxidase and scopoletin to the supernatant. Using this protocol, superoxide dismutase had no effect on the hydrogen peroxide production, and all the superoxide was accounted for by the hydrogen peroxide detected (Fig. 1B). Therefore, superoxide discharged by neutrophils dismutates to hydrogen peroxide without undergoing any other significant reactions. We conclude that when horseradish peroxidase and its phenolic substrate are present during respiratory burst activity, hydrogen peroxide production is underestimated because superoxide interacts with the assay system. The accumulation of hydrogen peroxide by stimulated neutrophils was also monitored with a hydrogen peroxide electrode. This electrode has the advantage that it consumes very little hydrogen peroxide and does not perturb the reaction system. 15 Using the electrode, we found that superoxide dismutase had no effect on the accumulation of hydrogen peroxide, and that all the superoxide generated was converted to hydrogen peroxide (Fig. 1B). This result confirms that superoxide interferes with the horseradish peroxidase assay. Our results are best explained by superoxide reducing phenoxyl radicals (RO') back to the parent phenol (ROH) (reaction 2). Phenoxyl radicals are produced in the peroxidase cycle (reactions 3 - 5 ) and normally react to form stable end products (reaction 6)? 7 However, if they reacted with superoxide generated by neutrophils, hydrogen peroxide would be consumed by horseradish peroxidase without nett oxidation of the phenol. Superoxide dismutase would prevent this and allow the true amount of hydrogen peroxide to be measured. In support of this explanation, the reaction of superoxide with tyrosyl radicals is fast (k2 1.7 x 109 M -t s -t) and inhibits formation of dityrosine, ts RO" + O2 + H + --' ROH + 02
(2)
HRP 3+ + H202 --' Compound I
(3)
Compound I + ROH--, Compound II + RO" + H +
(4)
Compound II + ROH --* HRP 3+ + RO" + H + (5) RO" + RO" --, Products HRP 3+ + 0 2 -* HRp2+o2 (Compound 111)
(6) (7)
163
Superoxide could inhibit horseradish Im~oxidase by forming compound HI (reaction 7). t7 However, this is unlikely because addition of twice the concentration of horseradish peroxidase had no effect on the hydrogen peroxide measured (data not shown). It could also react with compounds I and II and compete with phenol as an alternative peroxidase substrate. This possibility is also unlikely because reaction of superoxide with compound II is presumed to be slow. t9 Even if these reactions of superoxide with horseradish peroxidase were significant, the effect of superoxide dismutase would still be attributed to it preventing the artefactual interaction of superoxide with the detector, rather than distotting hydrogen peroxide generation by neutrophils. We conclude that assays that use horseradish peroxidase and a phenolic substrate underestimate hydrogen peroxide production by nentrophils because superoxide reacts with phenoxyl radicals generated during the assay. It is, therefore, necessary to include superoxide dismutase with cells to detect all the hydrogen peroxide they produce. Previous studies that have shown that superoxide dismutase, or its mimics, increases production of hydrogen peroxide by neutrophils must, therefore, be interpreted with caution. For example, the finding that Mn 2+, which displays superoxide dismutase activity, increases the amount of hydrogen peroxide measured in the scopoletin assay, should not be taken to indicate that Mn 2+ enhances hydrogen peroxide production by cells. 6 Acknowledgement--We thank the HealthResearchCouncilof New
Zealand for financialsupport. REFERENCES 1. IOebanoff,S. J. Phagocyticcells: Productsof oxygenmetabolism. In: Gallin, J. I.; Goldstein, I. M.; Snyderman~R., eds. Inflammafon: Basic principles and clinical correlates. New York: RavenPress; 1988:391-443. 2. Tauber, A. I.; Babior, B. M. Neutrophiloxygenreduction:The enzymesand the products.Adv. Free Radic. Biol. Me~ 1:265307; 1985. 3. Pipe, R. K. Generationof reactive oxygen metabolitesby the haemocytesof the musselMytilus edulis. Dev. Comp. Immunol. 16:111-122; 1992. 4. Rook R. K.; Metcalf,J. A. H202release from human granulocytesduringphagocytosis.Ralmionshipto superoxideanionformarion and cellular catabolismof H202: Studies with normal and cytochalasinB-treatedcells. J. Clin. Invest. 60:1266-1279; 1977. 5. Baggiolini,M.; Ruch,W.; Cooper,P. H. Measurementof hydrogen peroxideproductionby p h a g ~ usinghomovanilllcacid and horsera~sh peroxidese. Methods EnzymoL 132:395-400; 1986. 6. Klebanoff,S. J.; Schlechte,IL G.; Waltersck3q~A. M. Stimulation of the bactericidalactivityof p o l ~ u c l e m " lenkocytes by manganese. J. Le~oc. Biol. $3:666-672; 1993. 7. Winterbonrn, C. C.; Garcia, R.; Segal, A. W. Production of the superoxide adduct of myeloperoxidese(compoundHI) by stimulatedneutmphils,and its reactivitywithH202and chloride. B/ochent J. 7,28:583-592; 1985. 8. Kettle, A. J.; Winterbourn, C. C. Superoxide modulates the
164
9. 10.
11.
12.
13.
A. J. I~rr,.~ et aL
aclivity of myelcpemxidase and optimizes the production of hypochlorous ~ Biochem. J. 252:529-536; 1988. Cohen, M. S.; Britigun, B. E.; Hassott, D. J.; Rosen, G. M. Do human I~uB~dls form hydroxyl radical? Evaluation of an unmmlved controversy. Free Radic. Biol. MecL $:81-88; 1988. Ramos, C. L.; Pun, S.; Britigan, B. E.; Cohen, M. S.; Rosen, G. M. Spin trappin~ evidence for myelopemxidase-dependent hydroxyl radical formation by human ncotrophils and monocytes. J. BioL Chem. 267:8307-8312; 1992. Boyum, A. Isolation of mononuclear cells and granulucytes from human blood, hoLsqiun of mononuclear cells by one centrifugation; and of granulocytes by combined centrifugetion and sedimentation at 1 g. Scan~ Y. Ciir~ Lab. Invest. 21(Suppl 97):7789; 1968. Root, R. IL; Metcalf, J.; Oshino, N.; Chance, B. H202 release from hnmsn granttiocytes during phagocytosis I. Documentation, quantitaliun, and some regulating factors. J. Clin. Invest. $$:945-955; 1975. Smith, R. J.; Bowman, B. J. Generation of hydrogen peroxide by human neulrophils. Effects of soluble stimuli and requiremerits for divalent cations. Clin. lmmunol. ImmunopatheL 24:194-203; 1982.
14. Beers, R. J.; Sizer, I. W. A spectrophotometric method for measuring the breakdown of hydrogen peroxide by catalase. J. Biol. Chem. 195:133-140; 1952. 15. Test, S. T.; Weiss, S. J. Quantitative and temporal characterization of the extracellular hydrogen peroxide pool generated by human neutrophils. J. Biol. Chem. 259:399-405; 1984. 16. Fridovich, I. Cytochrome c. In: Greenwald, R. A., ed. Handbook of methods for oxygen radical research. Boca Raton: CRC Press, 1985:213-215. 17. Dunford, H. B. Horseradish peroxidase: Structure and kinetic properties. In: Everse, J.; Everse, K. E.; Grisham, M. B., eds. Peroxidases in chemistry and biology. Boca Raton: CRC Press, 1990:1-24. 18. Cudina, I.; Josimovic, Lj. The effect of oxygen on the radiolysis of tyrosine in aqueous solution. Radiat. Res. 109:206-215; 1987. 19. Bielski, B, H. J.; Gebicki, J. M. Study of peroxidase mechanisms by pulse radiolysis. HI. The rate of reaction of O~ and HO2 radicals with horseradish peroxidase compound I. Biochim. BIOphys. Acta 364:233-235; 1974.
Free RadicalBiology& Medicine,VoL 17, No. 2, pp. 161-164, 1994 Copyright© 1994Blsevia ~imce I ~ Prin~d in tlw USA.All flglmrmerved 0 8 9 1 o $ ~ $6.00 + .00
rergemn
0~l-.¢,~904)E00ZT-6
"~ BriefCommunication ASSAYS USING HORSERADISH PEROXIDASE AND PHENOLIC SUBSTRATES REQUIRE SUPEROXIDE DISMUTASE FOR ACCURATE DETERMINATION OF HYDROGEN PEROXIDE PRODUCTION BY NEUTROPHILS ANTHONY J. KEITLE, ANITRA C. CARR, and CHRISTINE C. WINTERBOURN Free Radical Research Unit, Department of Pathology, Christchurch School of Medicine, Christchurch, New Zealand
(Received 24 November 1993; Accepted 7 January 1994) Abstract--We used honeradish peroxidase and either scopoletin, homovanilllc acid, or phenol red to measure hydrogen peroxide ganentted by human neutrophils.With these assays, superoxidedismutasesj£nificantlyincreasedthe anmunt of bydrogan peroxide detected. In contrast, it had no effect when the accumul~on of hydrogen peroxide was measored with a hydmgan peroxide electrode. We propose that superoxideinterferes with horseradishp e m ~ t assays so that hydrogenperoxide is n ~ s t i m a t e d , Thus, when using these assays, superoxide dismutase must he added to neutmphils to ensure that all the hydrogen peroxide they produce is detected. Keyworde---Neutrophils, Hydrogen peroxide, Horseradish peroxidase, Free radicals
generated by neutmphils.2 Also, when measuring hydrogen peroxide, azide is added to inhibit myelopemxidase. Hence, other explanations are required to account for the effect of superoxide dismutase on the yield of hydrogen peroxide. We have measured hydrogen peroxide production by neutrophils using either horseradish peroxidase with phenolic substrates, or a hydrogen peroxide electrode. Hydrogen peroxide was detected in the presence and absence of superoxide dismutase, and was related to superoxide production. We show that when measured with a hydrogen peroxide electrode the amount of hydrogen peroxide produced by neutrophils is unaffected by superoxide dismutase, and it accounts for all the superoxide generated. We propose that superoxide interferes with horseradish peroxidase-dependent assays so that superoxide dismutase must he added to cells to ensure that all the hydrogen peroxide they produce is detected.
INTRODUCTION
Stimulated neutrophils generate superoxide which dismutates to hydrogen peroxideJ a The most commonly used methods for measuring hydrogen peroxide production by neutrophils employ horseradish peroxidase in combination with a phenofic substrate. 3-6 Hydrogen peroxide is detected because, in conjunction with horseradish peroxidase, it oxidizes the phenol to cause a change in fluorescence or absorbance. Superoxide dismutase is not normally added to these assays, but when included, it has been shown to increase the production of hydrogen peroxideJ -6 The usual explanation given for this increase is that not all the superoxide discharged by neum~phils spontaneously dismutates. It is presumed that some of it is involved in oxidation or reduction reactions, and that the addition of superoxide dismutase diverts superoxide from these reactions, leading to an artefactual increase in hydrogen peroxide production. Proposed reactions of superoxide, other than dismutation, include the reduction of redox intermediates of myeloperoxidaseTM and the formation of hydroxyl radicals. 9a° However, ff produced, hydroxyl radicals account for less than 1% of the superoxide
MATERIALS AND METHODS
Neutrophils were isolated from the blood of healthy donors by Ficoll-Hypaque centrifugefion, dextran sedimentation, and hypotonic lysis of contaminating red cells, n Horseradish peroxidase (Type I), bovine erythrocyte superoxide dismutase, bovine liver catalase,
Address c c ~ l a ~ to: Anthony J. Keltic, Free Radical Research Unit, Depmlmentof Pathology, ChristchurchSchool of Medicine, Christchurch, New Zealand. 161
162
A.J. KeTn~ et al.
phorbol myristate acetate, and cytochrome c (Type HI) were purchased from the Sigma Chemical Co. (St Louis, MO).
Hydrogen peroxide assays Scopoletin assay. Horseradish peroxidase (20 nM) and scopoletin (40/~M) were added to 105 cells in 10 mM phosphate buffer pH 7.4 containing 140 mM NaC1, 1 mM CaCI2, 0.5 mM MgC12, 1 mg ml -i of glucose, and 1 mM azide. Neutrophils were preincubated for 10 rain at 37°C and then stimulated with phorbol myristate acetate (100 ng ml-~). The final reaction volume was 1 ml. After 30 min, the reaction was stopped by placing mhes in melting ice and pelleting the cells by centrifugation. The hydrogen peroxide in the supernatant was determined by measuring the loss in fluorescence due to the oxidation of scopoletin (ex 350 nm; em 460 um).n Alternatively, hydrogen peroxide was allowed to accumulate for 30 rain before cells were pelleted, then horseradish peroxidase and scopoletin were added to the supernatant. HomovaniUic acid assay. Conditions were the same as in Scopoletin assay, except the concentration of homovanillic acid was 100/dvl. After 30 min of stimulation reactions were stopped by adding 120 #1 of 100 raM glycine-NaOH buffer pH 12 containing 25 mM EDTA and cells were pelleted. The hydrogen peroxide produced was dete~jned by measuring the increase in fluorescence in the supemat_ant due to the oxidation of homovanillic acid (ex 312 nm, em 420 nm). 5 Phenol red assay. Conditions were the same as in Scopoletin assay, except the concentratiom of horseradish peroxidase, cells, and phenol red were 170 nM, 5 x 105 ml -t, and 280 ~M, respectively. After the reactions were stopped, cells were pelleted by centrifugafion and 10 #1 of 1 M NaOH was added to 1 ml of supernatant. The hydrogen peroxide produced was determined by measuring the increase in A~m due to the oxidation of phenol red. 13
Each of the horseradish peroxidase-dependent assays was calibrated with known concentrations of hydrogen peroxide (e~o 43.6 M -1 c m - t ) J 4
Hydrogen peroxide electrode. Hydrogen peroxide production was monitored continuously with a YSI model 25 oxidase meter fitted with a YSI 2510 oxidase probe (Yellow Springs Instntment Co., Yellow Springs, O H ) as desorihed previously.~5Briefly, 106 neutrophils m1-1 were stimulated with phorbol mynstate acetate in the presence of 1 raM azide. Other conditions were as described in Scopoletiu assay, except for the omission
A
B
0.7 0.6
scopot~
Phe~ned
0.4 0.3 0.2
0.1 0.0 Fig. 1. Effects of supemxide dismutase on the ratio of hydrogen peroxide to superoxide produced by neulrophils. Cells were stimulated with phorbol myristate acetate in the presence (D) or absence (1) of 50 #g m1-1 of supe~oxide dismutase. (A) Hydrogen peroxide was determined by adding horseradish peroxidase and either scopoletin, homovanillic acid, or phenol red before stimulating the cells (n = 6). (B) Hydrogen peroxide was also measured by allowing it to accumulate for 30 rain before adding ho~_A~_h peroxidase and scopoletin to supetnatan~, and by using a hydrogen peroxide electrode (n = 3). Superoxide production, measured as superoxide dismutase-inhibitable cytochrome c reduction, was 18.0 -+ 2.0 nmol per 105 cells in 30 rain (n = 6). Reactions were cmxiad out in duplicate and dAt~ are represented as the ratio of the means of average values for cells from n individuals _+ the standard error of the mean. Conditions are given in the Materials and Methods section.
of horseradish peroxidase and scopoletin. Hydrogen peroxide accumulation was monitored for 20 rain. Superoxide production was measured as superoxide-dismutase-inhibitable cytochrome c reduction for 30 rain under the same conditions as described for the scopoletin assay. ~ Fluorescence was measured with an Aminco Bowman spectrophotofluorometer, and absorbances with a CarylE spectrophotometer. RESULTS AND DISCUSSION
Hydrogen peroxide production was measured by adding horseradish peroxidase and either scopoletin, homovanillic acid, or phenol red to neutrophils, and relating it to the amount of superoxide they generated. If all the superoxide spontaneously dismutated to hydrogen peroxide according to reaction (1), the ratio of hydrogen peroxide produced to superoxide produced would be 0.5. 0 2 + HO2 + H + --* H202 + 0 2
(1)
When cells were stimulated after addition of horseradish peroxidase and one of the phenolic subsidies, this ratio was significantly less than 0.5 (Fig. 1A). For scopoletin, only half the superoxide reacted to form hydrogen peroxide. Our results indicate that superoxide must have reacted in other ways besides spontane-
Assays for hydrogenperoxide productionby neutmphils ously dismutaflng. In agreement with previous studies, 3-6 superoxide dismutase increased the yield of hydrogen peroxide, so that the ratio of hydrogen peroxide to superoxide was close to theoretical. Because the amount of hydrogen peroxide detected varied with the phenolic substrate used, we tested the possibility that superoxide was reacting with components of the horseradish peroxidase assay. Cells were stimulated and hydrogen peroxide was allowed to accumulate for 30 rain before adding horseradish peroxidase and scopoletin to the supernatant. Using this protocol, superoxide dismutase had no effect on the hydrogen peroxide production, and all the superoxide was accounted for by the hydrogen peroxide detected (Fig. 1B). Therefore, superoxide discharged by neutrophils dismutates to hydrogen peroxide without undergoing any other significant reactions. We conclude that when horseradish peroxidase and its phenolic substrate are present during respiratory burst activity, hydrogen peroxide production is underestimated because superoxide interacts with the assay system. The accumulation of hydrogen peroxide by stimulated neutrophils was also monitored with a hydrogen peroxide electrode. This electrode has the advantage that it consumes very little hydrogen peroxide and does not perturb the reaction system. 15 Using the electrode, we found that superoxide dismutase had no effect on the accumulation of hydrogen peroxide, and that all the superoxide generated was converted to hydrogen peroxide (Fig. 1B). This result confirms that superoxide interferes with the horseradish peroxidase assay. Our results are best explained by superoxide reducing phenoxyl radicals (RO') back to the parent phenol (ROH) (reaction 2). Phenoxyl radicals are produced in the peroxidase cycle (reactions 3 - 5 ) and normally react to form stable end products (reaction 6)? 7 However, if they reacted with superoxide generated by neutrophils, hydrogen peroxide would be consumed by horseradish peroxidase without nett oxidation of the phenol. Superoxide dismutase would prevent this and allow the true amount of hydrogen peroxide to be measured. In support of this explanation, the reaction of superoxide with tyrosyl radicals is fast (k2 1.7 x 109 M -t s -t) and inhibits formation of dityrosine, ts RO" + O2 + H + --' ROH + 02
(2)
HRP 3+ + H202 --' Compound I
(3)
Compound I + ROH--, Compound II + RO" + H +
(4)
Compound II + ROH --* HRP 3+ + RO" + H + (5) RO" + RO" --, Products HRP 3+ + 0 2 -* HRp2+o2 (Compound 111)
(6) (7)
163
Superoxide could inhibit horseradish Im~oxidase by forming compound HI (reaction 7). t7 However, this is unlikely because addition of twice the concentration of horseradish peroxidase had no effect on the hydrogen peroxide measured (data not shown). It could also react with compounds I and II and compete with phenol as an alternative peroxidase substrate. This possibility is also unlikely because reaction of superoxide with compound II is presumed to be slow. t9 Even if these reactions of superoxide with horseradish peroxidase were significant, the effect of superoxide dismutase would still be attributed to it preventing the artefactual interaction of superoxide with the detector, rather than distotting hydrogen peroxide generation by neutrophils. We conclude that assays that use horseradish peroxidase and a phenolic substrate underestimate hydrogen peroxide production by nentrophils because superoxide reacts with phenoxyl radicals generated during the assay. It is, therefore, necessary to include superoxide dismutase with cells to detect all the hydrogen peroxide they produce. Previous studies that have shown that superoxide dismutase, or its mimics, increases production of hydrogen peroxide by neutrophils must, therefore, be interpreted with caution. For example, the finding that Mn 2+, which displays superoxide dismutase activity, increases the amount of hydrogen peroxide measured in the scopoletin assay, should not be taken to indicate that Mn 2+ enhances hydrogen peroxide production by cells. 6 Acknowledgement--We thank the HealthResearchCouncilof New
Zealand for financialsupport. REFERENCES 1. IOebanoff,S. J. Phagocyticcells: Productsof oxygenmetabolism. In: Gallin, J. I.; Goldstein, I. M.; Snyderman~R., eds. Inflammafon: Basic principles and clinical correlates. New York: RavenPress; 1988:391-443. 2. Tauber, A. I.; Babior, B. M. Neutrophiloxygenreduction:The enzymesand the products.Adv. Free Radic. Biol. Me~ 1:265307; 1985. 3. Pipe, R. K. Generationof reactive oxygen metabolitesby the haemocytesof the musselMytilus edulis. Dev. Comp. Immunol. 16:111-122; 1992. 4. Rook R. K.; Metcalf,J. A. H202release from human granulocytesduringphagocytosis.Ralmionshipto superoxideanionformarion and cellular catabolismof H202: Studies with normal and cytochalasinB-treatedcells. J. Clin. Invest. 60:1266-1279; 1977. 5. Baggiolini,M.; Ruch,W.; Cooper,P. H. Measurementof hydrogen peroxideproductionby p h a g ~ usinghomovanilllcacid and horsera~sh peroxidese. Methods EnzymoL 132:395-400; 1986. 6. Klebanoff,S. J.; Schlechte,IL G.; Waltersck3q~A. M. Stimulation of the bactericidalactivityof p o l ~ u c l e m " lenkocytes by manganese. J. Le~oc. Biol. $3:666-672; 1993. 7. Winterbonrn, C. C.; Garcia, R.; Segal, A. W. Production of the superoxide adduct of myeloperoxidese(compoundHI) by stimulatedneutmphils,and its reactivitywithH202and chloride. B/ochent J. 7,28:583-592; 1985. 8. Kettle, A. J.; Winterbourn, C. C. Superoxide modulates the
164
9. 10.
11.
12.
13.
A. J. I~rr,.~ et aL
aclivity of myelcpemxidase and optimizes the production of hypochlorous ~ Biochem. J. 252:529-536; 1988. Cohen, M. S.; Britigun, B. E.; Hassott, D. J.; Rosen, G. M. Do human I~uB~dls form hydroxyl radical? Evaluation of an unmmlved controversy. Free Radic. Biol. MecL $:81-88; 1988. Ramos, C. L.; Pun, S.; Britigan, B. E.; Cohen, M. S.; Rosen, G. M. Spin trappin~ evidence for myelopemxidase-dependent hydroxyl radical formation by human ncotrophils and monocytes. J. BioL Chem. 267:8307-8312; 1992. Boyum, A. Isolation of mononuclear cells and granulucytes from human blood, hoLsqiun of mononuclear cells by one centrifugation; and of granulocytes by combined centrifugetion and sedimentation at 1 g. Scan~ Y. Ciir~ Lab. Invest. 21(Suppl 97):7789; 1968. Root, R. IL; Metcalf, J.; Oshino, N.; Chance, B. H202 release from hnmsn granttiocytes during phagocytosis I. Documentation, quantitaliun, and some regulating factors. J. Clin. Invest. $$:945-955; 1975. Smith, R. J.; Bowman, B. J. Generation of hydrogen peroxide by human neulrophils. Effects of soluble stimuli and requiremerits for divalent cations. Clin. lmmunol. ImmunopatheL 24:194-203; 1982.
14. Beers, R. J.; Sizer, I. W. A spectrophotometric method for measuring the breakdown of hydrogen peroxide by catalase. J. Biol. Chem. 195:133-140; 1952. 15. Test, S. T.; Weiss, S. J. Quantitative and temporal characterization of the extracellular hydrogen peroxide pool generated by human neutrophils. J. Biol. Chem. 259:399-405; 1984. 16. Fridovich, I. Cytochrome c. In: Greenwald, R. A., ed. Handbook of methods for oxygen radical research. Boca Raton: CRC Press, 1985:213-215. 17. Dunford, H. B. Horseradish peroxidase: Structure and kinetic properties. In: Everse, J.; Everse, K. E.; Grisham, M. B., eds. Peroxidases in chemistry and biology. Boca Raton: CRC Press, 1990:1-24. 18. Cudina, I.; Josimovic, Lj. The effect of oxygen on the radiolysis of tyrosine in aqueous solution. Radiat. Res. 109:206-215; 1987. 19. Bielski, B, H. J.; Gebicki, J. M. Study of peroxidase mechanisms by pulse radiolysis. HI. The rate of reaction of O~ and HO2 radicals with horseradish peroxidase compound I. Biochim. BIOphys. Acta 364:233-235; 1974.