Chem -Bwl. Interactwns, 77 (1991) 303--314 Elsevier Scientific Publishers Ireland Ltd.

303

EVALUATION OF THE ABILITY OF THE ANGIOTENSIN-CONVERTING ENZYME INHIBITOR CAPTOPRIL TO SCAVENGE REACTIVE OXYGEN SPECIES

OKEZIE I. ARUOMAa, DOLA AKANMU~, RUBENS CECCHINIa,* and BARRY HALLIWELLb

aDepartment of Biochemistry, University of London King's College, Strand Campus, London WC~R ~LS (U.K.) and aDivision of Pulmonary-Cmtical Care Medw~ne, UC Davis Medical Center, Professional Building, 4301 X St., Sacramento, CA 95817 (U.S.A.) (Received September 27th, 1990) (Revision received October 29th, 1990) (Accepted November 1st, 1990)

SUMMARY

Captopril, an inhibitor of angiotensin-converting enzyme, has been suggested to have additional cardioprotective action because of its ability to act as an antioxidant. The rates of reaction of captopril with several biologically-relevant reactive oxygen species were determined. Captopril reacts slowly, if at all, with superoxide (rate constant < 108 M -1 s -1) or hydrogen peroxide (rate constant < 1 M-~ s-l). It does not inhibit peroxidation of lipids stimulated by iron ions and ascorbate or by the myoglobin]H202 system. Indeed, mixtures of ferric ion and captopril can stimulate lipid peroxidation. Captopril reacts rapidly with hydroxyl radical (rate constant > 109 M -1 s -1) but might be unlikely to compete with most biological molecules for "OH because of the low concentration of captopril that can be achieved in vivo during therapeutic use. Captopril did not significantly inhibit iron ion-dependent generation of hydroxyl radicals from hydrogen peroxide. By contrast, captopril is a powerful scavenger of hypochlorous acid: it was able to protect %-antiproteinase (%AP) against inactivation by this species and to prevent formation of chloramines from taurine. We suggest that the antioxidant action of captopril in vivo is likely to be limited, and may be restricted to protection against damage by hypochlorous acid derived from the action of neutrophil myeloperoxidase.

Key words: Captopril -- Antioxidant -- Reperfusion injury -- Hydroxyl radical -- Superoxide -- Hypochlorous acid Correspondence to: O.I. Aruoma, Department of Biochemistry, University of London King's College, Strand Campus, London WC2R 2LS, U.K. *Present address: Dept of General Pathology, University of Londrina, Londrina PR, Brazil. 0009-2797/91/$03.50 © 1991 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland

304 INTRODUCTION Various scavengers of reactive oxygen species have been shown to be protective against reoxygenation injury after ischaemia of the myocardium (reviewed in Refs. 1--4). Agents that have been shown to be effective in some systems include superoxide dismutase, catalase, iron ion chelators, scavengers of hydroxyl radical ('OH) and thiol compounds such as N-acetylcysteine and mercaptopropionylglycine [1--6]. It has therefore been suggested that re-oxygenation of ischaemic myocardium leads to generation of O: and H202 which can, in the presence of transition metal ions, become converted into highly-reactive "OH [1[. Potential sources of iron to catalyze "OH formation include the release of iron ions from injured cells and the ability of H20e in excess to degrade myoglobin with the release of iron ions [7,8]. In addition, the initial reaction of myoglobin with H202 leads to the formation of a powerful oxidant that can accelerate peroxidation of membrane lipids [9] and it has been suggested that myoglobinH202 interactions are an important component of reoxygenation injury [9]. This powerful oxidant does not appear to be "OH [10,11] and it may be a tyrosine peroxyl radical [12]. Yet another reactive oxygen species that can damage the myocardium is hypochlorous acid (HOC1), generated by the enzyme myeloperoxidase in activated neutrophils infiltrating the reoxygenated tissue [2]. Captopril, an inhibitor of angiotensin-converting enzyme, has beneficial effects on the reoxygenated myocardium and it has been suggested that some of these are mediated by free radical scavenging involving its thiol group [13--15]. Thus Das et al. [13] reported that captopril could scavenge 0~-, "OH and HOC1 in vitro. However, precise rates for these reactions were not determined: such information is essential in evaluating the possibility that captopril might act as a free radical scavenger in vivo (discussed in Ref. 16). The suggestion of 02scavenging by captopril has been challenged by others [17]. In addition, thiols can have pro-oxidant effects under certain circumstances [18--20]. Thus their reaction with oxygen radicals can produce sulphur-containing radicals that can damage biological molecules, including an acceleration of lipid peroxidation

[18-211. In the present paper, we have used established methods (reviewed in Ref. 16) to study the reactions of captopril with those reactive oxygen species that can be formed in the reoxygenated myocardium, viz. 02-, H202, HOC1 and "OH. In addition, the effect of captopril on "OH generation in the presence of iron ions and H202 and upon lipid peroxidation stimulated by the myoglobin/H202 system and by iron ions has been investigated. The results obtained allow us to evaluate the likelihood that captopril will act as an antioxidant in vivo. MATERIALSAND METHODS

Reagents Captopril was generously donated by Squibb, New Jersey. Other reagents, including myoglobin (horse-heart), xanthine oxidase (EDTA-free), superoxide dismutase (bovine copper-zinc containing enzyme) and %-antiproteinase (type

305 A9024) were from Sigma, except for HOC1 and pig pancreatic elastase (BDH Chemicals Ltd).

Assays Elastase and %-antiproteinase were assayed essentially as described in Ref. 21: full details are given in the legend to Table I. HOC1 was produced immediately before use by adjusting NaOC1 to pH 6.2 with dilute H2SO4 [22]. Generation of 02 by the hypoxanthine-xanthine oxidase system was carried out essentially as described in Aruoma et al. [23]. Reaction mixtures contained, in a final volume of 3 ml, 0.1 ml 30 mM EDTA, 10 ~l 30 mM hypoxanthine in 50 mM KOH, 100 ~l of 3 mM cytochrome c or 3 mM nitro-blue tetrazolium, and 50 mM (final concentration) KH2PO(KOH buffer (pH 7.4). Reaction was started by adding 0.2 ml of xanthine oxidase (freshly diluted in the above phosphate buffer to give one unit of enzyme activity per ml) and the rate of NBT or cytochrome c reduction measured at 560 or 550 nm, respectively, at 25°C. In early experiments, H202 was measured by a peroxidase-based assay system [23], but captopril was found to interfere. Hence loss of the --SH group was measured by reaction with DTNB, as described in Aruoma et al. [23]. Rat liver microsomes were prepared by differential pelleting and their peroxidation in the presence of FeC13 and ascorbate measured by the thiobarbituric acid (TBA) test as described in Cecchini et al. [24] except that the TBA reagents also contained 0.02% butylated hydroxytoluene to suppress peroxidation during the test itself [25]. Peroxidation of arachidonic acid by the myoglobin/H202 system was studied in reaction mixtures containing the following reagents at the final concentrations stated: 25 mM KH2PO4-KOH buffer (pH 7.4), 0.4 mM arachidonic acid, 50 ~M myoglobin, 0.5 mM H202 and 100 mM diethylenetriaminepentaacetic acid to bind any released metal ions. Tubes were incubated at 30°C for 10 min. Hydroxyl radical formation was measured in the presence of ascorbate, FeC13 (±EDTA) and H202, by the deoxyribose assay [26]. RESULTS

Does captopril scavenge superoxide? A mixture of hypoxanthine and xanthine oxidase at pH 7.4 generates 02, which can be detected by its ability to reduce ferricytochrome c or nitro-blue tetrazolium, NBT [27,28]. Any added compound that is itself able to react with O~ should decrease the rate of reduction of these substances. Captopril, tested at concentrations up to 3 mM (which did not themselves reduce cytochrome c or NBT under our reaction conditions), had no significant effect on the rate of reduction of 100 t~M ferricytochrome c, whereas superoxide dismutase inhibited its reduction by > 90%. Captopril (5 mM) inhibited reduction of 100 t~M NBT by < 10%, although NBT reduction was inhibited > 90% by superoxide dismutase. At pH 7.4 cytochrome c reacts with 02 with a rate constant of about 2.6 x 105 M -1 s -1, whereas NBT reacts with a rate constant of about 6 x 104 M -1 s -1 [29,30]. The inability of captopril, at a concentration fifty times greater than that

306

R

"i I lq

•

-@

10

1'5

Time (rain) Fig. 1. Reaction of captopril with H20.~. Reaction mixtures contained, in a final volume of 5 ml, 1.5 ml of 100 mM KH2PO4-KOH (pH 7.4), H202 (10.6 mM) and captopml at the concentrations stated (1 or 2 raM). At intervals, 0.25-ml aliquots of reaction mixture were added to 2.5 ml of buffer and 0.25 ml of 10 mM DTNB. After standing for 10 min, the A41z was measured. The X-axm shows the captopril concentration m the reaction mixture.

of NBT or cytochrome c, to inhibit their reduction by Of suggests that the reaction of captopril with O[ (if any) cannot proceed with a rate constant much above l0 s M -1 s-i.

Reaction of captopril with hydrogen peroxide Initial attempts to use a peroxidase-based assay method [23] to measure reaction of captopril with H202 showed that captopril interfered with the assay, perhaps by acting as a substrate for horseradish peroxidase like some other thiols [31]. Its reaction with H202 was therefore measured as loss of the --SH group, using 5,5 '-dithiobis-2-nitrobenzoic acid [32], as described in Aruoma et al. [23]. Measurement of A412 after reaction of captopril with DTNB gave a linear calibration plot up to captopril concentrations of 3 mM. Figure 1 shows the rate

307 TABLE I ACTION OF CAPTOPRIL ON THE ABILITY OF HOC1 TO INACTIVATE THE ELASTASEINHIBITORY CAPACITY OF a I ANTIPROTEINASE aIAP (1.2 mg/ml), HOCI (60 #M) and captopml (if present) were incubated in a final volume of 1.0 ml in phosphate-buffered saline at pH 7.4 (full details in Ref. 22) for 20 min at 25°C to allow HOC! to inactivate aI-AP. Then 2 ml of buffer and 0.05 m| of porcine pancreatic elastase solutmn [22] were added, followed by incubation at 25°C for a further 20 min to allow any active a~AP remaining to inhibit elastase (any HOCl remaining is diluted to a point at which it does not affect elastase itself). The uninhibited elastase is then measured by adding elastase substrate, which is hydrolyzed with a rise in A4~0. Concentrations of captopril quoted were those present in the first (1.0 ml) reaction mixture. Captopril was mixed with aIAP and buffer immediately before adding HOCI. Addition to first reaction mixture

Elastase activity in final reaction mixture AA4~0/s

Activity of %AP m inhibiting elastase (100 minus % of elastase activity detected)

Buffer only Buffer, %AP Buffer, alAP, HOCI + captopril (100 ~M) + captopril (200 ~M) + captopril (500 ~M) + captopril (1 mM)

0.050 0.000 0.049 0.037 0.023 0.000 0.000

-100 2 26 54 100 100

of loss of the --SH group when captopril (1 or 2 mM) was incubated with 10.6 mM H~O~ at pH 7.4. Even at this high H202 concentration the rate of reaction was slow: it took over 10 min for half the captopril to be oxidized. An approximate rate constant of 0.07--0.1 M -1 s -1 was calculated from the data presented. Inclusion of diethylenetriaminepentaacetic acid in the reaction mixture (tested up to 1.0 mM final concentration) to bind metal ions did not alter the rate of reaction of captopril with H202.

Scavenging of hypochlorous acid by captopril al-Antiproteinase (%AP), the major circulating inhibitor of serine proteases such as elastase, is inactivated by HOC1 extremely rapidly and loses its ability to inhibit elastase [33]. Although many compounds can react with HOC1, few do so sufficiently rapidly to protect %-AP against inactivation. A good test of whether a compound might be capable of scavenging HOC1 at a biologicallysignificant rate is therefore to examine its ability, at the concentrations present in vivo, to protect %AP against inactivation by HOC1 [16,34]. Table I shows that %AP inhibited elastase: a concentration just sufficient to inhibit completely was used, which is also within the range of concentrations of %AP present in human plasma [22]. Treatment of the %AP with 60 ~M HOC1 almost completely abolished its elastase-inhibitory capacity (Table I, third line). However, if captopril was included in the reaction mixture it protected %AP against inactivation by HOCh Protection was significant at 100 uM captopril and

308

t.

2-

I

i

J

i,

sqv

~o~

I 0

c-

.

l

l

I

•

g

t'~

sqv

I

¢.~

~

4.~

309

a concentration of 500 ~M protected completely. Control experiments showed that captopril did not inhibit elastase directly, nor did it interfere with the ability of %AP to inhibit elastase. It is concluded that captopril is a very good scavenger of HOC1. HOC1 generated in vivo can sometimes react with taurine to yield 'long lived' oxidants such as taurine chloramine [35,36]. When HOC1 was added to a 6.6-mM solution of taurine at pH 7.4, formation of these products was detected by a rise in absorbance at 250--260 nm. Inclusion of captopril in the reaction mixture at a concentration of 3 mM prevented this, suggesting that captopril reacts faster with HOC1 than does taurine (data not shown). However, when captopril was added to pre-formed chloramines, very little loss of these agents was observed (Fig. 2).

Scavenging of hydroxyl radical and inhibition of hydroxyl radical formation by captopril A mixture of FeCIs-EDTA, H202 and ascorbic acid at pH 7.4 generates "OH radicals, which can be detected by their ability to degrade the sugar deoxyribose into fragments that, on heating with thiobarbituric acid at low pH, generate a pink chromogen [26]. Provided that an excess of EDTA is used, any "OH that escapes direct scavenging by EDTA enters 'free solution' and is equally accessible to deoxyribose and to any other scavenger of "OH added [26]. Thus the ability of a substance to inhibit competitively deoxyribose degradation under these reac-

3-

l_. A

2'

Y

J

j

o

i

I

I

3

Fig. 3. Action of eaptepril on hydroxyl radical-dependent degradatmn of deoxyribose. Reaction mixtures contained, in a final volume of 1.2 ml, the following reagents at the final concentration stated: 10 mM Ktt~PO4-KOH buffer (pI-I 7.4), 2.8 mM I-I20z, 2.8 mM deoxyribose, 50 ~M FeCI a (pre-mixed with 100 ~M EDTA before addition to the reaction mixture, where stated). Aseorbate (0.1 raM) was Mded to start the reaction and tubes were incubated at 37°C for 1 h. Products of "OH attack upon deox3~ibose were measured as in ttalliwell et al. [26|. Line A, EDTA present. A rate constant of 3.66 x 10~ M -1 s -1 can be c~eulated from the data using the method described in Halliwell et al. [261. Line B, EDTA absent.

310 tion conditions is a measure of its ability to scavenge "OH and can be used to calculate a rate constant for reaction of "OH with the scavenger [16,26]. Figure 3 shows that captopril was a powerful scavenger of "OH competitive with deoxyribose. A rate constant of (3.6 ± 0.2) × 109 M -1 s -1 was calculated by the method described in Halliwell et al. [26] from the results of three such experiments. Control experiments shown that captopril did not interfere with the measurement of deoxyribose degradation (it had no effect when added with the TBA reagents) or itself react with "OH to give TBA-reactive material (omission of deoxyribose from the reaction mixture completely abolished colour formation). It was calculated that the slow reaction of captopril with H20 ~ (Fig. 1) could not interfere with this experiment because the loss of H~O2 by direct reaction would be too small to alter "OH generation. We cannot rule out the possibility that some of the captopril was oxidized but such a secondary reaction would be revealed by a deviation of the competition plot from linearity,which was not observed. When iron ions are added to the reaction mixture as FeC13 instead of as FeC13-EDTA, some of the ions form a complex with deoxyribose [37]. This complex can be reduced by ascorbate to give Fe z*, which can remain attached to deoxyribose [38] and subsequently react with H20 ~. This reaction appears to give "OH, which immediately attacks the deoxyribose in a site-specific manner [16,39]. The resulting deoxyribose degradation is not inhibited by "OH scavengers at moderate concentrations [39]. The only molecules that can inhibit deoxyribose degradation are those that have iron ion-binding capacity and can withdraw the iron ions from the deoxyribose and render them inactive or poorly active in Fenton reactions [16,39]. Indeed, the ability of a substance to inhibit deoxyribose degradation at pH 7.4 in reaction mixtures containing FeCI3, I-I20~ and ascorbate seems to be a measure of the ability of that substance to diminish iron ion-dependent "OH generation by binding the necessary iron ions [16,37--39]. Figure 3 (line B) shows that captopril at low concentrations had little inhibitory effect on deoxyribose degradation when FeCI3-EDTA was replaced by FeC13. Thus 5 mM captopril inhibited by only about 25%. We conclude that captopril is a good scavenger of "OH but a poor inhibitor of iron ion-dependent "OH generation from H202. The poor inhibitory action again illustrates that the reaction of H202 with captopril did not deplete H202 in the reaction mixture to an extent that significantly affected "OH generation. However, we cannot rule out the possibility that some of the captopril was oxidized by direct reaction with H202.

Action of captopril on lipid peroxidation Peroxidation of lipids mediated by myoglobin/H202 is thought to contribute to myocardial reoxygenation injury [9]. However captopril, tested at final concentrations up to 4 mM, had no significant effect on peroxidation of arachidonic acid by the myoglobirdH202 system. Captopril itself did not induce any peroxidation of arachidonic acid in the absence of either myoglobin or HeO2 from the reaction mixtures. Peroxidation of rat liver microsomes can be induced by adding FeCI 3 and

311 TABLE II ACTION OF CAPTOPRIL ON THE PEROXIDATION OF RAT LIVER MICROSOMES Reaction mixtures contained, in a final volume of 1 ml, 10 mM KH2PO4-KOH buffer (pH 7 4), 1.5 mg of mmrosomal protein, 100 ~M ascorbate and captopril at the final concentration stated. They were incubated at 37°C for 1 h and peroxldatlon measured by the TBA test m the presence of butylated hydroxytoluene, to stop peroxidation during the TBA assay itself. Reaction mixture

Conc. captopnl added (mM)

Amount of peroxidation

Microsomes/FeCl3/ascorbate

0 0.5 10 2.0 3.0

1.61 1.80 1.74 1 93 1.92

Microsomes/FeC13

0 08 1.0 1.5 2.0 30 4.O 5.0 6.0

0 100 0 219 0.118 0.125 0.163 O.384 O.288 0.310 0.337

A532

ascorbate [25,40]. Captopril, tested at concentrations up to 3 mM, had no significant inhibitory effect. When ascorbic acid was omitted from the reaction mixture, it was found that captopril itself could stimulate peroxidation in the presence of FeC18, to a small and variable extent (Table II). DISCUSSION

Captopril has been suggested to exert cardioprotective actions by scavenging reactive oxygen species such as 02, "OH and HOC1 [13-15]. However, we found, in agreement with Kukreja et al. [17], that the reaction of captopril with O~ is very slow (rate constant probably < 10 '~ M-1 s-~ at pH 7.4). The concentrations of captopril in the body fluids of patients treated with this drug are no greater than the micromolar range [41] and so scavenging of 02 is not a feasible mechanism of action in vivo. This conclusion is not unexpected, since, in general, thiols react slowly (if at all) with 02 [23,30]. Similarly, the slow reaction of captopril with H202 (rate constant < 1 M-1 s-1) suggests that removal of H20 z is not a likely mechanism of cardioprotective action in vivo. Captopril reacts very fast with "OH (rate constant > 109 M -1 s -~) but other biomolecules react equally fast and so captopril, present at concentrations in vivo that are much lower than the effective concentrations of most biological targets

312 [41], is unlikely to intercept "OH (this argument is elaborated further in Ref. 16). In any case, reaction of thiols with "OH can sometimes lead to generation of biologically-damaging sulphur-containing radicals [19--21]. No evidence was obtained for an ability of captopril to prevent iron ion-dependent generation of "OH. Captopril also failed to inhibit lipid peroxidation induced by FeC13/ascorbate or by myoglobin/H202. Indeed, it weakly stimulated peroxidation in the presence of FeCla, perhaps by reducing the ferric ions to the ferrous state. Several other thiols can also stimulate microsomal lipid peroxidation in the presence of ferric ions [18]. By contrast, captopril, like most thiols [23,24,34,42], was found to be a powerful scavenger of HOC1. Scavenging of HOC1, as originally suggested by Bagchi et al. [13], could indeed be a feasible mechanism of cardioprotective action by captopril in those ischaemia/reperfusion systems in which neutrophils play an important damaging role [2]. However, captopril does not seem able to remove the 'long-lived oxidants' generated by reaction of taurine with HOC1. ACKNOWLEDGEMENTS

We are grateful to the British Heart Foundation for research support. REFERENCES 1

J.M. McCord, Oxygen-derived free radicals ,n post-ischemm tissue injury, New Engl J. Med., 312 (1985) 159--163. 2 B.R. Lucchesl, S.W Werns and J.C. Fantone, The role of the neutroph,l and free radmals m lschemic myocardial injury, J Mol. Cell Cardlol., 21 (1989) 1241--1251 3 R. Bolh, Oxygen-demved free radmals and post-lschemm myocardial dysfunction ("stunned myocardium"), J. Am. Coll. Cardiol., 12 (1988) 239-249. 4 J.M. Downey, Free radicals and their involvement during long-term myocardial ~schemla and reperfusion, Annu. Rev. Physiol., 52 (1990) 487--504 5 M.B. Forman, D.W Puett, C.U. Cates, D.B. McCroskey, J.K. Beckman, H.L. Greene and R. Virmani, Glutathione redox pathway and reperfumon injury. Effect of N-acetylcysteme on infarct size and ventmcular function, C,rculation, 78 (1988) 202--213 6 R. Bolli, M.O. Jeroudi, B.S. Patel, O.I. Aruoma, B. Halhwell, E.K. Lai and P.B. McCay, Marked reduction of free radical generation and contractile dysfunction by antioxidant therapy begun at the time of reperfusion, Circ. Res, 65 (1989) 607--622. 7 B. Halliwell, Superoxide, iron, vascular endothelium and reperfusion injury, Free Rad. Res. Commun., 5 (1989) 315--318. 8 A. Puppo and B. Halliwell, Formation of hydroxyl radicals in biological systems Does myoglobin stimulate hydroxyl radmal formation from hydrogen peroxide? Free Rad. Res. Commun., 4 (1988) 415--422. 9 D Galans, D. Mira, A Sevanian, E. Cadenas and P. Hochstem, Co-ox~datlonof salicylate and cholesterol during the oxidation of metmyoglobin by H202, Arch. Biochem. Blophys, 262 (1988) 221--231. 10 A. Puppo, R. Cecchini, O.I. Aruoma, R. Bolh and B. Halhwell, Scavenging of hypochlorous acid and of myoglobm-derived oxidants by the cardioprotectlve agent mercaptopropionylglyclne, Free Radical Res. Commun., 10 (1990) 371--381. 11 J. Kanner and S. Harel, Initiation of membranal lipid peroxidation by activated metmyoglobin and methemoglobin, Arch Biochem. Biophys., 237 (1985) 314--321 12 M.J. Davies, Detection of myoglobm-derived radicals on reaction of metmyoglobm with

313

13 14

15 16 17 18 19

20 21

22

23

24

25 26 27 28 29 30 31 32 33

hydrogen peroxide and other peroxidic compounds, Free Rad. Res. Commun., 10 (1990) 361--370. D. Bagchi, R. Prasad and D.K. Das, Direct scavenging of free radicals by captopril, an angiotensin converting enzyme inhibitor, Biochem. Biophys. Res. Commun., 158 (1989) 52-57. D. Bagchi, I. Iyengar, P. Stockwell and D.K. Das, Enhanced prostaglandin production in the ischemic-reperfused myocardium by captopril linked with its free radical scavenging action, Prostagland. Leuk. Essent. Fatty Acids, 38 (1989) 145-150. X.J. Pi and X. Chen, Captopril and ramiprilat protect against free radical injury in isolated working rat hearts, J. Mol. Cell Cardiol., 21 (1989) 1261--1271. B. Halliwell, How to characterize a biological antioxidant, Free Rad. Res. Commun., 9 (1990) 1-32. R.C. Kukreja, H.A. Kontos and M.L. Hess, Captopril and enalaprilat do not scavenge the superoxide anion, Am. J. Cardlol., 65 (1990) 241--271. G.R.M.M. Haenen, N.P.E. Vermeulen, H. Timmerman and A. Bast, Effect of thiols on lipid peroxidation in rat liver microsomes, Chem.-Biol. Interact., 71 (1989) 201--212. C. Schoneich, K.D. Asmus, U. Dillinger and F. Bruchhausen, Thiyl radical attack on polyunsaturated fatty acids: a possible route to lipid peroxidatmn, Biochem. Biophys. Res. Commun., 161 (1989) 113--120. M.D. Sevilla, M. Yan, D. Becker and S. Gillich, E SR investigations of the reactions of radiationproduced thiyl and DNA peroxyl radicals: formation of sulfoxy radical, Free Rad. Res. Cornmum, 6 (1989) 21--24. O.I. Aruoma, B. Halliwell, J. Butler and B.M. Hoey, Apparent inactivation of %-antiproteinase by sulphur-containing radicals derived from penicillamine, Biochem. Pharmacol., 38 (1989) 4353--4357. M. Wasil, B. Halhwell, D.C.S. Hutchison and H. Baum, The antioxidant action of extracellular fluids. Effect of human serum and its protein components on the inactivation of ~l-antiproteinase by hypochlorous acid and by hydrogen peroxide, Biochem. J., 243 (1987) 219--223. O.I. Aruoma, B. Halliwell, B.M. Hoey and J. Butler, The antioxidant action of N-acetylcysteine: its reaction with hydrogen peroxide, hydroxyl radical, superoxide and hypochlorous acid, Free Rad. Biol. Med., 6 (1989) 593--597. R. Cecchini, O.I. Aruoma and B. Halliwell, The action of hydrogen peroxide on the formation of thiobarbituric acid-reactive material from mmrosomes, liposomes or from DNA damaged by bleomycin or phenanthrohne. Artefacts in the thiobarbituric acid test, Free Rad. Res. Commun., 10 (1990) 245--258. S.A. Buege and S.D. Aust, Microsomal lipid peroxidation, Methods Enzymol., 30 (1978) 302--310. B. Halliwell, J.M.C. Gutteridge and O.I. Aruoma, The deoxyribose method: a simple "test tube" assay for determination of rate constants for reactions of hydroxyl radicals, Anal. Biochem., 165 (1987) 215--219. J.M. McCord and I. Fridovich, Superoxide dismutase. An enzymic function for erythrocuprein (hemocuprein), J. Biol. Chem., 244 (1969) 6049--6055. C. Beauchamp and I. Fridovich, Superoxide dismutase. Improved assays and an assay applicable to acrylamide gels, Anal. Biochem., 44 (1971) 276--288. J. Butler, W.H. Koppenol and E. Margoliash, Kinetics and mechanism of the reduction of ferncytochrome c by the superoxide anion, J. Biol. Chem., 257 (1982) 10749--10752. B.H.J. Bielski, D.E. Cabelli and R.I. Arudi, Reactivity of HOz/O~ radicals in aqueous solution, J. Phys. Chem. Ref. Data, 14 (1985) 1041--1100. B. Halliwell and J. De Rycker, Superoxide and peromdase-catalysed reactions. Oxidation of dihydroxyfumarate, NADH and dithiothreitol by horseradish peroxidase, Photochem. Photobiol., 28 (1978) 757--763. G.L. Ellman, Tissue sulphydryl groups, Arch. Biochem. Biophys., 82 (1959) 70--77. R.A. Clark, P.J. Stone, A.E. Hag, J.D. Calore and C. Franzblau, Myeloperoxidase-catalyzed inactivation of a~-protease inhibitor by human neutrophils, J. Biol. Chem., 256 (1981) 3348--3353.

314 34

35 36 37

38 39

40 41

42

M. Wasil, B Halliwell. C P Moorhouse and D.C.S Hutchison, Biologically-sigmficant scavenging of the myeloperoxidase derived oxidant hypochlorous acid by some anti-mflammatory drugs, Biochem. Pharmacol , 36 (1987) 3847-3850 S.J Weiss, Tissue destruction by neutrophils, New Engl J Med , 320 (1989) 365-376 S.J. Weiss, M.D Lampert and ST Test, Long-hved oxidants generated by human neutrophils characterization and bioactivity, Science, 222 (1983) 625-628. 0 I Aruoma, M. Grootveld and B Halhwell, The role of iron m ascorbate-dependent deoxyrrbose degradation. Evidence consistent with a site-specific hydroxyl radical generation caused by iron ions bound to the deoxyribose molecule, J Inorg Biochem , 29 (1987) 289-299 0 I Aruoma, S. Chaudhary, M. Grootveld and B Halliwell, Bmdmg of u-on(H) ions to the pentose sugar 2-deoxyribose, J Inorg Biochem , 35 (1989) 149-155 J M.C Gutteridge, Reactivity of hydroxyl and hydroxyl-hke radical discrimmated by release of thiobarbituric-acid reactive material from deoxyribose, nucleosides and benzoate, Biochem J , 224 (1984) 761-767 E D. Wills, Lipid peroxide formation of microsomes The role of non-haem iron, Biochem J , 113 (1969) 325-332 H.J. Leis, M. Leis and W Welz, Determmation of captopril m human blood by gas chromatography-negative-ion chemical iomzation mass spectrometry with [“O,]captopril as internal standard, J Chromatogr , 529 (1990) 299-308 W R. Matheson, The effect of antiarthritic drugs and related compounds on the human neutrophil myeloperoxidase system, Biochem Biophys Res Commun , 108 (1982) 259-265