Document NGgdzaaYqkK1Op8kg0rYoqk6g

THEJOURNAL OF BIOLOGICCAHLEMISTRY 0 1985 by The American Society of Biological Chemists, Inc. Vol. 260, No. 30, Issue of December 25, pp. 16210-16214,1985 Printed in U.S.A. Metabolism of Diethylstilbestrol by Horseradish Peroxidaseand Prostaglandin-H Synthase GENERATION OF A FREE RADICAL INTERMEDIATE AND ITS INTERACTION WITH GLUTATHIONE* (Received for publication, May 17,1985) David RosslB, RolfJ. Mehlhornllll,Peter Moldeus$**, and Martyn T. Smith$$$ From the$Department of Biomedical and Environmental Health Sciences, School of Public Health, University of California, Berkeley, California 94720 and the llApplied Sciences Division, Lawrence Berkeley Laboratory, Berkeley, California 94720 Downloaded from www.jbc.org by guest, on April 1, 2010 Diethylstilbestrol is carcinogenic in rodents and in prevent miscarriages and as a postcoital contraceptive. DES humans and its peroxidatic oxidation in utero has been has been shown to be carcinogenic in animals (1)and itsuse associated with its carcinogenic activity. Horseradish as an abortifacient in humans hasbeen linked to theappear- peroxidase-catalyzed oxidation of [14C]diethylstilbes- ance of genital tract tumors in women whose mothers were trol and [14C]diethylstilbestroal nalogs inducedbinding treated with the drug (2, 3). Reproductive tract lesions in of radiolabel to DNA only when the compound con- human males (4-6), and in mice (7) exposed prenatally to tained a free hydroxy group (Metzler, M., and Epe, B. DES have also been reported. Furthermore, DES has been (1984) Chem. Biol. Interact. 50, 351-360). We have shown to cause kidney carcinoma in Syrian hamsters(8)and found that horseradish peroxidase or prostaglandin-H synthase-catalyzed oxidation of diethylstilbestrol in the presence of the spin trap 5,5-dimethyl-l-pyrrolineN-oxide caused the generationof an ESR signal indic- ative of a free radical intermediate(UN = 14.9 G , aH = 18.3 G ) .The identity of the trapped radicalcould not be identified on the basis of published hyperfine coupling constants, but the observation that horseradish to induce neoplastic transformation in Syrian hamster embryo fibroblasts i n vitro (9). The mechanisms underlying DES carcinogenicity are as yet unresolved but have been thought to involve the estrogenic activity of the compound (I). Recent data, however, from in vivo carcinogenicity studies (10,11) andstructure-activity studies on neoplastic transformation in vitro (9) suggest that peroxidase-catalyzed oxidation of 1-naphthol pro- hormonal activity alone is insufficient to explain the carcin- duced an identical ESR signal suggests that the radical ogenicity of DES. was eithear phenoxy or phenoxy-derived radical. Dur- DES is metabolized in both animals and man (12, 13) and ing horseradish peroxidase-catalyzed oxidation of di- has been shown to be extensively oxidized to reactive inter- ethylstilbestrol in thepresence of glutathione the thiol mediates by peroxidases in utero (14). Thus peroxidatic oxi- reduced the diethylstilbestrol radicaltogenerate a dation of DES is a possible determinant of its carcinogenicity. thiyl radical. This was shown by a thiol-dependent One product of the peroxidase-catalyzed oxidation of DES by oxygen uptakeduring horseradish peroxidase-cata- uterine peroxidase and horseradish peroxidase has been char- lyzed oxidation of diethylstilbestrol and the observa- acterized as DES quinone, which covalently binds to DNA tion of an ESR signal consistent with 5,5-dimethylpyr- and is thus a possible cause of the carcinogenic activity of roline-N-oxide-glutathionylradical adduct formation. A diethylstilbestrol analog devoid of free hydroxy groups, namely diethylstilbestrol dipropionate, did not produce an ESR signal above control levels during horseradish peroxidase-catalyzed metabolism in the presence of 5,5-dimethylpyrroline-N-oxide.Thus, free radicals are formed during peroxidatic oxidation of diethylstilbestrol and must be considered as possible determinants of the genotoxic activity of this com- pound. DES (15). A recent study showed, however,that horseradish peroxidase-catalyzedmetabolism of [14C]DESderivatives that could not form quinones resulted in DNA binding. This study showed that atleast one free hydroxy group was required for DNA binding and suggested that the phenoxy free radical derived from DES was the causative agent (16)..The formation of free radicals during peroxidase-catalyzed metabolism of DES has, however, never been shown. In thisstudy we have examined the metabolism of DES by horseradish peroxidase and prostaglandin synthase, a perox- idase that is present in organs susceptible to DES-induced Diethylstilbestrol (DES1) is a synthetic estrogen that has been used as a food additive for livestock and, in women, to * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Present address: School of Pharmacy, University of Colorado, Boulder, CO 80309. 11 Supported by National Institutes of Health GrantAG-04818. ** Permanent address: Department of Toxicology, Karolinska In- stitute, Stockholm, Sweden. `$$ Supported by the National Foundation for Cancer Research. The abbreviations used are: DES, diethylstilbestroh DMPO, 5,5dimethylpyrroline-N-oxide. carcinogenesis (17, 18).We show that both of these peroxidatic oxidations proceed via the production of free radical intermediates that can interact with theendogenous protective thiol glutathione (GSH)to produce glutathionyl radicals. MATERIALS AND METHODS Chemicals and Enzymes-DES,DES dipropionate, 1-naphthol, horseradish peroxidase Type VI, hydrogen peroxide, arachidonic acid, hematin, and indomethacin were obtained from Sigma.5-5-Dimethyl1-pyrroline-N-oxide (DMPO) was obtained from Aldrich. p-Phenetidine hydrochloride was synthesized as described previously (19). Purified prostaglandin-H synthase was a generous gift from Prof. L. Marnett, Department of Chemistry, Wayne State University, Detroit, MI. Incubations with Horseradish Peroxidase-The following condi- 16210 Downloaded from www.jbc.org by guest, on April 1, 2010 Peroxidase-catalyzed Metabolism of DES 16211 tions were used hydrogen peroxide (0.1 mM), horseradish peroxidase (3 pg/ml), DES or 1-naphthol (0.1 mM), and in some cases GSH (5 mM). Potassium phosphate buffer (0.1 M, pH = 7.4) was used for all experiments. In experiments usingESR spectroscopy, DMPO (0.1 M) was included in the reaction mixture. Reactions were performed at 25 "Cand were initiated by the addition of hydrogen peroxide. Incubations with Prostaglandin-H Synthase-The following con- ditions were used arachidonic acid (0.1 mM) or hydrogen peroxide (0.05 mM), purified prostaglandin-H synthase (equivalent to 12.3 pg of protein), DES (0.1 mM), hematin (1.0pM), and DMPO (0.1 M). Reactions were performed at 25 "C and were initiated by addition of either arachidonic acid or hydrogen peroxide. Oxygen Uptake-This was measured using a Clarke electrode in a sample volume of 3 ml at 25 "C. ESR Spectroscopy-This was performed on a Varian E109E spectrometer at room temperature. Instrument conditions, unless other- wise stated, werepower (10mW), modulation amplitude (1.25 G), time constant (0.064 s), receiver gain (2.5 X lo*), and scan time (1 min). RESULTS Horseradish peroxidase- and prostaglandin-H synthasecatalyzed oxidation of DES leads to thegeneration of colored metabolites and theinclusion of GSH in such reactions inhibits color development. This suggests that either GSH reacts with these metabolites or prevents their formation. Horseradish peroxidase-catalyzed co-oxidation of xenobiotics, driven by hydrogen peroxide, is not an oxygen-consuming reaction (20), and there was no immediate oxygen uptake observed during horseradish peroxidase-catalyzed oxidation of DES. Interestingly, a small delayedoxygen uptake was observed (Table I) reflecting, presumably, oxidation of a primary metabolite. In the presence of GSH, however, an immediate oxygen uptake was apparent (Table I) and the incubation mixture remained colorless.A GSH-dependent oxygenuptake TABLEI GSH-dependent oxygen uptake observed during the horseradish peroxidase-catalyed oxidation of diethylstilbestrol(A)and 1naphthol(B) complete incubations contained Hz02(0.1 mM), horseradish peroxidase (3 pg/ml), GSH (5 mM), and either DES (0.1 mM, A) or 1naphthol (0.1 mM, B). Incubations without DES contained an equivalent volume of the solvent dimethyl sulfoxide. Incubation mixture A) Complete minus GSH minus DES minus horseradish peroxidase Initial rate of oxygen uptake *nmol/ml/min 18 2 (n = 3) 0" 0 0 B) Complete 23 minus GSH 0 ~ A delayed rate of oxygen consumption was consistently observed in these reactions(delay 30 s, rate 5 -C 1nmol/ml/min). was also observed during horseradish peroxidase-catalyzed oxidation of another phenolic compound, 1-naphthol, in the presence of GSH (Table I). We have previously shown that thiol-dependent oxygenuptake occursduring horseradishperoxidase-catalyzed oxidation of both p-phenetidine and acetaminophen (21,22),reactions known to involvethe generation of cosubstrate-derived free radicals.(22, 23).The mechanism underlying this oxygen consumption involves the reduction of the cosubstrate-derivedradical by GSH to generate a thiyl radical that theneither reacts with oxygen to form a peroxysulfenyl radical or with GSH to form a glutathionyl anion radical that then autoxidizesto form superoxide (24). These 'results imply that free radical species are produced during the peroxidatic oxidation of DES, and we sought to confirm radical formation with ESR spectroscopy. Indeed, in the presence of the spin trap, DMPO, a 6-line paramagnetic signal (uN = 14.9G, UH = 18.3 G) could be detected during horseradish peroxidase-catalyzedoxidation of DES (Fig. L4). Control reactions without DES, inthe presence of boiled enzyme,without enzyme,without hydrogen peroxideo, r without DMPO produced signals indistinguishable fromnoise (Fig. 1B).This signal was not dependent on the presence of solvent (dimethyl sulfoxide), since control experiments using ethanol, where the solvent was evaporated prior to initiation of the enzyme reaction, still produced the 6-line signal (as in Fig. lA). The ESR signal shown in Fig. 1A was short-lived and decayed appreciablywithin three successive 1-min scans. When the reaction was performed in the absence of DMPO no signal could be detected. Similarly, if the spin trap was added after 15 min of reaction no signal was detected, confirming the transient nature of the radical. The identity of the radical species (Fig. 1A) could not be deduced from our data; the hyperfine coupling constants of the observed signal cannot be identified with available published constants (25). We also studied the horseradish peroxidase-catalyzed oxidation of another phenolic compound, 1-naphthol, and found an identical six-line signal, which was enzyme- and cosubstrate-dependent (Fig. 2). Horseradish peroxidase-catalyzed oxidation of DES dipropionate (0.1mM) and the aromatic amine p-phenetidine (0.1-0.5 mM) produced no ESR signals, showingthat a free hydroxy groupis needed fora cosubstrate to produce this particular radical signal, and that enzyme turnover induced by very efficient cosubstrates such as p phenetidine is notsufficient to generate this signal. The above experiments show that a free radical species is generated during horseradish peroxidase-catalyzed oxidation of DES. Moreover, the data shown in Table I suggest that such a radical can interact with GSH to form a thiyl radical. This was confirmed using DMPO to trap the glutathionyl radical (26-28) as shown in Fig. 3. An ESR signal identical to the signal observedwhen glutathione was treated with a hydroxyl radical-generating system (data not shown) was FIG. 1. ESR spectra observed during horseradish peroxidase-catalyzed oxidationof DES in the presence of DMPO. A , DES added as a dimethyl sulfoxide solution (0.1 m ~ ) , horseradish peroxidase (3pglml), hydro- gen peroxide (0.1 mM), and DMPO (0.1 M); B, as in A , but with DES omitted from the dimethyl sulfoxide. Similar spectra were seen when horseradish peroxidase was omitted, when hydrogen peroxide was omitted, when DMPO was omitted, or when boiled enzyme was substituted for horseradish peroxidase. 'IO GAUSS' A B 16212 Peroxidase-catalyzed Metabolismof DES FIG.2. ESR spectroabserved during horseradish peroxidase-cat- alyzedoxidation of 1-naphthol in the presence oDf MPO. A, 1-naphthol in dimethyl sulfoxide (0.1mM), horse- radish peroxidase (3 pglml), hydrogen peroxide (0.1 mM), and DMPO (0.1 M); B, as in A but with 1-naphthol omitted from the dimethyl sulfoxide. A similar spectrum was observed when horserad- ish peroxidase was omitted. - 10 GAUSS A .B Downloaded from www.jbc.org by guest, on April 1, 2010 FIG.3. ESR spectrraesulting fromthehorseradishperoxidasecatalyzed oxidation of DES in the presence ofGSH and DMPO. A, as in Fig. lA, but with 5 mM GSH added; B , as inA but with horseradish peroxidase omitted; C, DES omitted from the dimethyl sulfoxide; D, GSH added after incubating a GSH-free reaction mixture for 15 min. I - 10 GAUSS B C D observed during the horseradish peroxidase-catalyzed oxidation of DES in the presence of DMPO and GSH (Fig. 3 A ) , which was stable up to 3 min after initiation of the reaction and thensubsequentlydeclined. This signalwas enzyme- and DES-dependent (Fig. 3, B and C ) and was not produced if GSH was omitted from the initial reaction mixture or if it was added to thereaction after 15 min (Fig. 30). To show that these reactions occurred in thepresence of a biologically relevant peroxidase, the ESR investigations were repeated using prostaglandin-H synthase rather than horseradish peroxidase. Fig. 4A shows that the 6-line signal obtained during horseradish peroxidase-catalyzedoxidation was also obtained during prostaglandin-H synthase-catalyzedmetabolism of DES in thepresence of arachidonic acid. A small signalwith identical couplingconstants to the previous signal was also obtained in the absence of DES (Fig. 4B) and was presumably due to turnover of some component associated with the reaction mixture by the enzyme. The generation of the intense 6-line ESR signal was arachidonate- and prosta- glandin-H synthase-dependent (Fig. 4, C and D),and was almost totally inhibited by indomethacin (Fig. 4E),an inhib- itor of prostaglandin-H synthase. That the hydroperoxidase component of prostaglandin-H synthase was responsible for oxidation of DES to radical products is shownby the experi- ments described in the legend to Fig. 5. These demonstrate that hydrogen peroxide couldsubstitute for arachidonic acid as thesubstrate for prostaglandin-H synthase. Controls without DES (Fig. 5B), enzyme (Fig. 5C), or hydrogen peroxide (Fig.5 0 ) produced signalsof much lowerintensity than those obtained with the complete incubation system(Fig. 5 A ) . Furthermore, the possibility of nonenzymaticprocesses causing the generation of such a radical specieswas discounted as incubations containingboiled enzyme producnedo signal (Fig. 5E).As in thecase of horseradish peroxidase, prostaglandinH synthase-catalyzedoxidationof DESdipropionate (0.1mM) or p-phenetidine (0.1-0.5 mM) did not induce the formation of ESR signals above control levels (Fig.5, F and G)? DISCUSSION Peroxidase-catalyzedoxidationof DES occursin all in vitro systems in which DES has been found to be genotoxic (16, To ensure that the ESRsignal observed during prostaglandin H synthase-catalyzed metabolism of DES was not theresult of a marked increase in enzyme turnover, arachidonate-stimulated,prostaglandin H synthase-dependent oxygen uptake (using conditions identical to those used in the ESR experiments) was measured. The initial rate of oxygen uptake observed, indicative of enzyme turnover, was increased in the presence of both p-phenetidine (0.1mM) and DES (0.1 mM) by only 12 and 6%,respectively, of the control value. Peroxidase-catalyzed Metabolismof DES .1 0 C D 16213 rA Downloaded from www.jbc.org by guest, on April 1, 2010 . `10 GAUSS FIG.4. ESR spectra observed during arachidonic acid-de- pendent prostaglandin-H synthase-catalyzed oxidation of F DES in the presence of DMPO. A , DES dissolved in dimethyl sulfoxide (0.1mM), arachidonic acid (0.1 mM),purified prostaglandinH synthase (12.3 pg/ml protein), hematin (1.0 p ~ )a,nd DMPO (0.1 M);B, as in A but with DES omitted from the dimethyl sulfoxide; C, arachidonic acid omitted; D, prostaglandin-H synthase omitted; E , as inA but with indomethacin (0.1 mM) added. -G 10 GAUSS FIG.5. ESR spectra observed during hydrogen peroxide- dependent prostaglandin-H synthase catalyzed oxidation of ., 29) and also in organs susceptible to DES carcinogenicity DES in the presence of DMPO. A, conditions as in 4A but the (30). One product of peroxidase-catalyzedoxidation of DES, reaction was initiated by adding hydrogen peroxide (0.05 mM); B, as DES-quinone,binds to DNA and hastherefore been proposed as a possible carcinogenic metabolite of DES (15). A recent study, however, has shown that quinone formation isnot necessary for the binding of structural analogs of DES to in A but with DES missing from the dimethyl sulfoxide; C, without prostaglandin-H synthase; D, without hydrogen peroxide; E, boiled enzyme; F, with DES diproprionate in place of DES; G,p-phenetidine HCI(O.1 mM) in place of DES. DNA during peroxidase-catalyzedmetabolism (16)T. he latter study concluded that one free hydroxy group was necessary dase-catalyzed oxidation of DES in the presence of GSh is for the induction of DNAbinding during peroxidaticoxidation further evidence of the one electron mechanism operative in of analogs of DES and proposed the iemiquinone radical of these reactions and shows that the DES radical can interact DES as theputative genotoxic agent. Since the generation of with other biological constituents. The observationthat pros- such a species during peroxidase-catalyzedoxidation of DES taglandin-H synthase catalyzes the metabolism of DES to has never been shown, we investigated whether peroxidatic free radical species maybe particularly relevant as prosta- oxidation of DES occurred via the generation of free radical glandin-H synthase is present in many organs that are sus- intermediates. ceptible to DES-induced carcinogenesis (17,18). Our results show that both horseradish peroxidase- and In summary,our data showthat radical speciesare produced prostaglandin-H synthase-catalyzed oxidation of DES cause during peroxidatic oxidation of DES andtherefore could bea the formation of a radical species that can be trapped with determinant of the carcinogenic potency of DES either di- DMPO. The identity of this radical speciescannot be deduced rectly or as a result of further reactions such as redox cycling from our data,but the fact that 1-naphthol produces an (32) or lipid peroxidation processes (33). As one free hydroxy identical species suggeststhat theradical is eithera phenoxy group has been shown to be sufficient for the induction of radical or is derived from a phenoxy radical. The hyperfine DNA binding by DES analogsit seemsprobablethat phenoxy coupling constants observed in this study are substantially radicals or their rearrangement products can bind to DNA. smaller than constants that have been published for DMPO- This is not likely to be the only mechanism eliciting DNA phenyl radical adducts (31).Thus this datais not conclusive binding, as peroxidatic oxidation also produces the genotoxic, proof that the6-line ESR signal obtained during horseradish two-electron oxidized product-DES quinone. Whether DES peroxidase- or prostaglandin-H synthase-catalyzed metabo- radicals or quinones are more importantfor the induction of lism of DES in the presence of DMPO is indicative of a genotoxic effects must await further work, but it seems plau- DMPO-DES adduct. Attempts to characterize the putative sible,as both speciesare formedduringperoxidaticoxidations, DMPO-DES adduct by mass spectrometry are in progress. that the mechanism of DES-induced genotoxicity could in- The generation of a thiyl radical during horseradish peroxi- volve both compounds. 16214 Peroxidase-catalyzed Metabolism of DES Acknowledgment-Wewould like to thank Michael Murphy for secretarial assistance. REFERENCES 1. International Agency for Research on Cancer (1979) Monographs on the Evaluation of the CarcinogenicRisk of Chemicalsto Humans Vol. 21, Sex Hormones (11), International Agency for Research on Cancer, Lyon, France 2. Greenwald, P., Barlow, J. J., Nasca, P. C., and Burnett, W. S. (1971) N. Engl. J.Med. 285,390-393 3. Herbst, A. L., Ulfelder, H., and Poskanzer, D. C. (1971) N. Engl. J.Med. 284,878-881 4. Bibbo, M., AI-Nageeb, M., Baccarini, I., Gill, W., Newton, M., Sleeper, K. M., Sonek, M., and Weid, G. L. (1975) J.Reprod. Med. 15,29-32 5. Cosgrove, M. D., Benton, B., and Henderson, B.E. (1977) J. Urol. 117,220-222 6. Gill, W. B., Schumacher, G. F. B.,and Bibbo, M. (1976) J.Reprod. Med. 16,147-153 7. ,McLachlan, J. A., Newbold, R. R., and Bullock (1975) Science 190,991-992 8. Kirkman, H. (1959) Natl. Cancer Inst. Monogr. 1,l-58 9. McLachlan, J. A., Wong, A., Degen, G.H., and Barrett, J. C. (1982) Cancer Res. 42,3040-3045 10. Liehr, J. G. (1983) Mol. Pharmacol. 23,278-281 11. Li, J. J., Li, S. A., Klicka, J. K., Parsons, J. A., and Lam, L. K. T. (1983) Cancer Res. 43,5200-5204 12. Metzler, M. (1981) CRC Crit. Rev. Biochem. 1 0 , 171-212 13. Metzler, M., and McLachlan, J. A. (1978) Biochem. P h a r m o l . 27,1087-1094 14. Metzler, M., and McLachlan, J. A. (1978) Biochem. Biophys. Res. Commun. 85,874-884 15. Liehr, J. G., DaGue, B. B.,Ballatore, A. M., and Henkin, J. (1983) Biochem. Pharmacol. 32,3711-3718 16. Metzler, M., and Epe, B. (1984) Chem.Biol. Interact. 50, 351- 360 17. Davis, B. B., Mattamal, M. B., and Zenser, T. V. (1981) Nephron 27,187-196 18. Abel, M. H., and Baird, D. T.(1980) Endocrinology 1 0 6 , 1599- 1606 19. Anderson, B., Larsson, R., Rahimtula, A., and Moldbus, P. (1983) Biochem. Pharmacol. 32,1045-1050 20. Saunders, B. C. (1973) in Inorganic Biochemistry (Eichorn, G. L., ed) Vol. 2, pp. 988-1021, Elsevier/North-Holland, New York 21. Moldeus, P., and Jernstrom, B. (1983) in Functions of Glutathi- one. Biochemical,Physiological,Toxicological and Clinical As- pects (Larsson, A., Orrhenius, s.,Holingren, A., and Manner- vik, B., eds) pp. 99-108, Raven Press, New York 22. Ross, D., Larsson, R., Anderson, B., Nilsson, U., Lindquist, T., Lindeke, B., and Moldeus, P. (1985) Biochem. Phurmacol. 3 4 , 343-351 23. West, P. R., Harman, L. S., Josephy, P. D., and Mason,. R. P. (1984) Biochem. Pharmacol. 33,2933-2936 24. Ross, D., Norbeck, K., and Moldius, P. (1985) J.Bwl. Chem., in press 25. Janzen, E. G., and I-Ping Liu, J. (1973) J.Magn. Reson. 9,510- 512 26. Saez, G.,Thornalley, P. J., Hill, H. A. O., Hems, R., and Bannis- ter, J. V. (1982) Biochim. Biophys.Acta 719,24-31 27. Harman, L. S., Mottley, C., and Mason, R. P. (1984) J. Biol. Chem. 259,5606-5611 28. Ross, D., Albano, E., Nilsson, U., and Moldbus, P. (1984) Biochem. Biophys. Res. Commun. 125,109-115 29. Metzler, M. (1984) in BiochemicalBasis of Chemical Carcinogen- esis (Greim, H., Jung, R. Kramer, M., Marquardt, H., and Oesch, F., eds) p. 69, Raven Press, New York 30. Maydl, R., Newbold, R.R., Metzler, M., and McLachlan, J. A. (1983) Endocrinology 113,146-151 31. Hill, H. A. O., and Thornally, P. J. (1981) FEBS Lett. 125,235- 238 32. Kappus, H., and Sies, H. (1981) Experientiu 37,1233-1241 33. Bus, J. S., and Gibson, J. E. (1979) Rev.Biochem. Toxicol. 1 , 125-149 Downloaded from www.jbc.org by guest, on April 1, 2010