- Email: [email protected]
Reactivity of butylated hydroxytoluene
Free Radical Biology & Medicine,Vol. 21, No. 3, pp. 395400, 1996 Copyright © 1996 ElsevierScienceInc. Printed in the USA. All rights reserved 0891-5849/96 $15.00 + .00
PII S0891-5849(96)00050-0
ELSEVIER
-J~" Brief Communication REACTIVITY OF BUTYLATED HYDROXYTOLUENE CHRISTOPHERR. LAMBERT,* HOMER S. BLACK, t and T. GEORGE TRUSCOTT* *Wellman Laboratories of Photomedicine, Department of Dermatology, Harvard Medical School, Massachusetts General Hospital, Boston, MA 02114, USA; *Photobiochemistry laboratory, Veterans Affairs Medical Center and Baylor College of Medicine, Houston, TX 77030, USA; and *Chemistry Department, Keele University, Keele, Staffordshire, ST5 5BG, UK
(Received 12 September 1995; Revised 6 December 1995; Accepted 24 January 1996)
Abstract--Butylated hydroxytoluene (BHT) is a synthetic antioxidant that is widely used as an additive in foodstuffs to prevent spoiling. The physical-chemical properties of BHT and many related phenols have been examined previously although the mechanisms by which it exerts its antioxidant properties are poorly understood. The reactivity of BHT with singlet oxygen [O2(IAg)] and a number of radical species has been examined using the techniques of time resolved luminescence and pulse radiolysis. In benzene solution BHT reacted with O2(IAg) at a bimolecular rate constant of 1.3 × 106 M ~s-l. The one-electron oxidized, phenoxyl type BHT radical was generated using pulse radiolysis and the absorption spectrum showed a maximum at 400 nm. BHT reacts slowly with many radical species and upper limits for the bimolecular rate constant for reaction with several electron transfer processes are presented. The antioxidant role of BHT is discussed in terms of its reactivity, localization, and stability. Keywords--Butylated hydroxytoluene, Pulse radiolysis, Free radicals
sibility that BHT may act as a tumor promoter has been suggested, 6 and in addition, the precise safety limits for the use of this compound have yet to be established] The LDs0 in rats is reported in the Merck index 8 to be 1.6-3.2 g/kg from unpublished Eastman Kodak data. It has been assumed that the mechanism by which BHT exerts its antioxidant activity involves the quenching of reactive oxygen species and lipid soluble radicals. Reactive oxygen species may include singlet oxygen [O2(~Ag)], hydroxyl radicals ('OH), superoxide ( 0 2 " ) , and peroxyl radicals ( R O t ' ) . The reaction with O2(~Ag) has been previously investigated in methanol and benzene solution by Foote and co-workers. 9 These authors found that the rate constant for the reaction of BHT with OAIAg) in methanol was an order of magnitude higher than in benzene, and that in methanol, the fraction of quenching processes that led to a chemical reaction was decreased. This implies that BHT is a better antioxidant against O2(IAg)in polar environments, because not only does it react faster in polar solvents, but also a smaller fraction is destroyed by chemical reaction. In contrast, for the quenching of O2( j Ag) by a-tocopherol the fraction of quenching processes that lead to the formation of a chemical product is increased in more polar solvents. L°
INTRODUCTION
Butylated hydroxytoluene (BHT) is a phenolic antioxidant that has been widely used in the United States as a food additive since the Food and Drug Administration granted approval for its use in 1954. B H T is frequently added to dried cereals, cooking oils, and various animal foods. In some species the compound is not readily excreted, and it has been found at relatively high concentrations in human body fat. t B H T is known to give some protection against the toxicity and carcinogenicity of a number of chemicals and physical agents. B H T not only inhibits UV-carcinogenesis, as shown with animals receiving the compound as a dietary supplement, 2 but also provides protection to other physiological responses evoked by UV light. 3 At the cellular level, B H T inhibits lipid peroxidation and the photooxidation of cholesterol to the putative carcinogen, 5-a,6-a cholesterol-epoxide. 4 B H T also acts as an antiviral agent against lipid containing viruses, probably by changing the membrane fluidity and, hence, membrane function. 5 However, more recently the posAddress correspondence to: C. R. Lambert, Wellman Laboratories of Photomedicine, 50 Blossom Street, Wellman 224, Massachusetts General Hospital, Boston, MA 02114. 395
396
C . R . LAMBERT et al.
The reaction of BHT with peroxyl radicals has been examined by Boozer and Hammond.~l'12 The inhibition of radical chain reactions by B H T proceeds by reaction with alkyl peroxyl radicals rather than alkyl radicals except at low oxygen concentrations. These authors found that one B H T molecule is able to react with two peroxyl radicals to give products that are stable to further reaction. The mechanism is a two-step process involving a reversible reaction of BHT with a peroxyl radical to give a B H T - p e r o x y l radical complex that undergoes an irreversible reaction with a second radical. H'~2 The rate-limiting process in this reaction is the formation of the complex, at least in organic solvents. Richards and Adams ~3 attempted to correlate the antioxidant activity of B H T and a number of related molecules with the bimolecular rate constant for electron transfer to the semioxidized tryptophan radical, in aqueous micellar solution. The bimolecular rate constant for this process was reported as 2.9 × 10 6 M - i s 1. More recently, the activity of a number of antioxidants, including BHT, was compared with that of/3-carotene, both in solution and in a model solid food systemJ 4 It was found that despite large differences in solution chemistry, the antioxidant activities of B H T and/3-carotene were comparable. However, to fully investigate the antioxidant properties of BHT, a more comprehensive investigation of its reactivity towards a wider range of oxidizing species is required. Further, the interpretation of bimolecular rate constants in heterogeneous solution is complicated due to compartmentalization effects of the reactants. The present work was undertaken to investigate the reactivity of B H T towards a wider range of radical species. Although the experiments we report have been carried out in homogeneous solution, it is recognized that, in vivo B H T is located within more heterogeneous environments, which may effectively increase local concentrations and, thus, alter the relative importance of these reactions.
water to 40% acetonitrile (60:40 H20:ACN), which corresponds to a molar ratio of water to acetonitrile of 4.3:1.
Pulse radiolysis The pulse radiolysis facility at the Paterson Institute CRC laboratories has been described previouslyJ s Pulse widths were varied between 20 ns and 100 ns, which gave a maximum dose of 5 krad. Sample cells were 2.5 cm pathlength and the sample was changed after every electron pulse. The radiation chemistry of water has been well characterized and it is relatively easy to generate either purely oxidizing or reducing conditions. Although the radiation chemistry of acetonitrile has been studied previously, 16'17 the radiation chemistry of mixed solvent systems have not been commonly used in pulse radiolysis, and a number of side reactions have to be considered. Oxidizing conditions for pulse radiolysis are obtained routinely, in water, by using nitrous oxide saturated, sodium azide, solutions. The following reaction sequence is complete within a couple of hundred nanoseconds after the pulse. Radiation Pulse
H20
"OH + eaq + H" + O H - + H + eaq + NeO + HeO -" Ne + "OH + O H N3 + "OH --- N~ + O H -
Oxidizing conditions can also be generated using the same reaction sequence in micellar solution. In acetonitrile, the main radical species generated is the acetonitrile radical anion, CH3CN'-, which reacts reversibly with acetonitrile to give the dimeric species (CH3CN)2 ° . In the presence of water, the radical anion may protonate to form CH3C'NH and the dimeric species is not formed. The radical species generated from water radiolysis also react with acetonitrile with appreciable rate constants:
M A T E R I A L S AND M E T H O D S
Materials B H T was obtained from Sigma (Poole, UK) and sublimed under vacuo before use. Water was doubly distilled, the second time from alkaline permanganate. All other materials were of the purest available from Fluka (Gillingham, UK) and used as received. BHT is virtually insoluble in water and only sparingly soluble in micellar solution, but it was possible to carry out some experiments in 2% aqueous Triton X-100. Due to problems of solubility the pulse radiolysis experiments were carried out in a solvent mixture of 60%
CH3CN + H" ~ CH3C'NH (3.6 × 106 M - i s - l ) 17 CH3CN + "OH ~ CH3C'NOH (2.2 × 107 M-~s-~) ~8 CH3CN + e~q ~ CH3CN'- (3.7 × 107 M - i s - l ) 19 Using 60:40 H20:ACN, the concentration of acetonitrile is 7.6 M, which means that the radical adducts will be formed with pseudo first-order rate constants of 2 × 10 7 S-1, 2 × 108 S -1, and 4 × 10 7 S -1, respectively. However, azide reacts with the hydroxyl radical at 1.2 × 101° M - i s -~, which gives a corresponding rate constant for the growth of the azide radical, in the presence
Reactivity of butylated hydroxytoluene
of 0.05 M azide used in these experiments, of 6.0 x 108 s ~. The radical adducts of acetonitrile eventually form stable products such as ammonia, acetaldehyde, and carbon dioxide. Thus, although the solvent system used here is not as well characterized as pure water the environment after the pulse is still predominantly oxidizing, with more than 70% of "OH reacting to give N;. Reducing conditions for pulse radiolysis are obtained routinely, in water, by using nitrous oxide or nitrogen-saturated, sodium formate solutions. HCO2 + "OH --, CO2"- + H 2 0 HCO~ + H"
~
C02
°-
+
H2
Formate reacts with the hydroxyl radical with a rate constant of 3 x 109 M - I s - l , which gives a pseudo firstorder rate constant for the formation of CO2'- of 3 × 108 M-~s ~ in the presence of 0.1 M formate. This means that approximately 50% of the hydroxyl radicals will react to form CO2"- and 50% to form the acetonitrile adduct. The hydrogen atom reacts with formate with a bimolecular rate constant of 2.7 X 108 M - i s -~, which means that about 50% of the H" should react to give CO2"-. In oxygen-saturated formate solutions the superoxide radical is formed. This species is also formed in the radiolysis of acetonitrile, i.e., C 0 2 " - -t- 0 2 -'+ 0 2 " - -1--C 0 2
e~ + 02 ~ 02" CH3CN"
+ 0 2 ~ 0 2"
+
CH3CN
e~ represents the solvated electron in the mixed solvent system. No reaction was observed between BHT and any of the reducing species described above.
397 RESULTS
Generation of the BHT phenoxyl radical Pulse radiolysis was used to generate the semioxidized radical of BHT in the mixed solvent system, 60:40 HzO:ACN using the azide radical as oxidant. The time resolved spectrum of this species is shown in Figure 1. A similar spectrum was also obtained using aqueous 2% TX- 100 as solvent, confirming that under our conditions the initial radiation chemistry still gave mainly oxidizing conditions and there was little interference from other solvent derived radicals. The absorption band around 400 nm is typical of other phenoxyl radicals that have been reported in the literature, z° The bimolecular rate constant for reaction between the azide radical and BHT was estimated to be 4.5 × 10 9 M - i s -j. We also attempted to obtain the spectrum of the radical formed directly from "OH, in the absence of azide. Under these conditions the apparent yield of the phenoxyl radical is much reduced, which is a consequence of "OH reacting with acetonitrile. The spectrum obtained under these conditions is also distorted by a strong absorption around 300 nm observed under the same conditions in the absence of BHT.
Reactivity with known antioxidants and biological radicals The rate of electron transfer from B H T to the semioxidized tryptophan radical has been used as a measure of antioxidant activity by Richards and A d a m s ] 3 In the present work electron transfer was attempted with a number of radical species and these experiments are summarized in Table 1. The results are presented as an upper limit for the bimolecular rate constant because it was not possible to solubilize BHT at sufficient concentrations in 60:40
!
1
Singlet oxygen luminescence Singlet oxygen was generated in benzene solution by 532 nm laser irradiation of the photosensitizer, protoporphyrin IX dimethyl ester (PPIX DME). The 8 m m diameter, 8 ns frequency doubled line of a J. K. Lasers, System 2000 Nd:YAG laser was used as the excitation source and the time resolved near infrared (NIR) luminescence was monitored at 90 ° to the laser using an E.O.S.S. Inc. 5 mm: germanium diode. Samples were air saturated and the bimolecular rate constant for reaction with Oz(~Ag) was determined using a range of B H T concentrations.
_
i t . l ~
a
504gS-
x a
0.004-
-
c5 0.000
I
25O
i
350 W a v e l e n g t h (nm)
_
_
450
Fig. 1. Transient absorption spectrum of BHT (2 × 10 4 M) in N20 saturated solution of 0.05 M sodium azide (dose ~ 45 G).
398
C.R. LAMBERTet al.
Table 1. Upper Limits of the Rate Constants for Electron Transfer Reactions with BHT Oxidized Radical
Electron Donor
BHT BHT BHT Trolox Vitamin C Tryptophan
Donor Concentration
Trolox Vitamin C Cysteine BHT BHT BHT
1× 1X 1× 1 )< 1× 2×
10 4M 10-4M
10-5 M 10 4M 10 4 M 10-aM
10.00
Rate Constant Upper Limit <3 x 106M iS <3 X 106 M-as <2 × 107 M-~s-~ <106M Is I <106 M Is <106M Is I
6.00
2.00 H 2 0 : A C N to observe electron transfer in these experiments. W e were also unable to repeat the e x p e r i m e n t s p e r f o r m e d b y R i c h a r d s and A d a m s , again b e c a u s e insufficient B H T could be solubilized into SDS solution to affect the rate o f d e c a y o f the tryptophan radical.
Reaction with
h~
isc
~
3p,
02
~
I 0.04
I 0.06
The gradient o f this line gives the b i m o l e c u l a r rate constant for the reaction o f B H T with O2(1Ag), which was found to be 1.3 × 106 M is 1.
DISCUSSION
The quenching o f reactive o x y g e n species b y B H T has been studied extensively over the past 20 years b y c h e m i c a l methods. The nature o f the quenching process is k n o w n to vary, d e p e n d e n t on the reactive o x y g e n species and the local environment. The rate constants for reaction o f B H T with O2(~ Ag), p e r o x y l radicals, and h y d r o p e r o x i d e s have been m e a s u r e d p r e v i o u s l y and the products o f these reactions determined. H o w e v e r , a certain p a r a d o x still remains that despite the p o t e n c y of B H T as an antioxidant it reacts efficiently only with
p + O2(1mg)
The concentration o f B H T was varied over a concentration range o f up to 5 × 10 2 M. A plot o f the rate o f d e c a y o f O2(IAg) as a function o f B H T concentration is shown in Fig. 2.
•
/ 2 ROO ° •-
+
-
I
Fig. 2. Observed rate constant (k') vs. concentration plot for the determination of bimolecular rate constant for O2(~Ag) with BHT. The kinetics were monitored using the NIR luminescence.
O2(1mg)
__, x p ,
I 0.02
[BHT] M
It was not p o s s i b l e to solubilize B H T at sufficient concentration in D 2 0 - b a s e d m i c e l l a r solutions to observe any change in the rate o f d e c a y o f O:(1Ag); therefore, this value was d e t e r m i n e d in b e n z e n e solution. Singlet o x y g e n was generated using 532 n m excitation o f P P I X D M E according to the f o l l o w i n g reaction s c h e m e w h e r e P represents P P I X D M E . p
I 0
RO2H
OH
... O2(1~g)
Physical quenching to give ground state oxygen and B H T
Scheme 1. Reactions of BHT with reactive oxygen species.
Reactivity of butylated hydroxytoluene strongly oxidizing species such as the hydroxyl radical. These quenching reactions are summarized in Scheme 1. The scheme depicts the reaction of B H T with, hydroxyl radicals, peroxyl radicals, and O2(IAg). Hydroxyl radicals may react either to give an adduct or by electron transfer to give the phenoxyl radical. The phenoxyl radical is relatively stable but may eventually react with another radical or polymerize. The mechanism of reaction of "OH with phenols was originally investigated by Land and Ebert. 2° The reaction does not proceed by electron transfer directly but rather by "OH addition followed by acid and base-catalyzed elimination to give the phenoxyl radical. The mechanism of reaction of BHT with peroxyl radicals was proposed by Boozer and Hammond and involves the deactivation of two radicals. Reaction with the first radical is a reversible process to give a peroxyl r a d i c a l - B H T complex. This complex may then reduce a second peroxyl radical by hydrogen transfer. The mechanism of reaction of BHT with O2(lAg) involves the formation of an encounter complex. This reaction is reversible but the collapse of the complex may give product formation or the parent ground state molecules. The fraction of quenching processes that lead to product formation is controlled by the balance between the rate constants for product formation and deactivation and, as shown by Foote et al., 9 is solvent dependent. The one electron reduction potential of BHT has not been determined, but it is likely to be around 6 0 0 - 7 0 0 mV. This value is assumed by comparison with other substituted phenols. The reduction potential of phenol itself is reported to be more positive than 800 mV. Methyl substitution at the para position (p-cresol) lowers the potential to 770 mV. Substitution of two tertbutyl groups is likely to reduce the potential still further. The reduction potential of the compounds used in this study are given in Table 2. 21 Examination of Table 2 shows that it is thermodynamically favorable for the B H T radical to be reduced by ascorbate and Trolox and for BHT to reduce the tryptophyl and trichloromethyl peroxyl (CC13OO") radicals. The rate constant determined by Richards and Adams for electron transfer from BHT to the tryptophyl radical was 2.9 × 10 6 M-1s - l . This is bigger than the upper limit determined in the present work but probably reflects the effect of compartmentalization into the micelles. The bimolecular rate constant for reaction of BHT with CC13OO" has been measured in carbon tetrachloride and found to be 6 × 10 6 M - l s - l . 22 This last result clearly demonstrates that even for thermodynamically favorable peroxyl radicals the bimolecular rate constant for electron transfer is still relatively slow. The
399
Table 2. Reduction Potentials of Compounds Relevant to This Study Compound
E (mV)
Ascorbate Trolox Cysteine Tryptophan p-Cresol Trichloromethyl peroxyl
300 480 650 990 770 > 1000
Values taken from Ref. 21. reduction potential of cysteine is close to the value expected for B H T and, therefore, it is likely that electron transfer to a cysteine radical would have a rate constant significantly less than 10 6 M - I s -1. The correlation between physical-chemical parameters and biological function has been widely discussed in the literature. 23'24 From a consideration of both previous work and the results presented here, it is clear that BHT only reacts efficiently with strongly oxidizing species. Of course, this in itself is unsurprising because virtually all biological substrates react with strong oxidants. However, antioxidant activity is generally assumed to involve efficient quenching of a variety of reactive oxygen species, yet BHT quenches these species with rate constants of the order of 105-10 6 M-~s -1, an apparent paradox that has been noted. 14Other mechanisms of antioxidant activity, apart from simple competition kinetics, have to be considered to explain the mechanism of BHT activity. It is possible that BHT is not the active antioxidant in vivo, but rather a metabolite derived or induced by it. This may be the case in the eye where it has been shown that dietary BHT may increase the level of glutathione in the lens, retina, and cornea of the rat. 25 The biological metabolites of BHT have been found to include bezaldehyde and benzoic acid derivatives in the rat. 26 While metabolism may be a factor in the antioxidant properties of BHT in vivo, this is less likely to be the case for the mechanisms that operate to prevent the spoiling of foodstuffs. BHT is known to locate in lipid membranes where it may cause an increase in membrane fluidity, which is the mechanism proposed for its antiviral activity. 5 Localization of BHT in a membrane effectively increases the local concentration of the antioxidant and may be sufficient to allow BHT to compete with hydrogen abstraction processes that have rate constants typically of the order of about 10 5 M - l s - l . The greater mobility of BHT compared to phospholipids implies that it is also possible for BHT to move from the lipid core of the membrane to the surface where the BHT radical may be repaired by c~-tocopherol, thiol antioxidants, or ascorbate. This consideration also strongly
400
C.R. LAMBERTet al.
implies that bimolecular rate constants of the order 105 M - ' s -~ are very important in the context of organized media. The antioxidant power of BHT may also arise through synergistic effects, in that it increases the ability for endogenous antioxidants to operate. The reduction potential of BHT is probably lower than that of other radicals present in the membrane. Thus, it may act as a radical trap for hydrophobic radicals and may, itself, be repaired by endogenous antioxidants. Finally, the possibility has to be considered that the reason BHT is such an effective antioxidant is that it is n o t particularly reactive. If antioxidants such as/3-carotene or vitamin E were incorporated as additives into food stuffs in high concentrations, then these species, although some of the most effective antioxidants against O2(~Ag) and radical species, would be quickly destroyed because of their reactivity. However, BHT is effective against strongly oxidizing radicals and, therefore, can protect against gross oxidizing insults. In most complex biological systems there already exists some level of antioxidant that can effectively protect against less reactive species. Acknowledgements - - We are grateful to Dr. E. J. Land for useful discussion and for the use of the pulse radiolysis facility. We also acknowledge the Cancer Research Campaign, UK, and the Burroughs Wellcome Fund for a travel and research grant to H.S.B.
REFERENCES
1. Collings, A. J.; Sharratt, M. The BHT content of human adipose tissue. Food Cosmet. Toxicol. 8:409-412; 1970. 2. Black, H. S.; Chan, J. T.; Brown, G. E. Effects of dietary constituents on ultraviolet light-mediated carcinogenesis. Cancer Res. 38:1384--1387; 1978. 3. Black, H. S.; Tigges, J. Evaluation of structurally related phenols for anti-photocarcinogenic and photoprotective properties. Photochem. Photobiol. 43:403-408; 1986. 4. Lo, W. B.; Black, H. S. Inhibition of carcinogen formation in skin irradiated with ultraviolet light. Nature 246:489-491; 1973. 5. Snipes, W.; Person, S.; Keith, A.; Cupp, J. Butylated hydroxytoluene inactivates lipid-containing viruses. Science 188:64-66; 1975. 6. Witschi, H. The role of toxicological interactions in chemical carcinogenesis. Toxicol. Pathol. 12:84-88; 1984. 7. Babich, H. Butylated hydroxytoluene (BHT): A review. Environ. Res. 29:1-29; 1982.
8. Windholz, M. ed. The Merck index. Rahway, NJ: Merck & Co., Inc.; 1976; 1976. 9. Thomas, M. J.; Foote, C. S. Chemistry of singlet oxygen-XXVI. Photooxygenation of phenols. Photochem. Photobiol. 27:683~593; 1978. 10. Stevens, B.; Small, R. D., Jr.; Perez, S. R. The photoperoxidation of unsaturated organic molecules--XIII. Oz(1Ag) quenching by a-tocopherol. Photochem. Photobiol. 20:515-517; 1974. 11. Boozer, C. E.; Hammond, G. S.; Hamilton, C. E.; Sen, J. N. Air oxidation of hydrocarbons. II. The stoichiometry and fate of inhibitors in benzene and chlorobenzene. J. Am. Chem. Soc. 77:3233-3237; 1955. 12. Hammond, G. S.; Boozer, C. E.; Hamilton, C. E.; Sen, J. N. Air oxidation of hydrocarbons. III. Mechanism of inhibitor action in benzene and chlorobenzene solutions. J. Am. Chem. Soc. 77:3238-3244; 1955. 13. Richards, J. T.; Adams, F. A. Study of reaction rates of the antioxidants gallic acid, BHT and BHA using the technique of pulse radiolysis. Int. J. Food Sci. TechnoL 22:501-508; 1987. 14. Katusin-Razem, B.; Razem, D. Activity of antioxidants in solution and in irradiated heterogeneous solution. J. Am. Oil Chem. Soc. 71:519-523; 1994. 15. Keene, J. P. Pulse radiolysis apparatus. J. Sci. lnstrum. 41:493496; 1964. 16. Bell, I. P.; Rodgers, M. A. J.; Burrows, H. D. Kinetic and thermodynamic character of reducing species produced on pulse radiolysis of acetonitrile. J. Chem. Soc. Faraday Trans. 1 73:315326; 1977. 17. Draganir, I.; Draganir, Z.; Petkovir, L.; Nikolir, A. The radiation chemistry of aqueous solutions of simple RCN compounds. J. Am. Chem. Soc. 95:7193-7199; 1973. 18. Neta, P.; Schuler, R. H. Rate constants for the reaction of O'radicals with organic substrates in aqueous solution. J. Phys. Chem. 79:1-6; 1975. 19. Buxton, G. V.; Greenstock, C. L.; Helman, W. P.; Ross, A. B. Critical review of rate constants for reactions of hydrated electrons, hydrogen atoms and hydroxyl radicals ('OH/O'-) in aqueous solution. J. Phys. Chem R e f Data 17:513-886; 1988. 20. Land, E. J.; Ebert, M. Pulse radiolysis studies of aqueous phenol. Water elimination from dihydroxycyclohexadienyl radicals to form phenoxyl. Trans. Faraday Soc. 63:1181-1190; 1967. 21. Wardman, P. Reduction potentials of one-electron couples involving free radicals in aqueous solution. J. Phys. Chem. R e f Data 18:1637-1755; 1989. 22. Hunter, E. P. L.; Simic, M. G. Kinetics of peroxyl radical reactions with antioxidants. In: Cohen, G.; Greenwald, R. A. eds. Oxy radicals and their scavenging systems, Vol 1, Molecular aspects. New York: Elsevier Biomedical; 1983:32-37. 23. Halliwell, B. How to characterize a biological antioxidant. Free Radic. Res. Commun. 9:1-32; 1990. 24. Wardman, P.; Clarke, E. D. Redox properties and rate constants in free-radical mediated damage. Br. J. Cancer 55(Suppl. VIII): 172-177; 1987. 25. Ahmad, H.; Sharma, R.; Mansour, A.; Awasthi, Y. C. t-Butylated hydroxytoluene enhances intracellular levels of glutathione and related enzymes of rat lens in vitro organ culture. Exp. Eye Res. 54:41-48; 1992. 26. Hathway, D. E. Metabolic fate in animals of hindered phenolic antioxidants in relation to their safety evaluation and antioxidant function. Adv. Food Res. 15:1-56; 1966.
PII S0891-5849(96)00050-0
ELSEVIER
-J~" Brief Communication REACTIVITY OF BUTYLATED HYDROXYTOLUENE CHRISTOPHERR. LAMBERT,* HOMER S. BLACK, t and T. GEORGE TRUSCOTT* *Wellman Laboratories of Photomedicine, Department of Dermatology, Harvard Medical School, Massachusetts General Hospital, Boston, MA 02114, USA; *Photobiochemistry laboratory, Veterans Affairs Medical Center and Baylor College of Medicine, Houston, TX 77030, USA; and *Chemistry Department, Keele University, Keele, Staffordshire, ST5 5BG, UK
(Received 12 September 1995; Revised 6 December 1995; Accepted 24 January 1996)
Abstract--Butylated hydroxytoluene (BHT) is a synthetic antioxidant that is widely used as an additive in foodstuffs to prevent spoiling. The physical-chemical properties of BHT and many related phenols have been examined previously although the mechanisms by which it exerts its antioxidant properties are poorly understood. The reactivity of BHT with singlet oxygen [O2(IAg)] and a number of radical species has been examined using the techniques of time resolved luminescence and pulse radiolysis. In benzene solution BHT reacted with O2(IAg) at a bimolecular rate constant of 1.3 × 106 M ~s-l. The one-electron oxidized, phenoxyl type BHT radical was generated using pulse radiolysis and the absorption spectrum showed a maximum at 400 nm. BHT reacts slowly with many radical species and upper limits for the bimolecular rate constant for reaction with several electron transfer processes are presented. The antioxidant role of BHT is discussed in terms of its reactivity, localization, and stability. Keywords--Butylated hydroxytoluene, Pulse radiolysis, Free radicals
sibility that BHT may act as a tumor promoter has been suggested, 6 and in addition, the precise safety limits for the use of this compound have yet to be established] The LDs0 in rats is reported in the Merck index 8 to be 1.6-3.2 g/kg from unpublished Eastman Kodak data. It has been assumed that the mechanism by which BHT exerts its antioxidant activity involves the quenching of reactive oxygen species and lipid soluble radicals. Reactive oxygen species may include singlet oxygen [O2(~Ag)], hydroxyl radicals ('OH), superoxide ( 0 2 " ) , and peroxyl radicals ( R O t ' ) . The reaction with O2(~Ag) has been previously investigated in methanol and benzene solution by Foote and co-workers. 9 These authors found that the rate constant for the reaction of BHT with OAIAg) in methanol was an order of magnitude higher than in benzene, and that in methanol, the fraction of quenching processes that led to a chemical reaction was decreased. This implies that BHT is a better antioxidant against O2(IAg)in polar environments, because not only does it react faster in polar solvents, but also a smaller fraction is destroyed by chemical reaction. In contrast, for the quenching of O2( j Ag) by a-tocopherol the fraction of quenching processes that lead to the formation of a chemical product is increased in more polar solvents. L°
INTRODUCTION
Butylated hydroxytoluene (BHT) is a phenolic antioxidant that has been widely used in the United States as a food additive since the Food and Drug Administration granted approval for its use in 1954. B H T is frequently added to dried cereals, cooking oils, and various animal foods. In some species the compound is not readily excreted, and it has been found at relatively high concentrations in human body fat. t B H T is known to give some protection against the toxicity and carcinogenicity of a number of chemicals and physical agents. B H T not only inhibits UV-carcinogenesis, as shown with animals receiving the compound as a dietary supplement, 2 but also provides protection to other physiological responses evoked by UV light. 3 At the cellular level, B H T inhibits lipid peroxidation and the photooxidation of cholesterol to the putative carcinogen, 5-a,6-a cholesterol-epoxide. 4 B H T also acts as an antiviral agent against lipid containing viruses, probably by changing the membrane fluidity and, hence, membrane function. 5 However, more recently the posAddress correspondence to: C. R. Lambert, Wellman Laboratories of Photomedicine, 50 Blossom Street, Wellman 224, Massachusetts General Hospital, Boston, MA 02114. 395
396
C . R . LAMBERT et al.
The reaction of BHT with peroxyl radicals has been examined by Boozer and Hammond.~l'12 The inhibition of radical chain reactions by B H T proceeds by reaction with alkyl peroxyl radicals rather than alkyl radicals except at low oxygen concentrations. These authors found that one B H T molecule is able to react with two peroxyl radicals to give products that are stable to further reaction. The mechanism is a two-step process involving a reversible reaction of BHT with a peroxyl radical to give a B H T - p e r o x y l radical complex that undergoes an irreversible reaction with a second radical. H'~2 The rate-limiting process in this reaction is the formation of the complex, at least in organic solvents. Richards and Adams ~3 attempted to correlate the antioxidant activity of B H T and a number of related molecules with the bimolecular rate constant for electron transfer to the semioxidized tryptophan radical, in aqueous micellar solution. The bimolecular rate constant for this process was reported as 2.9 × 10 6 M - i s 1. More recently, the activity of a number of antioxidants, including BHT, was compared with that of/3-carotene, both in solution and in a model solid food systemJ 4 It was found that despite large differences in solution chemistry, the antioxidant activities of B H T and/3-carotene were comparable. However, to fully investigate the antioxidant properties of BHT, a more comprehensive investigation of its reactivity towards a wider range of oxidizing species is required. Further, the interpretation of bimolecular rate constants in heterogeneous solution is complicated due to compartmentalization effects of the reactants. The present work was undertaken to investigate the reactivity of B H T towards a wider range of radical species. Although the experiments we report have been carried out in homogeneous solution, it is recognized that, in vivo B H T is located within more heterogeneous environments, which may effectively increase local concentrations and, thus, alter the relative importance of these reactions.
water to 40% acetonitrile (60:40 H20:ACN), which corresponds to a molar ratio of water to acetonitrile of 4.3:1.
Pulse radiolysis The pulse radiolysis facility at the Paterson Institute CRC laboratories has been described previouslyJ s Pulse widths were varied between 20 ns and 100 ns, which gave a maximum dose of 5 krad. Sample cells were 2.5 cm pathlength and the sample was changed after every electron pulse. The radiation chemistry of water has been well characterized and it is relatively easy to generate either purely oxidizing or reducing conditions. Although the radiation chemistry of acetonitrile has been studied previously, 16'17 the radiation chemistry of mixed solvent systems have not been commonly used in pulse radiolysis, and a number of side reactions have to be considered. Oxidizing conditions for pulse radiolysis are obtained routinely, in water, by using nitrous oxide saturated, sodium azide, solutions. The following reaction sequence is complete within a couple of hundred nanoseconds after the pulse. Radiation Pulse
H20
"OH + eaq + H" + O H - + H + eaq + NeO + HeO -" Ne + "OH + O H N3 + "OH --- N~ + O H -
Oxidizing conditions can also be generated using the same reaction sequence in micellar solution. In acetonitrile, the main radical species generated is the acetonitrile radical anion, CH3CN'-, which reacts reversibly with acetonitrile to give the dimeric species (CH3CN)2 ° . In the presence of water, the radical anion may protonate to form CH3C'NH and the dimeric species is not formed. The radical species generated from water radiolysis also react with acetonitrile with appreciable rate constants:
M A T E R I A L S AND M E T H O D S
Materials B H T was obtained from Sigma (Poole, UK) and sublimed under vacuo before use. Water was doubly distilled, the second time from alkaline permanganate. All other materials were of the purest available from Fluka (Gillingham, UK) and used as received. BHT is virtually insoluble in water and only sparingly soluble in micellar solution, but it was possible to carry out some experiments in 2% aqueous Triton X-100. Due to problems of solubility the pulse radiolysis experiments were carried out in a solvent mixture of 60%
CH3CN + H" ~ CH3C'NH (3.6 × 106 M - i s - l ) 17 CH3CN + "OH ~ CH3C'NOH (2.2 × 107 M-~s-~) ~8 CH3CN + e~q ~ CH3CN'- (3.7 × 107 M - i s - l ) 19 Using 60:40 H20:ACN, the concentration of acetonitrile is 7.6 M, which means that the radical adducts will be formed with pseudo first-order rate constants of 2 × 10 7 S-1, 2 × 108 S -1, and 4 × 10 7 S -1, respectively. However, azide reacts with the hydroxyl radical at 1.2 × 101° M - i s -~, which gives a corresponding rate constant for the growth of the azide radical, in the presence
Reactivity of butylated hydroxytoluene
of 0.05 M azide used in these experiments, of 6.0 x 108 s ~. The radical adducts of acetonitrile eventually form stable products such as ammonia, acetaldehyde, and carbon dioxide. Thus, although the solvent system used here is not as well characterized as pure water the environment after the pulse is still predominantly oxidizing, with more than 70% of "OH reacting to give N;. Reducing conditions for pulse radiolysis are obtained routinely, in water, by using nitrous oxide or nitrogen-saturated, sodium formate solutions. HCO2 + "OH --, CO2"- + H 2 0 HCO~ + H"
~
C02
°-
+
H2
Formate reacts with the hydroxyl radical with a rate constant of 3 x 109 M - I s - l , which gives a pseudo firstorder rate constant for the formation of CO2'- of 3 × 108 M-~s ~ in the presence of 0.1 M formate. This means that approximately 50% of the hydroxyl radicals will react to form CO2"- and 50% to form the acetonitrile adduct. The hydrogen atom reacts with formate with a bimolecular rate constant of 2.7 X 108 M - i s -~, which means that about 50% of the H" should react to give CO2"-. In oxygen-saturated formate solutions the superoxide radical is formed. This species is also formed in the radiolysis of acetonitrile, i.e., C 0 2 " - -t- 0 2 -'+ 0 2 " - -1--C 0 2
e~ + 02 ~ 02" CH3CN"
+ 0 2 ~ 0 2"
+
CH3CN
e~ represents the solvated electron in the mixed solvent system. No reaction was observed between BHT and any of the reducing species described above.
397 RESULTS
Generation of the BHT phenoxyl radical Pulse radiolysis was used to generate the semioxidized radical of BHT in the mixed solvent system, 60:40 HzO:ACN using the azide radical as oxidant. The time resolved spectrum of this species is shown in Figure 1. A similar spectrum was also obtained using aqueous 2% TX- 100 as solvent, confirming that under our conditions the initial radiation chemistry still gave mainly oxidizing conditions and there was little interference from other solvent derived radicals. The absorption band around 400 nm is typical of other phenoxyl radicals that have been reported in the literature, z° The bimolecular rate constant for reaction between the azide radical and BHT was estimated to be 4.5 × 10 9 M - i s -j. We also attempted to obtain the spectrum of the radical formed directly from "OH, in the absence of azide. Under these conditions the apparent yield of the phenoxyl radical is much reduced, which is a consequence of "OH reacting with acetonitrile. The spectrum obtained under these conditions is also distorted by a strong absorption around 300 nm observed under the same conditions in the absence of BHT.
Reactivity with known antioxidants and biological radicals The rate of electron transfer from B H T to the semioxidized tryptophan radical has been used as a measure of antioxidant activity by Richards and A d a m s ] 3 In the present work electron transfer was attempted with a number of radical species and these experiments are summarized in Table 1. The results are presented as an upper limit for the bimolecular rate constant because it was not possible to solubilize BHT at sufficient concentrations in 60:40
!
1
Singlet oxygen luminescence Singlet oxygen was generated in benzene solution by 532 nm laser irradiation of the photosensitizer, protoporphyrin IX dimethyl ester (PPIX DME). The 8 m m diameter, 8 ns frequency doubled line of a J. K. Lasers, System 2000 Nd:YAG laser was used as the excitation source and the time resolved near infrared (NIR) luminescence was monitored at 90 ° to the laser using an E.O.S.S. Inc. 5 mm: germanium diode. Samples were air saturated and the bimolecular rate constant for reaction with Oz(~Ag) was determined using a range of B H T concentrations.
_
i t . l ~
a
504gS-
x a
0.004-
-
c5 0.000
I
25O
i
350 W a v e l e n g t h (nm)
_
_
450
Fig. 1. Transient absorption spectrum of BHT (2 × 10 4 M) in N20 saturated solution of 0.05 M sodium azide (dose ~ 45 G).
398
C.R. LAMBERTet al.
Table 1. Upper Limits of the Rate Constants for Electron Transfer Reactions with BHT Oxidized Radical
Electron Donor
BHT BHT BHT Trolox Vitamin C Tryptophan
Donor Concentration
Trolox Vitamin C Cysteine BHT BHT BHT
1× 1X 1× 1 )< 1× 2×
10 4M 10-4M
10-5 M 10 4M 10 4 M 10-aM
10.00
Rate Constant Upper Limit <3 x 106M iS <3 X 106 M-as <2 × 107 M-~s-~ <106M Is I <106 M Is <106M Is I
6.00
2.00 H 2 0 : A C N to observe electron transfer in these experiments. W e were also unable to repeat the e x p e r i m e n t s p e r f o r m e d b y R i c h a r d s and A d a m s , again b e c a u s e insufficient B H T could be solubilized into SDS solution to affect the rate o f d e c a y o f the tryptophan radical.
Reaction with
h~
isc
~
3p,
02
~
I 0.04
I 0.06
The gradient o f this line gives the b i m o l e c u l a r rate constant for the reaction o f B H T with O2(1Ag), which was found to be 1.3 × 106 M is 1.
DISCUSSION
The quenching o f reactive o x y g e n species b y B H T has been studied extensively over the past 20 years b y c h e m i c a l methods. The nature o f the quenching process is k n o w n to vary, d e p e n d e n t on the reactive o x y g e n species and the local environment. The rate constants for reaction o f B H T with O2(~ Ag), p e r o x y l radicals, and h y d r o p e r o x i d e s have been m e a s u r e d p r e v i o u s l y and the products o f these reactions determined. H o w e v e r , a certain p a r a d o x still remains that despite the p o t e n c y of B H T as an antioxidant it reacts efficiently only with
p + O2(1mg)
The concentration o f B H T was varied over a concentration range o f up to 5 × 10 2 M. A plot o f the rate o f d e c a y o f O2(IAg) as a function o f B H T concentration is shown in Fig. 2.
•
/ 2 ROO ° •-
+
-
I
Fig. 2. Observed rate constant (k') vs. concentration plot for the determination of bimolecular rate constant for O2(~Ag) with BHT. The kinetics were monitored using the NIR luminescence.
O2(1mg)
__, x p ,
I 0.02
[BHT] M
It was not p o s s i b l e to solubilize B H T at sufficient concentration in D 2 0 - b a s e d m i c e l l a r solutions to observe any change in the rate o f d e c a y o f O:(1Ag); therefore, this value was d e t e r m i n e d in b e n z e n e solution. Singlet o x y g e n was generated using 532 n m excitation o f P P I X D M E according to the f o l l o w i n g reaction s c h e m e w h e r e P represents P P I X D M E . p
I 0
RO2H
OH
... O2(1~g)
Physical quenching to give ground state oxygen and B H T
Scheme 1. Reactions of BHT with reactive oxygen species.
Reactivity of butylated hydroxytoluene strongly oxidizing species such as the hydroxyl radical. These quenching reactions are summarized in Scheme 1. The scheme depicts the reaction of B H T with, hydroxyl radicals, peroxyl radicals, and O2(IAg). Hydroxyl radicals may react either to give an adduct or by electron transfer to give the phenoxyl radical. The phenoxyl radical is relatively stable but may eventually react with another radical or polymerize. The mechanism of reaction of "OH with phenols was originally investigated by Land and Ebert. 2° The reaction does not proceed by electron transfer directly but rather by "OH addition followed by acid and base-catalyzed elimination to give the phenoxyl radical. The mechanism of reaction of BHT with peroxyl radicals was proposed by Boozer and Hammond and involves the deactivation of two radicals. Reaction with the first radical is a reversible process to give a peroxyl r a d i c a l - B H T complex. This complex may then reduce a second peroxyl radical by hydrogen transfer. The mechanism of reaction of BHT with O2(lAg) involves the formation of an encounter complex. This reaction is reversible but the collapse of the complex may give product formation or the parent ground state molecules. The fraction of quenching processes that lead to product formation is controlled by the balance between the rate constants for product formation and deactivation and, as shown by Foote et al., 9 is solvent dependent. The one electron reduction potential of BHT has not been determined, but it is likely to be around 6 0 0 - 7 0 0 mV. This value is assumed by comparison with other substituted phenols. The reduction potential of phenol itself is reported to be more positive than 800 mV. Methyl substitution at the para position (p-cresol) lowers the potential to 770 mV. Substitution of two tertbutyl groups is likely to reduce the potential still further. The reduction potential of the compounds used in this study are given in Table 2. 21 Examination of Table 2 shows that it is thermodynamically favorable for the B H T radical to be reduced by ascorbate and Trolox and for BHT to reduce the tryptophyl and trichloromethyl peroxyl (CC13OO") radicals. The rate constant determined by Richards and Adams for electron transfer from BHT to the tryptophyl radical was 2.9 × 10 6 M-1s - l . This is bigger than the upper limit determined in the present work but probably reflects the effect of compartmentalization into the micelles. The bimolecular rate constant for reaction of BHT with CC13OO" has been measured in carbon tetrachloride and found to be 6 × 10 6 M - l s - l . 22 This last result clearly demonstrates that even for thermodynamically favorable peroxyl radicals the bimolecular rate constant for electron transfer is still relatively slow. The
399
Table 2. Reduction Potentials of Compounds Relevant to This Study Compound
E (mV)
Ascorbate Trolox Cysteine Tryptophan p-Cresol Trichloromethyl peroxyl
300 480 650 990 770 > 1000
Values taken from Ref. 21. reduction potential of cysteine is close to the value expected for B H T and, therefore, it is likely that electron transfer to a cysteine radical would have a rate constant significantly less than 10 6 M - I s -1. The correlation between physical-chemical parameters and biological function has been widely discussed in the literature. 23'24 From a consideration of both previous work and the results presented here, it is clear that BHT only reacts efficiently with strongly oxidizing species. Of course, this in itself is unsurprising because virtually all biological substrates react with strong oxidants. However, antioxidant activity is generally assumed to involve efficient quenching of a variety of reactive oxygen species, yet BHT quenches these species with rate constants of the order of 105-10 6 M-~s -1, an apparent paradox that has been noted. 14Other mechanisms of antioxidant activity, apart from simple competition kinetics, have to be considered to explain the mechanism of BHT activity. It is possible that BHT is not the active antioxidant in vivo, but rather a metabolite derived or induced by it. This may be the case in the eye where it has been shown that dietary BHT may increase the level of glutathione in the lens, retina, and cornea of the rat. 25 The biological metabolites of BHT have been found to include bezaldehyde and benzoic acid derivatives in the rat. 26 While metabolism may be a factor in the antioxidant properties of BHT in vivo, this is less likely to be the case for the mechanisms that operate to prevent the spoiling of foodstuffs. BHT is known to locate in lipid membranes where it may cause an increase in membrane fluidity, which is the mechanism proposed for its antiviral activity. 5 Localization of BHT in a membrane effectively increases the local concentration of the antioxidant and may be sufficient to allow BHT to compete with hydrogen abstraction processes that have rate constants typically of the order of about 10 5 M - l s - l . The greater mobility of BHT compared to phospholipids implies that it is also possible for BHT to move from the lipid core of the membrane to the surface where the BHT radical may be repaired by c~-tocopherol, thiol antioxidants, or ascorbate. This consideration also strongly
400
C.R. LAMBERTet al.
implies that bimolecular rate constants of the order 105 M - ' s -~ are very important in the context of organized media. The antioxidant power of BHT may also arise through synergistic effects, in that it increases the ability for endogenous antioxidants to operate. The reduction potential of BHT is probably lower than that of other radicals present in the membrane. Thus, it may act as a radical trap for hydrophobic radicals and may, itself, be repaired by endogenous antioxidants. Finally, the possibility has to be considered that the reason BHT is such an effective antioxidant is that it is n o t particularly reactive. If antioxidants such as/3-carotene or vitamin E were incorporated as additives into food stuffs in high concentrations, then these species, although some of the most effective antioxidants against O2(~Ag) and radical species, would be quickly destroyed because of their reactivity. However, BHT is effective against strongly oxidizing radicals and, therefore, can protect against gross oxidizing insults. In most complex biological systems there already exists some level of antioxidant that can effectively protect against less reactive species. Acknowledgements - - We are grateful to Dr. E. J. Land for useful discussion and for the use of the pulse radiolysis facility. We also acknowledge the Cancer Research Campaign, UK, and the Burroughs Wellcome Fund for a travel and research grant to H.S.B.
REFERENCES
1. Collings, A. J.; Sharratt, M. The BHT content of human adipose tissue. Food Cosmet. Toxicol. 8:409-412; 1970. 2. Black, H. S.; Chan, J. T.; Brown, G. E. Effects of dietary constituents on ultraviolet light-mediated carcinogenesis. Cancer Res. 38:1384--1387; 1978. 3. Black, H. S.; Tigges, J. Evaluation of structurally related phenols for anti-photocarcinogenic and photoprotective properties. Photochem. Photobiol. 43:403-408; 1986. 4. Lo, W. B.; Black, H. S. Inhibition of carcinogen formation in skin irradiated with ultraviolet light. Nature 246:489-491; 1973. 5. Snipes, W.; Person, S.; Keith, A.; Cupp, J. Butylated hydroxytoluene inactivates lipid-containing viruses. Science 188:64-66; 1975. 6. Witschi, H. The role of toxicological interactions in chemical carcinogenesis. Toxicol. Pathol. 12:84-88; 1984. 7. Babich, H. Butylated hydroxytoluene (BHT): A review. Environ. Res. 29:1-29; 1982.
8. Windholz, M. ed. The Merck index. Rahway, NJ: Merck & Co., Inc.; 1976; 1976. 9. Thomas, M. J.; Foote, C. S. Chemistry of singlet oxygen-XXVI. Photooxygenation of phenols. Photochem. Photobiol. 27:683~593; 1978. 10. Stevens, B.; Small, R. D., Jr.; Perez, S. R. The photoperoxidation of unsaturated organic molecules--XIII. Oz(1Ag) quenching by a-tocopherol. Photochem. Photobiol. 20:515-517; 1974. 11. Boozer, C. E.; Hammond, G. S.; Hamilton, C. E.; Sen, J. N. Air oxidation of hydrocarbons. II. The stoichiometry and fate of inhibitors in benzene and chlorobenzene. J. Am. Chem. Soc. 77:3233-3237; 1955. 12. Hammond, G. S.; Boozer, C. E.; Hamilton, C. E.; Sen, J. N. Air oxidation of hydrocarbons. III. Mechanism of inhibitor action in benzene and chlorobenzene solutions. J. Am. Chem. Soc. 77:3238-3244; 1955. 13. Richards, J. T.; Adams, F. A. Study of reaction rates of the antioxidants gallic acid, BHT and BHA using the technique of pulse radiolysis. Int. J. Food Sci. TechnoL 22:501-508; 1987. 14. Katusin-Razem, B.; Razem, D. Activity of antioxidants in solution and in irradiated heterogeneous solution. J. Am. Oil Chem. Soc. 71:519-523; 1994. 15. Keene, J. P. Pulse radiolysis apparatus. J. Sci. lnstrum. 41:493496; 1964. 16. Bell, I. P.; Rodgers, M. A. J.; Burrows, H. D. Kinetic and thermodynamic character of reducing species produced on pulse radiolysis of acetonitrile. J. Chem. Soc. Faraday Trans. 1 73:315326; 1977. 17. Draganir, I.; Draganir, Z.; Petkovir, L.; Nikolir, A. The radiation chemistry of aqueous solutions of simple RCN compounds. J. Am. Chem. Soc. 95:7193-7199; 1973. 18. Neta, P.; Schuler, R. H. Rate constants for the reaction of O'radicals with organic substrates in aqueous solution. J. Phys. Chem. 79:1-6; 1975. 19. Buxton, G. V.; Greenstock, C. L.; Helman, W. P.; Ross, A. B. Critical review of rate constants for reactions of hydrated electrons, hydrogen atoms and hydroxyl radicals ('OH/O'-) in aqueous solution. J. Phys. Chem R e f Data 17:513-886; 1988. 20. Land, E. J.; Ebert, M. Pulse radiolysis studies of aqueous phenol. Water elimination from dihydroxycyclohexadienyl radicals to form phenoxyl. Trans. Faraday Soc. 63:1181-1190; 1967. 21. Wardman, P. Reduction potentials of one-electron couples involving free radicals in aqueous solution. J. Phys. Chem. R e f Data 18:1637-1755; 1989. 22. Hunter, E. P. L.; Simic, M. G. Kinetics of peroxyl radical reactions with antioxidants. In: Cohen, G.; Greenwald, R. A. eds. Oxy radicals and their scavenging systems, Vol 1, Molecular aspects. New York: Elsevier Biomedical; 1983:32-37. 23. Halliwell, B. How to characterize a biological antioxidant. Free Radic. Res. Commun. 9:1-32; 1990. 24. Wardman, P.; Clarke, E. D. Redox properties and rate constants in free-radical mediated damage. Br. J. Cancer 55(Suppl. VIII): 172-177; 1987. 25. Ahmad, H.; Sharma, R.; Mansour, A.; Awasthi, Y. C. t-Butylated hydroxytoluene enhances intracellular levels of glutathione and related enzymes of rat lens in vitro organ culture. Exp. Eye Res. 54:41-48; 1992. 26. Hathway, D. E. Metabolic fate in animals of hindered phenolic antioxidants in relation to their safety evaluation and antioxidant function. Adv. Food Res. 15:1-56; 1966.