BIOCHEMISTRY and MOLECULAR BIOLOGY INTERNATIONAL DIFFERENTIAL INHIBITION OF AFLATOXIN B1 OXIDATION BY GESTODENE ACTION ON HUMAN LIVER MICROSOMES

Transcription

1 Vol. 43, No. 4, November 1997 Pages DIFFERENTIAL INHIBITION OF AFLATOXIN B1 OXIDATION BY GESTODENE ACTION ON HUMAN LIVER MICROSOMES Bok Ryang Kim,* Hyun Sook Oh,** and Dong-Hyun Kim ~t'l *Department of Biochemistry, School of Medicine, Won Kwang University, Iksan **Department of Clinical Pathology, Won Kwang Junior College, Iksan, and rdoping Control Center, Korea Institute of Science and Technology, SeouL Korea Received July 8, 1997 Received after revision September 9, 1997 SUMMARY Human cytochrome P45 (P45) 3A is known to be involved in the formation of both aflatoxin Bl-exo-8,9-epoxide (exo-epoxidation) and aflatoxin Ql (3~hydroxylation). Gestodene, a known inactivator of P45 3A4, inhibited the formation of AFBt metabolites in a variety of ways depending on the incubation condition. Preincubation of gestodene with human liver microsomes prior to the addition of AFB~ inhibred both exo-epoxidation and 3~-hydroxylation whereas simultaneous incubation of gestodene with AFBt only inhibited 3r These results suggest that two independent substrate binding sites exist in P45 3A4, and AFB~ binds to both of the binding sites. Gestodene selectively binds to one of the binding sites leading to the formation of AFQ1. whereas it does not affect the formation of exo-epoxide via the other binding site. Key words: Aflatoxin Bb Gestodene, P45 3A4, Inhibition INTRODUCTION Cytochrome P45 (P45) enzymes play a major role in the oxidation of a wide variety of chemicals including drugs, environmental contaminants, and endogenous compounds (1,2). P45 consists of multiple forms and each displays unique substrate and catalytic activity profiles. Among them, P45 3A4 is of special interest because this enzyme metabolizes structurally diverse substrates ranging from small molecules to cyclosporin (3,4), and its activity can be enhanced in vitro by a number of compounds (5-7). P45 3A4 enzyme activities may be directly activated by 7,8-benzoflavone (6,7) whereas some reactions such as Abbreviations used are: P45, cytochrome P45; AFBI, aflatoxin B6 AFQI, aflatoxin Q1; AFM~, aflatoxin M1. 1To whom correspondence should be addressed /97/4839~855./ Copyright by Academic Press Australi~t. All rights of reproduction in any form reserved.

2 alfentanil hydroxylation are inhibited (8,9). 7,8-Benzoflavone itself may serve as a substrate for P45 3A4 (1). These phenomena may be due to the multiple substrate binding sites in the active site of the enzyme or the existence of different protein conformations. Recently, Koley et al. (11) demonstrated that P45 3A4 is composed of multiple kinetically distinguishable conformers using a CO binding kinetic analysis. Moreover, studies by (12) dearly showed direct evidence for the simultaneous binding of two substrates in the P45 3A4 active site by a kinetic analysis of 7,8-benzoflavone and phenanthrene metabolism. In the present investigation we report additional evidence of two substrate binding sites in the P45 3A4 active center by showing the differential effects of gestodene on the inhibition kinetics of AFB1-8,9-epoxidation and 3ct-hydroxylation. MATERIALS AND METHODS.Chemicals. AFB1, troleandomycin, glucose-6-phosphate, NADP+, glutathione and glucose- 6-phosphate dehydrogenase were purchased from Sigma Co. (St. Louis, MO). HPLC solvent was obtained from Merck Co. (Darmstadt, Germany). Gestodene was kindly donated by Dr. F.P. Guengenich (Vanderbilt University, USA). All other reagents were of the highest grade commercially available. Preparation of human liver mierosomes. The human liver sample IL 11) was donated by Dr. F.P. Guengerich (Vanderbilt) and the microsomal fraction was prepared using the method described of (13). The content of P45 was estimated as.87 nmol/mg protein as measured by the method of(14). AFBI metabolism. Measurement of AFB1 metabolism was made using the method described in (15). Briefly, incubation mixtures consisted of 5 mm potassium phosphate buffer (ph 7.5) containing 2 mm glutathione,.6 mg mouse liver cytosol, 5 ktm AFBt, 5 mm glucose 6- phosphate, 1. mm NADP ~, and 1. IU glucose 6-phosphate dehydrogenase. The mixtures were incubated for 1 min at 37 ~ and the reaction was terminated by the addition of 36 p/of formic acid. The reactants were centrifuged at 3 rpm for 1 min and the supernatant fraction was taken for analysis. Treatment of Gestodene. Treatments with gestodene were carried out in two different ways. In the preincubation experiment, gestodene was incubated with microsomes in the presence of an NADPH generating system for 3 rain prior to the addition of AFBt. In the other type of experiment, gestodene was simultaneously added to the incubation mixture with AFB~. Chromatography. Exo-epoxide and AFQ~ were quantified by HPLC with UV detection according to the modified procedure of (7). The supernatants were loaded onto the octadecylsilyl (C18) column (4.6 x 25cm, 5 #m, Beckman) with a mobile phase of solvent A (2 mm ammonium acetate, ph 5.4) containing 1% solvent B (acetonitrile: methanol: H2 = 4.5: 4.5:1, v/v/v). Elution was achieved using a gradient of 1-7 % solvent B over 16 min at a flow rate of 1.5 me/min. The eluate was monitored at 362 nm. Quantitation of exoepoxide-glutathione conjugate and AFQ1 was made by measuring the peak intensity and making comparisons with the response of an external standard. 84

3 RESULTS AFBt metabolism by human liver microsomes. The incubation of AFBI with HL 11 microsomes only generated exo-epoxide and AFQx at the ratio of 2.4. AFM1 that is predominantly generated by P45 1A2 (16,17) was not detected under the present experimental conditions (Fig. 1). The product ratio by the microsomes is consistent with the value range of two to three found for the reconstituted P45 3A4 system (18), and the failure to detect AFM1 suggests that AFB~ metabolism in the incubation system used here was predominantly mediated by P45 3A4. Inhibition of AFB1 metabolism by gestodene. Gestodene is known to be a selective inhibitor of P45 3A4 through the inactivation of heme (19). Preincubation of gestodene with microsomes for 3 min inhibited both AFB~-8,9-epoxidation and 3 ~z -hydroxylation in a dose dependent manner (Fig. 2A). The degree of inhibition was linear up to 2 I.tM gestodene and saturated at concentrations above 5 I-tM. The simultaneous addition of gestodene with AFBI to the microsomes exhibited a different inhibition pattern (Fig 2B). Little inhibition was observed in the case of 3 a -hydroxylation up to t [am gestodene. On the other hand, AFBl-8,9-expoxidation was inhibited in a dose dependent manner within the concentration range of 1-1 [am, suggesting that different inhibition mechanisms exist in AFB1 metabolism. In order to characterize the inhibition kinetics of gestodene, AFB~ metabolism was measured by varing the preincubation time of gestodene with the microsomes (Fig. 3). AFB~ exo-epoxidation was inhibited by 1% in the no preincubation condition whereas the 3 a-hydroxylation reaction was inhibited by 58%. AFBl-exo-epoxidation was gradually inhibited with increasing preincubation time, and the magnitude of the inhibition of exoepoxidation and 3 ct -hydroxylation became the same after 4 min of preincubation. DISCUSSION The P45 3A4 enzyme has complex substrate binding characteristics (2 and references therein). P45 3A4 plays a major role in both AFBt-8,9-epoxidation and 3~hydroxylation (7). The above two reactions were reported to be regulated differently by an in vitro activator and/or a selective inhibitor for P45 3A4 and were interpreted in terms of allosteric effects or different protein configurations (7). 841

4 BIOCHEMISTRYond MOLECULAR BIOLOGY INTERNATIONAL 23 3 A B 2 A 352 L., i,, I.... r,, j tr, min Fig. 1. HPLC analysis of the oxidation products of AFBI formed by human liver microsomes. Human liver microsomes were incubated with AFB1 and an NAPPH-generating system in the absence ofgestodene (A) or in the presence of 2 ~tm gestodene 3) for 1 rain at 37~ The indicated peaks were identified as the glutathione conjugate of AF.Bl-exo-8,9-epoxide (1), AFQ1 (2) and AFB1 (3) by comparison with authentic samples of these compounds. 842

5 BIOCHEMISTRYond MOLECULAR BIOLOGY INTERNATIONAL 1( A 1( c- O ~ t, t, t 2 ' t t, t, I I Gestodene (~tm) Fig. 2. Effect of gestodene on the oxidation of AFB1 by human liver microsomes. (A) gestodene was added to the reaction mixtures and preincubated for 3 min at 37~ prior to the addition of AFB1. (B) gestodene was immediately added to the reaction mixtures containing AFB1. The incubation for AFB~ oxidation was then carried out for 1 min at 37~ The formation of AFBl-exo-8,9-epoxide () and AFQi (Q)) were measured by HPLC. The results are shown as mean + SE for triplicate experiments. Several lines of evidence suggest that there exist at least two different binding sites in the active center of P45 3A4. 7,8-Benzoflavone stimulated several P45 3A4-mediated oxidative reactions such as testosterone-613-hydroxylation and AFBl-epoxidation (5-7). Some reactions such as alfentanil hydroxylation were reported to be inhibited by 7,8- benzoflavone (8,9), suggesting that different regulatory mechanisms acted on the P45 3A4 reactions. Recently, more direct evidence for the simultaneous binding of two substrates in the P45 3A4 active site was reported by (12). P45 3A4 catalyzes the oxidative metabolism of 7,8-benzoflavone and phenanthrene, and these two substrates affected the Vmax of each reaction without changing their respective Km values. This kinetic data may be interpreted as an instance of independent binding within the active site, because the compounds were shown not to be competitive inhibitions. Similar results were seen in the selective inhibition of AFBI metabolism by AFB2 and dehydronifedipine. AFB2, the saturated derivative of AFB1, preferentially inhibited the 8,9-epoxidation (9,2) while 843

6 BIOCHEMISTRYond MOLECULAR BIOLOGY INTERNATIONAL 1 I 1 I I 8 m i.. t- O o o~ 6 4o<- 2O I I I I, I + I Time, min Fig. 3. Inhibition of AFB~ oxidation activities of human liver microsomes as a function of preincubation time with gestodene. Human liver microsomes were incubated with 5 /~ M gestodene and an NADPH-generating system for the designated time at 37~ AFB~, glutathione and mouse liver cytosol were then added to all reaction mixtures and incubated for 1 rain at 37~ The formation of AFBl-exo-8,9-epoxide () and AFQ1 (O) were measured by HPLC. The results are shown as mean + SE for triplicate experiments. dehydronifedipine inhibited only the 3c~-hydroxylation reaction (15, 2). These reports collectively suggest that AFB~ can access the heme via two independent substrate binding sites and is converted to either AFBl-epoxide or AFQ~ depending on the binding pathway. The present investigation presents further evidence that P45 3A4 contains two independent substrate binding sites. Preincubation of gestodene in the presence of NADPH destroyed the heme of P45 3A4 and resulted in the inhibition of both 8,9-epoxidation and 3a-hydroxylation. Only the 3c~-hydroxylation was inhibited in the case of simultaneous treatment of gestodene with AFB1. It is well known that the addition of a competitive inhibitor to the reaction mixture protects the mechanism-based inactivator-induced loss of enzyme activity by the preoccupation of the binding site. The results presented here suggest that gestodene competed with AFBx for the binding site which leads to the 3c~hydroxylation of AFB1 and thus decreased the inactivation rate of P45 3A4, resulting in 844

7 partial protection of the exo-epoxidation activity which occurs through an independent binding pathway. competition with gestodene. gestodene-mediated The inhibition of 3ot-hydroxylation activity appears to be due to the Additional evidence for this proposition is discrepancy in inhibition of AFB, metabolism which gradually disappeared by increasing the preincubation time of gestodene with the microsomes. The protection of 8,9-epoxidation activity gradually decreased with increasing preincubation time and was not noted after 2 rain of preincubation. This incubation time profile correlates with the inactivation profile of P45 3A4-mediated nifedipine oxidation previously reported by (19)_ Troleandomycin, another mechanism-based inactivator, inhibited both epoxidation and 3ot- hydroxylation in the same ratio regardless of the incubation conditions (15), suggesting that the inactivation kinetics of troleandomycin may be different to that of gestodene. In conclusion, further evidence supporting the proposition that two independent substrate binding sites exist in P45 3A4. This was shown by the differential inhibition of AFB1 8,9-epoxidation and 3ot-hydroxylation by gestodene in human liver microsomes. AFB1 appears to bind to both sites in P45 3A4, leading to AFQ1 and 8,9-epoxide formation, and gestodene selectively binds to the site which gives rise to the formation of AFQ1. ACKNOWLEDGEMENTS This work was supported in part by Grants from Won Kwang University and Grants from the Ministry of Health and Social Affairs REFERENCES 1. Porter, T. D. and Coon, M. J. (1991) J. Biol. Chem. 266, Guengerich, F. P. (1995) in Cytochrome P45 (Ortiz de Montellano, P. R., Ed.) 2nd ed. pp , Plenum Press, New York. 3. Wrighton, S. A., Maural, P., Scheutz, E. G., Watkins, P B., Young, B., and Guezelian, P (t 985) Biochemistry 24, , Brian, W. R., Sari, M,, Iwasaki, M., Shimada, T., Kaminsky, L. S., and Guengerich, F. P. (199) Biochemistry 29, Buening, M. K., Fortner, J. G., Kappas, A. and Conney, A. H. (1978) Biochem. Biophys. Res. Commun. 82, Lundgren, K., Andries, M., Thompson, C., and Lucier, G.W. (1987) Cancer Res. 49, Raney,.K.D., Shimada, T., Kim, D. H., Groopman, J. D., Harris, T. M. and Guengerich, F. P. (1992) Chem. Res. Ibxicol. 5, Yun, C. H., Wood, M., Wood, A. J. J., and Guengerich, F.P. (1992)Anesthesiology 77, Labroo, R. B., Thummel, K. E., Kunze, K. L., Podoll, T., Trager, W. F., and Kharasch, E. D. (1995) DrugMetab. Dispos. 23,

8 1. Lee, H. S., Jin, C. B., Chong, H. S., Yun, C. H., Park, J. S., and Kim, D. H. (1994) Xenobiotica 24, Koley, A. P., Buters, J. T. M., Robinson, R. C., Markowitz, A., and Friedman, F. K. (1995) J. Biol. Chem. 27, Shou, M., Grogan, J., Mancewicz, J. A., Krausz, K. W., Gonzalez, F. J., Gelboin, H. V., and Korzekwa, K. R. (1994) Biochemistry 33, Lindstrom, T. D., Hanssen, B. R., and Wrighton, S. A (1993) Antimicrobial Agents and Chemotherapy 37, Omura, T. and Sato, R. (1964) J. Biol. Chem. 239, Kim, B. R, Oh, H. S., and Kim, H. J. (1995) Korean J. Toxicology 11, Guo, Z., Gillam, E M. J., Ohmori, S., Tukey, R.H. and Guengerich, F. P. (1994)Arch, Biochem. Biophys. 312, Ueng, Y.-F., Shimada, T., Yamazaki, H., and Guengerich, F. P. (1995) Chem. Res. Toxicol. 8, , Gillam, E M. J., Baba, T., Kim, B. R., Ohmori, S. and Guengerich, F. P. (1993) Arch. Biochem. Biophys. 35, Guengerich, F. P. (199) Chem. Res. Toxicol, 3, Guengerich, F. P., Kim, B. R., Gillam, E. M. J., and Shimada, T. (1994) in Proceedings, 8th Int. conference on cytochrome P45: Biochemistry, Biophysics and Molecular Biology (Lechner M.C., ed.), pp , John Libbey Eurotext, Chichester, U.K. 846