722
Chem. Res. Toxicol. 1996, 9, 722-728
The Effect of Fluoro Substituents on Reactivity of 7-Methylbenz[a]anthracene Diol Epoxides Wanda Baer-Dubowska,† Raghunathan V. Nair,† Adam Dubowski,† Ronald G. Harvey,‡ Cecilia Cortez,‡ and John DiGiovanni*,† Science Park-Research Division, The University of Texas M. D. Anderson Cancer Center, P.O. Box 389, Smithville, Texas 78957, and The Ben May Institute, University of Chicago, Chicago, Illinois 60637 Received December 11, 1995X
The present study has examined potential mechanisms for the influence of F-substituents on the biologic activity of methylbenz[a]anthracenes. DNA adducts derived from reaction of the racemic bay-region anti-diol epoxides of 7-methylbenz[a]anthracene, and its 9- and 10fluoro derivatives, with calf thymus DNA in vitro were partially characterized. All three hydrocarbon diol epoxides produced similar DNA adduct profiles upon reaction with calf thymus DNA in vitro that were composed of two deoxyguanosine and two deoxyadenosine adducts (tentatively identified as trans addition products). The extent of covalent binding to calf thymus DNA, as estimated by 32P-postlabeling, was similar for all three diol epoxides. The reactivity of the unsubstituted and 10-F-substituted diol epoxide was further assessed by measuring overall pseudo-first-order rate constants for hydrolysis in water or 0.1 M Tris-HCl buffer, pH 7.0, and in the presence or absence of native or denatured DNA. The rate constant for hydrolysis of 7-methylbenz[a]anthracene diol epoxide in the absence of DNA was similar to that of 10-F-7-methylbenz[a]anthracene diol epoxide (t1/2 ) 138 min vs 115 min in water, respectively, and 93 vs 83 min in 0.1 M Tris-HCl buffer, respectively). In addition, the presence of DNA accelerated hydrolysis rates to similar extents for both diol epoxides. The skin tumorinitiating activities of the 9- and 10-F-substituted 3,4-diols of 7-methyl-, 12-methyl-, and 7,12dimethylbenz[a]anthracene were determined in SENCAR mice. The presence of F-substituents in the 9- or 10-position did not enhance or in some cases reduced the tumor-initiating activity of the 3,4-diols of these hydrocarbons. Collectively, these results, as well as previous results from our laboratory, suggest that the influence of a F-substituent at position 10 of the benz[a]anthracene nucleus is not due to increased or altered reactivity of the bay-region diol epoxide but rather likely on the initial formation of the 3,4-diol.
Introduction Polycyclic aromatic hydrocarbons (PAH)1 are widely distributed environmental pollutants, which are carcinogenic only after metabolism to reactive intermediates (1). For most PAH, the ultimate carcinogenic metabolites are vicinal diol epoxides (DEs), most frequently of the “bay-region” type (1-3). Mouse epidermis is a target tissue for PAH-induced carcinogenesis. Among the benz[a]anthracenes (BAs), the unsubstituted parent PAH has only weak activity as a tumor initiator in mouse skin (4), while analogs with methyl substituents at positions 7,12 or both are considerably more active. Of the 12 possible * To whom correspondence should be addressed. † The University of Texas M. D. Anderson Cancer Center. ‡ The Ben May Institute. X Abstract published in Advance ACS Abstracts, April 15, 1996. 1 Abbreviations: PAH, polycyclic aromatic hydrocarbons; BA, benz[a]anthracene; MBA, methylbenz[a]anthracene; 7-MBA, 7-methylbenz[a]anthracene; 12-MBA, 12-methylbenz[a]anthracene; DMBA, 7,12dimethylbenz[a]anthracene; F, fluorine; 9-F-7-MBA, 9-fluoro-7-methylbenz[a]anthracene; 10-F-7-MBA, 10-fluoro-7-methylbenz[a]anthracene; B[a]P, benzo[a]pyrene; DE, diol epoxide; BADE, (()-trans-3,4-dihydro3,4-dihydroxy-anti-1,2-epoxy-1,2,3,4-tetrahydrobenz[a]anthracene; 7-MBADE, (()-trans-3,4-dihydro-3,4-dihydroxy-anti-1,2-epoxy-1,2,3,4tetrahydro-7-methylbenz[a]anthracene; 9-F-7-BADE, (()-trans-3,4dihydro-3,4-dihydroxy-anti-1,2-epoxy-1,2,3,4-tetrahydro-9-F-7-methylbenz[a]anthracene; 10-F-7-MBADE, (()-trans-3,4-dihydro-3,4,dihydroxy-anti-1,2-epoxy-1,2,3,4-tetrahydro-10-F-7-methylbenz[a]anthracene; DB[a,j]A, dibenz[a,j]anthracene; B[a]PDE, (()-trans-7,8-dihydro7,8-dihydroxy-anti-9,10-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene; 5-MeC, 5-methylchrysene; B[c]Ph, benzo[c]phenanthrene; dGuo, deoxyguanosine; dAdo, deoxyadenosine; 2-ME, 2-mercaptoethanol; TPA, 12-Otetradecanoylphorbol 13-acetate.
S0893-228x(95)00207-4 CCC: $12.00
monomethylbenz[a]anthracenes, 7-methylbenz[a]anthracene (7-MBA) is the most potent carcinogen in mouse skin and other tissues (5-8), and the dimethyl derivative, 7,12-dimethylbenz[a]anthracene (DMBA), is one of the most potent skin carcinogens known (9). As a group, these three compounds differ greatly in their biological activity, but they are all thought to be activated, at least in part, via structurally similar “bay-region” DEs, of the type exemplified by trans-3,4-dihydro-3,4-dihydroxy-1,2epoxy-1,2,3,4-tetrahydro-3,4-benz[a]anthracene (BADE). Replacement of hydrogens by fluorine (F) in PAH is known to alter their metabolism, DNA binding, and carcinogenicity (10-13). Introduction of an F atom on the A-ring of DMBA, for example (i.e., positions 1, 2, 3, or 4), dramatically reduces tumor-initiating activity. This effect of a bay-region benzo-ring substituent in DMBA and other PAH is believed to be due primarily to blocking of metabolic activation to the corresponding bay-region DEs (reviewed in refs 14 and 15). An F atom at position C5 of DMBA or C11 of 5-methylchrysene (5-MeC), referred to as the “peri” position, also significantly reduces tumorinitiating activity of the parent compound (reviewed in refs 14 and 15). In contrast, derivatives with F-substituents in positions farther removed from the bay region of various PAH usually possess significant tumor-initiating and mutagenic activity approximately equal to that of the parent molecule (16-19). However, we have shown that an F-substituent in position 10 of DMBA and 7-MBA markedly enhanced both tumor-initiating activity © 1996 American Chemical Society
F-Substituents and Reactivity of Diol Epoxides
in mouse skin and mutagenic activity in a human hepatoma (HepG2) cell-mediated assay (20-22). The present study was designed to further explore the mechanism for enhancement of tumor-initiating activity in BAs resulting from a 10-F-substituent. To achieve this goal, we performed the following types of experiments: (i) comparison of the type and levels of DNA adducts formed following reaction of (()-trans-3,4-dihydro-3,4dihydroxy-anti-1,2-epoxy-1,2,3,4-tetrahydro-7-methylbenz[a]anthracene (anti-7-MBADE) and 9- and 10-F-7MBADE with calf thymus DNA; (ii) comparison of the hydrolysis rates of the same DEs in aqueous solution; and (iii) comparison of the tumor-initiating activity of the 3,4-diols derived from 7-MBA, 9-F-7-MBA, and 10-F-7MBA, as well as the corresponding analogs from 12-MBA and 7,12-DMBA. On the basis of our results, we conclude that the presence of an F atom at position 10 of the BA nucleus does not significantly affect reactivity of the bayregion anti-diol epoxide but rather hypothesize that the enhancing effect on tumor initiation is due to enhanced metabolic formation of the bay region 3,4-diol.
Materials and Methods Chemicals. Caution: The BA derivatives described within this paper have been determined to be carcinogenic to laboratory animals. Hence, protective clothing and appropriate safety procedures should be followed when working with these compounds. 12-O-Tetradecanoylphorbol 13-acetate (TPA) was obtained from LC Services (Woburn, NY). Sodium salts of 2′-deoxyguanosine 5′-monophosphate and 2′-deoxyadenosine 5′-monophosphate, calf thymus DNA, DNase (bovine pancreas, EC 3.1.21.1), snake venom phosphodiesterase (Crotalus atrox, EC 3.1.4.1), and Escherichia coli alkaline phosphatase (type III, EC 3.1.3.1) were purchased from Sigma Co. (St. Louis, MO). [3H]dGuo and [3H]dAdo having specific activities of 10-22 Ci/mmol were obtained from ICN Laboratories (Irvine, CA). Racemic 3,4-diols of 7-MBA, 12-MBA, DMBA, and their 9-F and 10-F derivatives were synthesized as previously described (23). Racemic antiDEs of 7-MBA, 9-F-7-MBA, and 10-F-7-MBA were synthesized by modifications of previous methods (24). Instrumentation. Ultraviolet absorption spectra were obtained in CH3OH on a Shimadzu UV-160 UV-visible recording spectrophotometer. Proton NMR spectra were obtained in CD3OD on a General Electric GN-500 spectrometer at the University of Texas in Austin. HPLC separations were carried out by using a Shimadzu LC-6A equipped with Shimadzu SCL-6A system controller, SPD-6A UV detector, and CR-501 data processor. DNA Binding. Calf thymus DNA labeled with [3H]dGuo or [3H]dAdo was prepared as previously described (25) by using a modification of the method supplied with the nick translation kit from Bethesda Research Laboratories, Gaithersburg, MD. Labeled (413 µg) and unlabeled (1 mg) calf thymus DNA was reacted with DEs (100 µL of a 1 mg/mL solution in anhydrous acetone) in 0.01 M Tris-HCl buffer (pH 7.0) at 37 °C for 16 h. DNA was subsequently precipitated, washed, dissolved in 0.01 M Tris-MgCl2 buffer (pH 7.0), and hydrolyzed to deoxyribonucleosides as previously described (26). The hydrolysates were purified on Sephadex LH-20 columns according to published procedures (26) to afford the modified deoxyribonucleosides. The relative extent of binding of DEs with DNA was measured by 32P-postlabeling using a 1-butanol enrichment procedure as described previously (27). Initial experiments were performed to determine that the amount of [γ-32P]ATP was not rate limiting in the detection of total adduct levels from the DEs used in the current study (data not shown). Synthesis of 7-MBA and 9-F- and 10-F-7-MBA-anti-DE Adducts. anti-DE DNA adducts were prepared essentially as described previously (25). Briefly, to a solution of 2′-deoxyguanosine 5′-monophosphate or 2′-deoxyadenosine 5′-monophos-
Chem. Res. Toxicol., Vol. 9, No. 4, 1996 723 phate sodium salt (1 g) in 50 mM Tris-HCl buffer (50 mL, pH 7.0) was added a solution of the appropriate racemic DE (3 mg, in 6 mL of acetone). The mixture was slowly shaken for 20 h in a water bath maintained at 37 °C. The reaction mixture was then extracted with ethyl acetate and ether. The combined organic extracts were evaporated under reduced pressure to afford a mixture of the respective isomeric tetrols. Following the removal of the trace of organic solvents, the diluted aqueous phase was digested overnight at 37 °C with E. coli alkaline phosphatase and purified on Sephadex LH-20 columns. Methanol phases containing DNA adducts were analyzed by HPLC. Analytical HPLC of the adducts was performed on an Ultrasphere ODS column (4.6 mm × 25 cm) using a linear gradient of 21-25% acetonitrile in water over 60 min at 0.7 mL/min (dGuo adducts) or a linear gradient of 21-25% acetonitrile in water followed by a 20 min hold at 25% acetonitrile in water and then a linear gradient of 25-100% acetonitrile in water over 20 min (dAdo adducts). DE Hydrolysis Rates. Stock solutions of (()-anti-7-MBADE and (()-anti-10-F-7-MBADE were prepared by dissolving in anhydrous acetone at a concentration of 2.6 mM. Native calf thymus DNA solution was prepared in 0.01 M Tris-HCl buffer (pH 7.0). Denatured DNA solution was prepared from native calf thymus DNA by heating it in boiling water for 20 min and then rapidly cooling it in an ice bath. UV absorbance enhancement of 43% at 260 nm was observed. The kinetics of the hydrolysis of DEs were monitored by HPLC after trapping unreacted epoxides with 2-mercaptoethanol (2-ME) (28). The reactions were initiated by the addition of 50 µL of the stock solution of DE to 10 mL of water (pH 7.0) or 0.01 M Tris buffer (pH 7.0) in the presence or absence of native or denatured DNA. In addition, rate constants were determined in Tris-HCl buffer plus native DNA with 0.1 M NaCl. The reactions were carried out at 25 ( 1 °C. At different time intervals, 1 mL aliquots were removed from the reaction mixture and immediately quenched with 0.1 mL of 2-ME solution (29). Each aliquot was extracted with 3 × 1 mL of ethyl acetate. The organic solvent was removed, and the residue was dissolved in methanol and analyzed by HPLC using a Beckman ODS column (4.6 × 250 mm) with the following gradient: 30-60% methanol in water over 90 min, followed by 60-100% methanol in water over 10 min at a flow rate of 1.0 mL/min. Products were detected by UV absorption at 254 nm. The pseudo-first-order rate constants for the disappearance of (()-anti-DEs were determined by plotting the negative logarithm of residual DE measured as the thioether adduct versus time. The rate constants were determined from the least-squares linear fit slopes. Tumor Induction Experiments. Female SENCAR mice were obtained from the National Cancer Institute (Frederick, MD), and when 7-9 weeks of age, shaved on the dorsal side. Mice were allowed to stabilize for at least 2 days, and only those mice in the resting phase of the hair growth cycle were subsequently used. All chemicals were applied topically to the shaved area in 0.2 mL of acetone, and control animals were treated with an equal volume of acetone. Each experimental group contained 25 preshaved mice. Mice were initiated with the 3,4-diols of 7-MBA, 12-MBA, DMBA, and their 9- and 10-F derivatives at the doses indicated. Two weeks after initiation, mice began receiving twice-weekly treatments with 3.4 nmol of TPA. The incidence and number of papillomas was observed and recorded weekly. Promotion was continued in all groups until the average number of papillomas per mouse reached a plateau. Statistical analyses of the difference between mean papilloma responses (i.e., papillomas per mouse) were performed using the Mann-Whitney U-test. The level of significance was set at p < 0.05.
Results DNA Adducts Derived from (()-anti-7-MBADE and its 9-F and 10-F Derivatives. The reaction of (()anti-7-MBADE and its 9-F and 10-F derivatives with calf
724 Chem. Res. Toxicol., Vol. 9, No. 4, 1996
Baer-Dubowska et al.
Figure 2. Rates of (()-anti-7-MBADE (9), and (()-anti-10-F7-MBADE ()) hydrolysis in water (means from two experiments).
Figure 1. HPLC elution profiles of deoxyribonucleoside adducts obtained from reactions of (()-trans-3,4-dihydro-3,4dihydroxy-anti-1,2-epoxy-1,2,3,4-tetrahydro-7-methylbenz[a]anthracene (panel A) and its 9-F (panel B) and 10-F (panel C) derivatives with calf thymus DNA. The adducts were eluted on an Ultrasphere ODS column (4.6 mm × 25 cm) with 47% CH3OH for 50 min followed by sequential linear gradients of 4760% CH3OH for 50 min and 60-100% CH3OH for 15 min. The identity of deoxyguanosine (dGuo) and deoxyadenosine (dAdo) adduct peaks was determined by measuring the radioactivity in HPLC fractions obtained from analysis of separate reactions of tritium-labeled DNA with each epoxide. Peaks 1, 1′, and 1′′ and 2, 2′, and 2′′ are derived from reaction with dGuo, whereas peaks 3, 3′, and 3′′ and 4, 4′, and 4′′ are derived from reaction with dAdo.
thymus DNA in vitro followed by enzymatic degradation and isolation of modified deoxyribonucleosides provided the HPLC profiles shown in Figure 1 (panels A-C). Incorporation of tritiated dGuo and dAdo residues into DNA in separate experiments by utilizing a nick translation procedure prior to reaction with diol epoxides unambiguously identified peaks 1,2, 1′,2′, and 1′′,2′′ as containing dGuo residues, whereas peaks 3,4, 3′,4′, and 3′′,4′′ were identified as containing dAdo residues. All three DEs afforded qualitatively similar adduct profiles upon reaction with calf thymus DNA in vitro. In this regard, all three DEs reacted more extensively with dGuo than dAdo residues. The extent of binding as estimated by the 32P-postlabeling method was 1.58 ( 0.29, 1.74 ( 0.10, and 1.40 ( 0.13 pmol/µg of DNA for 7-MBADE, 9-F7-MBADE, and 10-F-7-MBADE, respectively. These values were not significantly different (p > 0.05). Thus, the presence of an F-substituent at positions 9 or 10 of 7-MBADE did not appear to influence the extent of the reaction with calf thymus DNA in vitro. Quantities of peaks 3′ and 3′′ sufficient for structural characterization were prepared from reaction of DEs of 7-MBA and its F-substituted derivatives as described under Materials and Methods. 1H-NMR spectra were obtained only for these peaks since the main difference
in the three DNA adduct profiles shown in Figure 1 was in retention time of the adducts derived from the F derivatives. Previous experiments analyzing the adducts obtained from DMBA, 9-F-DMBA, and 10-F-DMBA in mouse epidermis in vivo had shown similar shifts in retention times for adducts derived from the F derivatives (30). Based on the J3,4 coupling exhibited by the H4 resonance signal as described previously for dibenz[a,j]anthracene (DB[a,j]A) adducts (25, 31), these adducts can be assigned as derived from trans addition of dAdo to the respective DEs [9-F-7-MBA-dAdo3′: δH4 ) 4.96, J3,4 ) 8.41 Hz; 10-F-7-MBA-dAdo3′′: δH4 ) 4.96, J3,4 ) 8.73 Hz]. Based upon analogy with the previously reported DB[a,j]A-dAdo1 adduct and taking into consideration the data of Peltonen (32), the following tentative structures (except 3′ and 3′′ which were confirmed) have been assigned: peaks 1, 1′, and 1′′ are derived from (-)-antitrans-N2-dGuo, peaks 2, 2′, and 2′′ are derived from (+)anti-trans-N2-dGuo, peaks 3, 3′′, and 3′′ are derived from (+)-anti-trans-N6-dAdo, and peaks 4, 4′, and 4′′ are derived from (-)-anti-trans-N6-dAdo. Based on the previous report of Peltonen et al. (32), adducts derived from cis addition of the (+) or (-)-anti-DEs of these closely related PAH analogs with calf thymus DNA would have been formed at very low levels under the current experimental conditions. Rates of Hydrolysis of DEs of 7-MBA and 10-F-7MBA. The rates of hydrolysis of 7-MBADE and 10-F-7MBADE were monitored by HPLC after trapping unreacted DEs with 2-ME (28). The half-lives (t1/2) in solution for each DE were estimated from the time-dependent decrease in formation of their respective 2-ME adducts. Figure 2 shows a plot of the log of the peak areas of the 2-ME adduct versus reaction time for an assay performed in water (pH 7.0). The hydrolysis of both DEs followed pseudo-first-order kinetics with rate constants of 0.883 and 1.000 × 10-4 s-1, for the 7-MBADE and 10-F-7MBADE, respectively. Table 1 summarizes the results of kinetic measurements for hydrolysis of DEs in TrisHCl buffer carried out in the presence of native or denatured calf thymus DNA, and in the presence of native calf thymus DNA with 0.1 M NaCl. As shown in Table 1, the rates of disappearance of the DEs in the presence of native DNA were increased ∼3 and 16 times for DNA concentrations of 15 and 30 µg/mL, respectively, as compared to the rate in Tris-HCl (pH 7.0) buffer alone. The presence of denatured DNA or addition of 0.1 M NaCl to native DNA reduced the rate constants but not
F-Substituents and Reactivity of Diol Epoxides
Chem. Res. Toxicol., Vol. 9, No. 4, 1996 725
Table 1. Pseudo-First-Order Rate Constants for Hydrolysis of (()-anti-DEs of 7-MBA and 10-F-7-MBA solution
addition
H2O 0.01 M Tris, pH 7.0 0.01 M Tris, pH 7.0
none none denatured DNA: 15 µg/mL 30 µg/mL native DNA: 15 µg/mL 30 µg/mL 15 µg/mL + 0.1 M NaCl
0.01 M Tris, pH 7.0
Table 2. Tumor-Initiating Activity of 7-MBA-, 12-MBA-, and 7,12-DMBA-3,4-diols and Their 9- and 10-F Derivatives in Female SENCAR Micea initiator acetone 7-MBA-3,4-diol 9-F-7-MBA-3,4-diol 10-F-7MBA-3,4-diol 12-MBA-3,4-diol 9-F-12-MBA-3,4-diol 10-F-12-MBA-3,4-diol DMBA-3,4-diol 9-F-DMBA-3,4-diol 10-F-DMBA-3,4-diol
dose (nmol)
papillomas per mouseb
% of mice with papillomas
0.2 200 400 200 400 200 400
0.0 3.96 ( 0.59 5.08 ( 0.56 2.68 ( 0.48 3.04 ( 0.95c 4.52 ( 0.55 5.33 ( 0.68
0.0 88.0 95.8 76.0 70.8 100.0 96.0
200 400 200 400 200 400
16.80 ( 2.20 22.60 ( 2.23 11.96 ( 1.38 11.20 ( 1.38c 22.68 ( 2.37 19.88 ( 2.33
100.0 100.0 96.0 100.0 100.0 100.0
4 10 4 10 4 10
21.44 ( 2.27 23.80 ( 2.17 11.44 ( 1.49c 20.16 ( 2.42 20.92 ( 2.42 13.60 ( 1.77c
100.0 100.0 100.0 100.0 95.8 97.7
a Twenty-five female SENCAR mice were used for each treatment group. Animals were initiated and 2 weeks later received twice-weekly applications of 3.4 nmol of TPA for 19 weeks. b Average number of papillomas per mouse ((SEM) after 19 weeks of promotion. c Significantly different from parent 3,4-diol (p
