Comparative Metabolism of the Aza Polynuclear Aromatic

State University of New York College at Buffalo, 1300 Elmwood Avenue,. Buffalo, New York 14222. Received January 6, 2004. To assess the role of human ...
0 downloads 0 Views 81KB Size
Download PDF
672

Chem. Res. Toxicol. 2004, 17, 672-678

Comparative Metabolism of the Aza Polynuclear Aromatic Hydrocarbon Dibenz[a,h]acridine by Recombinant Human and Rat Cytochrome P450s Zhi-Xin Yuan, Subodh Kumar, and Harish C. Sikka* Environmental Toxicology and Chemistry Laboratory, Great Lakes Center, State University of New York College at Buffalo, 1300 Elmwood Avenue, Buffalo, New York 14222 Received January 6, 2004

To assess the role of human and rat cytochrome P450s in the metabolism of aza-polynuclear aromatic hydrocarbons (aza-PAHs) and to examine the influence of heterocyclic nitrogen on the metabolism of these chemicals, we have investigated the biotransformation of dibenz[a,h]acridine (DB[a,h]ACR), an aza-PAH with two nonidentical bay regions, by recombinant human cytochromes P450 1A1, 1B1, and 3A4 and rat P450 1A1. Among the three P450s, 1A1 was the most effective in metabolizing DB[a,h]ACR followed by 1B1 and 3A4. The major DB[a,h]ACR metabolites produced by human P450 1A1 and 1B1 were the dihydrodiols with a bay region double bond, namely, DB[a,h]ACR-3,4-diol and DB[a,h]ACR-10,11-diol (putative proximate carcinogen). P450 1A1 produced a higher proportion of DB[a,h]ACR-10,11-diol (derived from the benzo ring adjacent to the nitrogen) (44.7%) than of DB[a,h]ACR-3,4-diol (derived from benzo ring away from the nitrogen) (23.8%). In contrast, 1B1 produced a much greater proportion of 3,4-diol (54.7%) than of 10,11-diol (6.4%). These data indicate that (i) human P450 1A1 and 1B1 differ dramatically with respect to the regiospecific metabolism of DB[a,h]ACR, (ii) human P450 1A1 is substantially more active than human P450 1B1 in the metabolic activation of the aza-PAH to its 10,11-diol, and (iii) the presence of nitrogen influences the relative extent to which the two benzo ring diols with a bay region double bond are formed by human P450s 1A1 and 1B1. In contrast to human P450s 1A1 and 1B1, rat P450 1A1 showed no regioselectivity in the metabolism of DB[a,h]ACR producing nearly equal proportions of 10,11-diol and 3,4-diol. Despite significant differences in their regioselectivity, human P450 1A1 and 1B1 and rat P450 1A1 showed similar stereoselectivity in the metabolism of DB[a,h]ACR to its diols having a bay region double bond, producing primarily the R,R enantiomers (>94%). The data of these studies indicte that human and rat P450 1A1 differ in their regioselectivity in the metabolism of DB[a,h]ACR to its two benzo ring diols with a bay region double bond and consequently in their ability to metabolically activate the parent aza-PAH. However, human and rat P450 1A1 do not differ with respect to their stereoselectivity in the metabolism of DB[a,h]ACR to the diols.

Introduction The aza-PAHs,1 in which at least one aromatic carbon is substituted by an aza group, are commonly occurring environmental contaminants formed during the pyrolysis of nitrogen-containing organic materials (1, 2). These chemicals are found in significant amounts in tobacco smoke, gasoline engine exhaust, and effluents from coal combustion processes (3-7). This class of chemicals includes a number of dibenzacridines, including DB[a,h]ACR, DB[a,j]ACR, and dibenz[c,h]acridine (8). The concentration of dibenzacridines in cigarette smoke condensate ranges from 0.01 to 1 mg/100 cigarettes as compared to 0.5 mg B[a]P/100 cigarettes (4). Among the dibenzacridines tested, DB[a,h]ACR is a highly potent mutagen * To whom correspondence should be addressed. Tel: 716-878-5422. Fax: 716-878-5400. E-mail: [email protected]. 1 Abbreviations: PAHs, polynuclear aromatic hydrocarbons; azaPAHs, aza analogues of PAHs; DB[a,h]ACR, dibenz[a,h]acridine; DB[a,j]ACR, dibenz[a,j]acridine; B[a]P, benzo[a]pyrene; diol, dihydrodiol; DMSO, dimethyl sulfoxide; HPLC, high-pressure liquid chromatography.

(9) and carcinogen (10); it exhibits tumorigenic activity similar to that of its carbon analogue dibenz[a,h]anthracene (11). DB[a,h]ACR is more potent in inducing pulmonary adenomas than the well-studied carcinogenic PAH B[a]P (12). Aza-PAH, like PAHs, require metabolic activation by cytochrome P450-dependent monoxygenases and epoxide hydrolase to highly reactive bay region diol epoxides that react with cellular DNA to initiate cancer (13-17). Structurally, DB[a,h]ACR is unique among the abovementioned dibenzacridines in that, because of the presence of nitrogen heteroatom at position 7, it has two dissimilar bay regions in which diol epoxides could form. In one of the diol epoxides, nitrogen is located near the bay region while in the other, nitrogen is distant from the bay region. Our previous studies have shown that DB[a,h]ACR is metabolically activated via the formation of its 10,11-diol to the bay region DB[a,h]ACR 10,11-diol8,9-epoxide (9, 10, 17) (Figure 1). The bay region diol epoxide derived from the 8,9,10,11-benzo ring of DB[a,h]ACR is more mutagenic and tumorigenic as compared

10.1021/tx049979i CCC: $27.50 © 2004 American Chemical Society Published on Web 04/17/2004

Metabolism of DB[a,h]ACR by Human and Rat P450s

Chem. Res. Toxicol., Vol. 17, No. 5, 2004 673

the regio- and stereoselectivity of human P450 1A1 and P450 1B1 in the metabolism of DB[a,h]ACR, and (iii) compared the regio- and stereoselectivity of DB[a,h]ACR metabolism by human 1A1 and rat 1A1. The aim of the study was to assess directly the role of individual human P450s in the metabolism of DB[a,h]ACR and to provide support for the metabolic activation pathway of this carcinogenic aza-PAH in humans. Because of the presence of two dissimilar bay regions of DB[a,h]ACR, the study of the metabolism of the chemical will allow us to assess the effect of the nitrogen heteroatom on the regioand stereoselectivity of individual human P450s in the metabolism of DB[a,h]ACR.

Materials and Methods

Figure 1. Metabolic pathways for the formation of the major metabolites of DB[a,h]ACR by rat liver microsomes. The heavy arrows represent the pathway by which DB[a,h]ACR is converted into mutagenic and carcinogenic metabolites.

to the corresponding bay region diol epoxide derived from the 1,2,3,4-benzo ring of DB[a,h]ACR (9, 10). Furthermore, among the bay region diol epoxide isomers, only (+)-DB[a,h]ACR-10S,11R-diol-8R,9S-epoxide-2 and its precursor (-)-DB[a,h]ACR-10R,11R-diol displayed significant tumorigenic activity (10). These data indicate that both the regio- and the stereoselectivity of cytochrome P450s play an important role in the metabolic activation of DB[a,h]ACR. Although we have characterized the metabolic activation pathways of DB[a,h]ACR in rats (18, 19), it is difficult to extrapolate these data to humans because of notable differences among species with regard to the expression and catalytic activity of P450s. Therefore, it is important to assess directly the role of individual human P450 isoforms in the metabolism of chemical carcinogens in order to predict human response to a given carcinogen. Humans show considerable interindividual variations in the levels of expression of individual P450s and in their corresponding activities (20). Knowledge about which human P450(s) is involved in the metabolic activation of DB[a,h]ACR will be useful in assessing susceptibility of individuals to this potent carcinogen. Several P450s are known to be involved in the metabolic oxidation of PAHs (21-28). Among the various forms of P450 determined so far, P450 1A1 and P450 1B1 are known to be the most important human P450 enzymes in the metabolism of PAHs and PAH diols (2932). Both enzymes are expressed predominantly in the extrahepatic organs such as the lungs and mammary glands. Other P450 enzymes such as P450 3A4 may have some role in PAH metabolism. It is well-established that individual P450s differ with respect to overall substrate specificity and regio- and stereoselectivity for the metabolism of individual substrates that reflect different tertiary structure of the proteins (24, 33-37). In the present investigation, we have (i) determined the catalytic activity of human P450 1A1, P450 1B1, and P450 3A4 toward the metabolism of DB[a,h]ACR, (ii) examined

Materials. DB[a,h]ACR and [1,2-14C]DB[a,h]ACR were purchased from the Midwest Research Insititute (Kansas City, MO) and New England Nuclear (Boston, MA), respectively. Radiolabeled DB[a,h]ACR was purified to >98% purity by HPLC. DB[a,h]ACR-3,4-diol, 8,9-diol, 10,11-diol, and epoxides were synthesized as described previously (38, 39). The following were obtained from Gentest Corporation (Woburn, MA): (i) microsomes prepared from human lymphoblastoid cells in which human cytochromes P450 1A1, 1B1, and 3A4 and epoxide hydrolase were expressed and (ii) microsomes prepared from insect cells in which rat P450 1A1 was expressed. The P450 content of microsomes expressing their P450s was described in the data sheet provided by the vendor. Recombinant rat microsomal epoxide hydrolase was kindly provided by Dr. Bruce Hammock, University of California (Davis, CA). Metabolism of [14C]DB[a,h]ACR by Recombinant Cytochrome P450s. A typical 1 mL reaction mixture containing 50 pmol of human or rat P450, ∼100 pmol of human or rat epoxide hydrolase, 100 µM potassium phosphate buffer (pH 7.4), 2.5 µmol of MgCl2, and 1.0 µmol of NADPH was preincubated at 37 °C for 2 min. The reaction was initiated thereafter by the addition of [14C]DB[a,h]ACR (40 µM containing 2 µCi, dissolved in DMSO). An incubation mixture without NADPH served as control. After the mixture was incubated for 10 min with shaking, the extent of DB[a,h]ACR metabolism was determined as the amount of total metabolites formed according to the procedure of Van Cantfort et al. (40). The data were statistically analyzed by Newman-Keuls multiple comparison method after one way ANOVA. For analysis of DB[a,h]ACR metabolites, the reaction was termined by the addition of 1 mL of ice-cold acetone after 10 min of incubation. The incubation mixture was then extracted thrice with 2 vol of ethyl acetate. The ethyl acetate layers containing DB[a,h]ACR and its metabolites were pooled, dried over anhydrous sodium sulfate, taken to dryness at room temperature under a stream of nitrogen, and stored at -80 °C. Prior to HPLC analysis, the residue was dissolved in 100 µL of freshly distilled THF containing 0.1% NH4ΟΗ. DB[a,h]ACR, its metabolites, and the synthetic reference standards were resolved on a Hewlett-Packard 5000 highpressure liquid chromatograph equipped with a solvent delivery system, solvent programmer, and a diode array detector. The metabolites were separated using a Brownlee RP8 column (5 µm, 25 cm × 4.6 mm i.d.). The column was eluted with methanol/water gradient at a flow rate of 1 mL/min as follows: 60-65% methanol for 15 min, isocratic elution at 65% for 5 min, followed by a gradient of 65-90% methanol for 10 min, 90% methanol for 5 min, and 90-100% methanol for 3 min. At the end of the gradient, the flow was maintained at 100% methanol for an additional 5 min. The peaks were detected by UV absorbance at 280 nm. The metabolites were identified by comparing their UV spectra and retention times with those of authentic standards. To quantitate the radiolabeled metabolites, 20 s fractions were collected, mixed with Scintisafe scintillation fluid, and assayed for radioactivity by liquid scintillation counting.

674

Chem. Res. Toxicol., Vol. 17, No. 5, 2004

Yuan et al.

Table 1. Metabolism of DB[a,h]ACR by Recombinant Human and Rat Cytochrome P40sa P450

DB[a,h]ACR metabolized (pmol/min/pmol P450)

human 1A1 human 1B1 human 3A4 rat 1A1

5.38 ( 0.56b 0.67 ( 0.07 0.20 ( 0.03 11.85 ( 1.31

a The P450s (50 pmol P450/mL) were incubated with 40 µM [14C]DB[a,h]ACR for 10 min in the presence of human or rat epoxide hydrolase, and the extent of DB[a,h]ACR metabolism was determined according to Van Cantfort et al. (33). Results are means ( SD of triplicate determinations. b Statistically significant from human 1B1, human 3A4, and rat 1A1 at P e 0.01.

Enantiometric Composition of DB[a,h]ACR diols Formed by Human P450 1A1 and 1B1. The enantiomeric composition of DB[a,h]ACR-3,4-diol and DB[a,h]ACR-10,11-diol formed by the P450s was determined as previously described (19). HPLC fractions containing a mixture of DB[a,h]ACR-3,4and DB[a,h]ACR-10,11-diol from several microsomal incubations were pooled and evaporated to dryness. The residue was treated with (-)-menthyloxyacetyl chloride in pyridine for 24 h at 0-5 °C. The mixture was partitioned between ethyl acetate and 5% HCl. The organic layer was then washed with 5% sodium bicarbonate and dried over anhydrous sodium sulfate. The resulting pairs of diastereomers of bis-(-)-menthyloxy esters of DB[a,h]ACR-3,4-and 10,11-diol were separated by normal phase HPLC on a Rainin Microsorb silica column (5 µm, 25 cm × 4.6 mm) and eluted with 6% ether in cyclohexane at a flow rate of 1 mL/min (19). Individual diastereomers were detected by UV absorption at 254 nm and quantitated by liquid scintillation counting.

Results Rate of Metabolism of DB[a,h]ACR by P450s. The capacity of the recombinant human P450s 1A1, 1B1, and 3A4 and rat P450 1A1 to metabolize DB[a,h]ACR {calculated from the amount of total DB[a,h]ACR metabolites formed according to Van Cantfort et al. (40)} is shown in Table 1. Among the human P450s, 1A1 was the most active in the total metabolism of DB[a,h]ACR. P450 1B1 was moderate in its activity whereas P450 3A4 exhibited a much lower catalytic activity for metabolizing the azaPAH. Human P450 1A1 metabolized DB[a,h]ACR at an 8- and 27-fold higher rate than human P450s 1B1 and 3A4, respectively. However, the rate of total DB[a,h]ACR metabolism by human P450 1A1 was less than half of that noted with rat P450 1A1. Regioselectivity of Individual P450s. Because of the nonsymmetry of the DB[a,h]ACR molecule, its enzymatic oxidation yields two benzo diols with a bay region double bond, e.g., DB[a,h]ACR-3,4-diol and DB[a,h]ACR10,11-diol. Because DB[a,h]ACR-10,11-diol is a precursor of the ultimate carcinogen DB[a,h]ACR-10,11-diol-8,9epoxide, the formation of the diol was used as an estimate for determining which P450s are capable of metabolically activating DB[a,h]ACR. A representative HPLC profile of the ethyl acetate soluble DB[a,h]ACR metabolites formed by human P450 1A1 incubated with [14C]DB[a,h]ACR for 10 min is shown in Figure 2. The following DB[a,h]ACR metabolites were identified on the basis of cochromatography with authentic standards: DB[a,h]ACR-3,4-diol, DB[a,h]ACR-10,11diol, DB[a,h]ACR-8,9-diol, and 3-hydroxyDB[a,h]ACR. Table 2 shows the relative proportions of the metabolites formed by human P450s 1A1 and 1B1. The major DB[a,h]ACR metabolites produced by human P450s 1A1

Figure 2. HPLC profile of [14C]DB[a,h]ACR metabolites formed by human P450 1A1. See Table 1 for incubation conditions. The metabolites were resolved on a Brownlee RP8 column as described in the text. Fractions were collected every 20 s and counted for radioactivity. Table 2. Profiles of [14C]DB[a,h]ACR Metabolites Formed by Recombinant Human and Rat Cytochrome P450sa % of total metabolites metabolites

human P450 1A1

24.5 ( 0.8 (1.31 ( 0.04)c 10,11-diol 48.4 ( 5.2 (2.60 ( 0.28) 8,9-diol 1.4 ( 0.6 (0.07 ( 0.00) 3-OH-DB[a,h]ACR + 7.0 ( 2.8 epoxides (0.37 ( 0.12) unknown metabolites 12.5 ( 2.3 (0.67 ( 0.12)

3,4-diol

human P450 1B1

rat P450 1A1

54.8 ( 3.8 (0.37 ( 0.00) 6.4 ( 1.3 (0.04 ( 0.00) ND (ND) 13.5 ( 1.6 (0.09 ( 0.01) 25.4 ( 0.8 (0.17 ( 0.00)

28.7 ( 0.3 (3.40 ( 0.03) 27.3 ( 0.7 (3.23 ( 0.08) 1.4 ( 0.4 (0.16 ( 0.04) 22.6 ( 0.1 (2.67 ( 0.00) 15.5 ( 0.3 (1.83 ( 0.03)

a Values obtained by incubating 40 µM [14C]DB[a,h]ACR with the p450s (50 pmol P450/mL) for 10 min in the presence of human or rat epoxide hydrolase. b Values are (SD of triplicate determinations and represent the percentage of total radioactivity, which emerges from the HPLC column in the defined metabolite fractions prior to DB[a,h]ACR. c The values in parentheses represent the amount of DB[a,h]ACR metabolites formed (pmol/min/pmol P450).

and1B1 were the diols with a bay region double bond, namely, DB[a,h]ACR-10,11-diol (proximate carcinogen) and DB[a,h]ACR-3,4-diol. Although the two P450s produced qualitatively similar DB[a,h]ACR metabolites, they exhibited remarkably different regioselectivity in the metabolism of the aza-PAH. The two P450s differed considerably from each other with respect to the proportions of DB[a,h]ACR-10,11-diol and -3,4-diol. Human P450 1A1 preferentially catalyzed the formation of DB[a,h]ACR-10,11-diol (derived from the benzo ring adjacent to the nitrogen), which accounted for nearly 50% of the ethyl acetate soluble metabolites. DB[a,h]ACR-3,4-diol (derived from the benzo ring away from the nitrogen) was formed to a much smaller extent (∼25%) by human P450 1A1. In contrast, human P450 1B1 showed a strong preference for metabolism at the 3,4-position rather than at the 10,11-position, forming DB[a,h]ACR-3,4-diol to an extent of nearly 55% of the total ethyl acetate soluble metabolites. This P450 produced DB[a,h]ACR-10,11-diol in only minor amounts (6.5% of total metabolites). Both 1A1 and 1B1 produced DB[a,h]ACR-8,9-diol as a minor or nondetectable metabolite. While human P450 1A1 preferentially oxidized the 10,11-position, rat P450 1A1 showed no regioselectivity,

Metabolism of DB[a,h]ACR by Human and Rat P450s

Chem. Res. Toxicol., Vol. 17, No. 5, 2004 675 Table 3. Enantiomeric Composition (%) of trans-DB[a,h]ACR-3,4-diol and 10,11-Diol Formed by Recombinant Human and Rat Cytochrome P450sa

Figure 3. Separation by HPLC of R,R and S,S enantiomers of [14C]DB[a,h]ACR 3,4- and 10,11-diol formed upon incubation of [14C]DB[a,h]ACR with human P540 1A1. The metabolites cochromatographing with (()DB[a,h]ACR 3,4-diol and (()10,11diol were isolated by HPLC, converted to (-)-menthyloxyacetyl ester, then resolved by direct phase HPLC as described in the text, and assayed for radioactivity as described in the text.

producing almost equal proportions of DB[a,h]ACR-3,4diol and -10,11-diol. The data on the proportion of the two diols formed by rat P450 1A1 are similar to those noted with the studies on the metabolism of DB[a,h]ACR by 3-methylcholanthrene-induced rat liver microsomes (18). Stereoselectivity of Individual P450s. The enatiomeric composition of DB[a,h]ACR-3,4-diol and DB[a,h]ACR-10,11-diol produced by human P450s 1A1 and 1B1 and rat P450 1A1 was determined after pooling the products from several incubations, as described in the Materials and Methods. Each sample showed the presence of four peaks (Figure 3). The first two early eluting peaks at 14.2 and 16.1 min were identified as the bis-()-menthyloxyacetyl esters of DB[a,h]ACR-10,11-diol with R,R and S,S absolute configurations, respectively, based on cochromatography with authentic standards (19). The remaining two peaks with retention times of 19.0 and 21.6 min were tentatively identified as the corresponding bis-(-)-menthyloxyacetyl esters of DB[a,h]ACR-3,4-diol with R,R and S,S absolute configurations, respectively, based on the general observation (41-44) that the analogous bis-(-)-menthyloxyacetyl esters with the R,R absolute configuration elute earlier than the corresponding bis-(-)-menthyloxyacetyl esters with the S,S absolute configuration. Quantitative analysis of each bis-(-)menthyloxyacetyl ester indicated that all of the three P450s produced predominately the R,R enantiomer of the metabolically formed DB[a,h]ACR-3,4-diol and DB[a,h]ACR-10,11-diol (Table 3). These were formed with high enantiomeric purity ranging from 83 to 100%.

Discussion Because aza-PAHs such as DB[a,h]ACR require metabolic activation by cytochrome P450 in order to express their mutagenic/carcinogenic activity, assessment of individual P450s contributing to the metabolic activation of DB[a,h]ACR is of considerable interest. For assessing the contribution of individual P450s to the metabolic activation of an individual carcinogen, their substrate specificity, regioselectivity, and stereoselectivity need to be determined. The present investigation was undertaken to gain insight into the relative contribution of specific forms of human cytochrome P450 to the metabolic activation of DB[a,h]ACR. The formation of the bay

absolute configurationb

human P450 1A1

human P450 1B1

10R,11R 10S,11S enantiomeric purity (10R,11R) 3R,4R 3S,4S enantiomeric purity (3R,4R)

95.7 4.3 91.4

100 NDc 100

94.4 5.6 88.8

100 ND 100

rat P450 1A1 100 ND 100 91.5 8.5 83.0

a The P450s (50 pmol P450/mL) were incubated with 40 µM [14C]DB[a,h]ACR for 10 min in the presence of human or rat epoxide hydrolase. The R,R and S,S enantiomers of DB[a,h]ACR3,4- and 10,11-diols were quantitated as described in the text. The data are based on three pooled incubations. b Percent enantiomeric purity is defined as 100 (mol predominate enantiomer R,R-mol S,S)/(mol R,R + mol S,S) (19). c Not detectable.

region DB[a,h]ACR diol epoxides requires two sequential oxidations by P450 enzymes at 3,4-, and 10,11-positions. Because DB[a,h]ACR-10,11-diol is an essential intermediate in the pathway of metabolic activation of the azaPAH (9, 10), the ability of individual P450s for metabolic activation of DB[a,h]ACR was assessed by determining the level of the 10,11-diol. The study investigated the role of individual recombinant human P450 1A1, 1B1, and 3A4 in the metabolism of DB[a,h]ACR. All three P450s metabolized DB[a,h]ACR, but major differences were observed in their catalytic efficiencies (Table 1). The ability of the human P450s to metabolize the aza-PAH was in the following order: 1A1 > 1B1 > 3A4. This rank order is consistent with other studies with human P450 enzymes using phenanthrene (34), B[a]P (24, 25, 31), dibenz(a,h)anthracene (27), dibenz(a,1)pyrene (26, 30), benzo[c]phenanthrene (35), and DB[a,j]ACR (36) as substrates. The human P450 1A1 was the most active in metabolizing these PAHs. Our data support the view that both human P450 1A1 and 1B1 play an important role in the metabolism of PAHs. Because of the higher activity of P450 1A1 and P450 1B1 to metabolize DB[a,h]ACR, we examined the regioand stereoselectivity of the two P450s in the metabolism of the aza-PAH. The metabolic pathways characterized previously in rat liver microsomes (18), i.e., the formation of DB[a,h]ACR diols, phenols, and epoxides, were all observed in the case of human P450 1A1 and P450 1B1. Therefore, the study shows that both recombinant human P450s and the rat liver microsomal preparations catalyze the oxidation of DB[ah]ACR to similar metabolites. Regioselectivity in the oxidation of PAHs is of major importance for their metabolic activation (45). Although the metabolism of DB[a,h]ACR by human P450s 1A1 and 1B1 was qualitatively similar, there were considerable differences in the relative proportions of the individual metabolites formed. Our data indicate that human P450s 1A1 and 1B1 are remarkably dissimilar with respect to the regioselective oxidation of DB[a,h]ACR. A striking feature of DB[a,h]ACR metabolism by human P450s 1A1 and 1B1 was that the P450s differed substantially with respect to their ability to oxidize the two benzo rings to form the respective diols with a bay region double bond (DB[a,h]ACR-3,4-diol and DB[a,h]ACR-10,11-diol). P450 1A1 was considerably more regioselective for oxidation at the 10,11-position than at the 3,4-position. This P450

676

Chem. Res. Toxicol., Vol. 17, No. 5, 2004

metabolized DB[a,h]ACR to its 10,11-diol (the putative proximate carcinogen) and 3,4-diol in a ratio of 2:1. On the other hand, P450 1B1 showed a high regioselectivity for attacking the 3,4-position producing DB[a,h]ACR-3,4diol and DB[a,h]ACR-10,11-diol in a ratio of 9:1. Different regioselectivities between human P450s 1A1 and 1B1 were also previously noted in studies with other PAHs such as benzo[c]phenathrene (35) and 7,12-dimethylbenz[a]anthracene (37). In contrast, human P450s 1A1 and 1B1 showed a remarkable similarity in the regioselective oxidation of the potent carcinogenic PAH dibenz[a,1]pyrene (26). The favored formation of DB[a,h]ACR-10,11diol by P450 1A1 is of considerable toxicological significance because further oxidation of this diol leads to the formation of the ultimate carcinogen of DB[a,h]ACR (9, 10, 19). The data of these studies show that although P450 1B1 is able to metabolize DB[a,h]ACR, it plays a minor role in the oxidation at the 10,11-position, suggesting that P450 1B1, unlike P450 1A1, does not make a major contribution to the initial key step of metabolic activation of DB[a,h]ACR, namely, the conversion of DB[a,h]ACR to DB[a,h]ACR-10,11-diol. This is in contrast to the studies of Shimada et al. (31), which demonstrated that P450 1B1 was more active than P450 1A1 in catalyzing the metabolism of B[a]P to B[a]P-7,8-diol, the proximate carcinogenic B[a]P metabolite. In another study, Shimada et al. (29) observed that P450 1B1 was more active than P450 1A1 in oxidizing the diols of many PAHs to mutagenic metabolites, presumably via the formation of their respective diol epoxides. Although our data show that P450 1B1 is less active in metabolizing DB[a,h]ACR to its 10,11-diol (the approximate carcinogenic DB[a,h]ACR metabolite), the possibility exists that this P450 may play a more important role in the second key step in the metabolic activation of DB[a,h]ACR, namely, the epoxidation of DB[a,h]ACR-10,11-diol. Further studies are planned to investigate this possibility. In addition to regioselectivity, stereoselectivity of P450 enzymes plays an important role in the metabolic activation of PAHs. For a number of PAHs, only the R,R-diol with a bay region double bond and bay region R,S-diolS,R-epoxides display high mutagenic and carcinogenic activity (45). Both P450 1A1 and P450 1B1 showed a high degree of stereoselectivity in the metabolism of DB[a,h]ACR to 10,11-diol and 3,4-diol with the (-)-(R,R) enantiomer predominating in each case. The data of our studies are in accordance with those of Roberts-Thompson et al. (36) who demonstrated that the metabolism of the aza-PAHs DB[a,j]ACR and 7-methylbenz[c]acridine to their respective non-K-region diols by human P450 1A1 proceeds with a high degree of stereoselectivity with the (-)-(R,R) enantiomer being prdominant. The observed stereoselectivity of human P450s 1A1 and 1B1 noted in the metabolism of DB[a,h]ACR is of toxicological significance because DB[a,h]ACR-10,11-diol with an [R,R] configuration is substantially more tumorigenic than the diol with an [S,S] configuration (10). Further oxidation of this stereoisomer produces the ultimate carcinogenic (+)-(8R,9S,10S,11R)-diol epoxide (19). The data from these studies show that human P450 1A1 is the major P450 contributing to the metabolic activation of DB[a,h]ACR. Because differences exist between species with respect to the regio- and stereoselectivity of individual P450s (24, 34), we compared the metabolism of DB[a,h]ACR by human and rat recombi-

Yuan et al.

nant P450 1A1. Such information is necessary for assessing the relevance of animal experimental data for humans. Rat P450 1A1 was considerably more active (>2fold) in metabolizing DB[a,h]ACR as compared to human P450 1A1. Although both human P450 1A1 and rat P450 1A1 produced similar DB[a,h]ACR metabolites, the two P450s differed greatly from each other with respect to regiospecific metabolism of the aza-PAH. The human P450 1A1 demonstrated a high degree of regioselectivity in oxidizing the two benzo rings with a bay region double bond, favoring the formation of DB[a,h]ACR-10,11-diol over DB[a,h]ACR-3,4-diol by a ratio of 2:1. In contrast, rat P450 1A1 showed no regioselectivity in the formation of the two diols producing them in almost equal proportions. Despite significant differences in their regioselectivity, human P450 1A1 and rat P450 1A1 exhibited similar stereoselectivity in the metabolism of DB[a,h]ACR to its diols having a bay region double bond, producing predominantly the R,R enantiomers. Because of the presence of the dissimilar bay regions, DB[a,h]ACR is an ideal compound in which to assess the effect of the nitrogen atom on the regioselective metabolism of the adjacent 8,9,10,11-benzo ring and remote 1,2,3,4-benzo ring of the aza-PAH. As noted above, human P450s 1A1 and 1B1 exhibit a high degree of regioselectivity in oxidation at 10,11- and 3,4-position of DB[a,h]ACR, respectively. These differences in the regioselectivity of human P450 1A1 and P450 1B1 suggest that the presence of the nitrogen heteratom affects the orientation of the 3,4- and 10,11-double bonds of DB[a,h]ACR to the active site of the two P450s. However, a similarity noted in the stereoselective metabolism of DB[a,h]ACR at these double bonds by P450s 1A1 and 1B1 suggests that the active sites of the two P450s involved in the metabolism of DB[a,h]ACR are stereochemically identical. In conclusion, the results of this investigation demonstrate that human P450s 1A1 and 1B1, the two important human P450s involved in the metabolism of PAHs, differ greatly from each other both with respect to the substrate specificity and regioselectivity but not stereoselectivity in the metabolism of DB[a,h]ACR. On the basis of these data, human P450 1A1, but not P450 1B1, plays a major role in the metabolic activation of DB[a,h]ACR to DB[a,h]ACR-10,11-diol. The nitrogen heteroatom significantly influences the regioselective metabolism of the two DB[a,h]ACR benzo rings having a bay region double bond by human P450s 1A1 and 1B1. Human P450 1A1 and rat P450 1A1 exhibit considerable differences in the regioselective metabolism of the aza-PAH.

Acknowledgment. This research was supported in part by Grant ES03346 from the National Institute of Environmental Health Sciences (S.K). We thank Dr. Bruce Hammock, University of California, Davis, CA, for providing recombinant rat microsomal epoxide hydrolase and Emily Dehoff for her help in preparing the manuscript.

References (1) Sawicki, E., Meeker, J. E., and Morgan, M. J. (1965) The quantitative composition of air pollution: source of effluents in terms of aza heterocyclic compounds and polynuclear aromatic hydrocarbons. Int. J. Air Water Pollut. 9, 291-298. (2) Hoffmann, D., and Wynder, E. L. (1977) Organic particulate pollutants: chemical analysis and bioassay for carcinogenicity.

Metabolism of DB[a,h]ACR by Human and Rat P450s

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

(12)

(13)

(14)

(15)

(16)

(17)

(18)

(19)

(20)

(21)

(22)

In Air Pollution (Stern, A. C., Ed.) 3rd ed., Vol. 2, pp 361-455, Academic Press, New York. Schmeltz, I., and Hoffmann, D. (1977) Nitrogen containing compounds in tobacco and tobacco smoke. Chem. Rev. 77, 295311. Van Duuren, B. L., Bilbao, J. A., and Joseph, C. A. (1960) The carcinogenic nitrogen heterocyclics in cigarette-smoke condensate. J. Natl. Cancer Inst. 25, 53-61. International Agency for Research on Cancer (1983) Polynuclear Aromatic Compounds, Part 1. IARC Monogr. Eval. Carcinog. Risk Chem. Hum. 32. International Agency for Research on Cancer (1972) Certain Polynuclear Aromatic Hydrocarbons and Heterocyclic Compounds. IARC Monogr. Eval. Carcinog. Risk Chem. Hum. 3. Dong, M., Schmelz, I., LaVoie, E., and Hoffmann, D. (1978) Azaarenes in the respiratory environment: analysis and assays for mutangenicity. In Carcinogenesis, Vol. 3, Polynuclear Aromatic Hydrocarbons (Jones, P. W., and Freudenthal, R. I., Eds.) pp 97108, Raven Press, New York. Dipple, A., Moschel, R. C., and Bigger, A. H. (1984) Polycyclic aromatic carcinogens. In Chemical Carcinogens (Searle, C. E., Ed.) 2nd ed., pp 43-163, American Chemical Society, Washington, DC. Wood, A. W., Chang, R. L., Katz, M., Conney, A. H., Jerina, D. M., Sikka, H. C., Levin, W., and Kumar, S. (1989) Mutagenicity of dihydrodiols and diol epoxides of dibenz[a,h]acridine in bacterial and mammalian cells. Cancer Res. 49, 6981-6984. Kumar, S., Chang, R. L., Wood, A. W., Xie, J. G., Huang, M. T., Cui, X. X., Kole, P. L., Sikka, H. C., Balani, S. K., Conney, A. H., and Jerina, D. M. (2001) Turorigenicity of racemic and optically pure bay region diol epoxides and other derivatives of the nitrogen heterocycle dibenz[a,h]acridine on mouse skin. Carcinogenesis 22, 951-955. Chang, R. L., Levin, W., Conney, A. H., Boyd, D. R., Balani, S. K., Yagi, H., Jerina, D. M., and Wood, A. W. (1989) Mutagenicity and tumorigenicity of the four optically active bay-region 3,4-diol1,2-epoxide isomers of dibenz[a,h]anthracene. Proc. Am. Assoc. Cancer Res. 30, 144. Andervont, H. B., and Shimkin, M. B. (1940) Biologic testing of carcinogens, II. Pulmonary tumor induction technique. J. Natl. Cancer Inst. 1, 225-239. Wood, A. W., Chang, R. L., Levin, W., Ryan, D. E., Thomas, P. E., Lehr, R. E., Kumar, S., Schaefer-Ridder, M., Engelhardt, U., Yagi, H., Jerina, D. M., and Conney, A. H. (1983) Mutagenicity of diol-epoxides and tetrahydroepoxides of benz[a]acridine and benz[c]acridine in bacteria and mammalian cells. Cancer Res. 43, 1656-1662. Lehr, R. E., Wood, A. W., Levin, W., Conney, A. H., and Jerina, D. M. (1988) Benzacridines and dibenzacridines: metabolism, mutagenicity, and carcinogenicity. In Polycyclic Aromatic Hydrocarbon Carcinogenesis: Structure-Activity Relationships (Yang, S. K., and Silverman, B. D., Eds.) L. 1, pp 31-58, CRC Press, Boca Raton, FL. Chang, R. L., Wood, A. W., Kumar, S., Lehr, R. E., Naohiro, S., Jerina, D. M., and Conney, A. H. (2000) Tumorigenicity of four optically active bay-region 3,4-diol 1,2-epoxides and other derivatives of the nitrogen heterocycle dibenz[c,h]acridine on mouse skin and in newborn mice. Carcinogenesis 21, 1997-2003. Warshawsky, D. (1992) Environmental sources, carcinogenicity, mutagenicity, metabolism and DNA binding of nitrogen and sulfur heterocyclic aromatics. J. Environ. Sci. Health C10, 1-71. Chang, R. L., Battista, S., Wong, C. Q., Kumar, S., Kole, P. L., Sikka, H. C., Balani, S. K., Jerina, D. M., Conney, A. H., and Wood, A. W. (1993) Bacterial and mammalian cell mutagenicity of four optically active bay-region 10,11-diol 8,9-epoxides of the nitrogen heterocycle dibenz[a,h]acridine. Carcinogenesis 14, 22332237. Steward, A. R., Kumar, S., and Sikka, H. C. (1987) Metabolism of dibenz[a,h]acridine by rat liver microsomes. Carcinogenesis 8, 1043-1050. Kumar, S., Singh, S. K., Kole, P. L., Elmarakby, S., and Sikka, H. C. (1995) Stereoselective metabolism of dibenz[a,h]acridine to bay-region diol epoxides by rat liver microsomes. Carcinogenesis 16, 525-530. Guengerich, F. P. (1995) Human cytochrome P450 enzymes. In Cytochrome P450 (Ortiz de Montellano, P. R., Ed.) 2nd ed., pp 473-535, Plenum, New York. Wilson, N. M., Christou, M., Turner, C. R., Wrighton, S. A., and Jefcoate, C. R. (1984) Binding and metabolism of benzo[a]pyrene and 7,12-dimethylbenz[a]anthracene by seven purified forms of cytochrome P-450. Carcinogenesis 5, 1475-1483. Hall, M., Forrester, L. M., Parker, D. K., Grover, P. L., and Wolf, C. R. (1989) Relative contribution of various forms of cytochrome

Chem. Res. Toxicol., Vol. 17, No. 5, 2004 677

(23)

(24)

(25)

(26)

(27)

(28)

(29)

(30)

(31)

(32) (33)

(34)

(35)

(36)

(37)

(38)

(39)

(40)

(41)

P450 to the metabolism of benzo[a]pyrene by human liver microsomes. Carcinogenesis 10, 1815-1821. Roberts-Thompson, S. J., McManus, M. E., Tukey, R. H., Gonzalez, F. F., and Holder, G. M. (1993) The catalytic activity of four expressed human cytochrome P450s towards benzo[a]pyrene and the isomers of its proximate carcinogen. Biochem. Biophys. Res. Commun. 192, 1373-1379. Shou, M., Korzekwa, K. R., Crespi, C. L., Gonzalez, F. J., and Gelboin, H. V. (1994) The role of 12 cDNA-expressed human, rodent, and rabbit cytochromes P450 in the metabolism of benzo[a]pyrene and benzo[a]pyrene trans-7,8-dihydrodiol. Mol. Carcinog. 10, 159-168. Bauer, E., Guo, Z., Ueng, U., Bell, C., Zeldin, D., and Guengerich, F. P. (1995) Oxidation of benzo[a]pyrene by recombinant human cytocrome P450 enzymes. Chem. Res. Toxicol. 8, 136-142. Shou, M., Krausz, W. K., Gonzalez, F. J., and Gelboin, H. V. (1996) Metabolic activation of the potent carcinogen dibenzo[a,l]pyrene by human recombinant cytochromes P450, lung and liver microsomes. Carcinogenesis 17, 2429-2433. Shou, M., Krausz, W. K., Gonzalez, F. J., and Gelboin, H. V. (1996) Metabolic activation of the potent carcinogen dibenzo[a,h]anthracene by cDNA-expressed human cytochromes P450. Arch. Biochem. Biophys. 328, 201-207. Jacob, J., Raab, G., Soballa, V., Schmalix, W. A., Grimmer, G., Greim, H., Doehmer, J., and Seidel, A. (1996) Cytochrome P450mediated activation of phenanthrene in genetically engineered V79 Chinese hamster cells. Environ. Toxicol. Pharmocol. L, 1-11. Shimada, T., Hayes, C. L., Yamazaki, H., Amink, S., Hecht, S. S., Guengerich, F. P., and Sutter, T. R. (1996) Activation of chemically diverse procarcinogens by human cytochrome P-450 1B1. Cancer Res. 56, 2979-2984. Shimada, T., Oda, Y., Gillam, E. M. J., Guengerich, F. P., and Inoue, K. (2001) Metabolic activation of polycyclic aromatic hydrocarbons and other procarcinogens by cytochromes P450 1A1 and P450 1B1 allelic variants and other human cytochromes P450 in Salmonella typhimurium NM2009. Drug Metab. Dispos. 29, 1176-1182. Shimada, T., Gillam, E. M. J., Oda, Y., Tsumura, F., Sutter, T. R., Guengerich, F. P., and Inoue, K. (1999) Metabolism of benzo[a]pyrene by trans-7,8-dihyroxy-7,8-dihyrobenzo[a]pyrene by recombinant human cytochrome P450 1B1 and purified liver epoxide hydrolase. Chem. Res. Toxicol. 12, 623-629. Shimada, T. (2000) Human cytochrome P450 1B1 and chemical carcinogenesis. Rev. Toxicol. 3, 139-166. Deutsch, J., Leutz, J. C., Yang, S. K., Gelboin, H. V., Chiang, Y. L., Vatsis, K. P., and Coon, M. J. (1978) Regio- and stereoselectivity of various forms of purified cytochrome P-450 in the metabolism of benzo[a]pyrene and (-)trans-7,8-dihydroxy-7,8dihydrobenzo[a]pyrene as shown byproduct formation and binding to DNA. Proc. Natl. Acad. Sci. U.S.A. 75, 3123-3127. Shou, M., Korzekwa, K. R., Krausz, K. W., Crespi, C. L., Gonzalez, F. J., and Gelboin, H. V. (1994) Regio- and stereoselective metabolism of phenanthrene by twelve cDNA-expressed human, rodent, and rabbit cytochromes P-450. Cancer Lett. 83, 305-313. Seidel, A., Soballa, V. J., Raab, G., Frank, H., Greim, H., Grimmer, G., Jacob, J., and Doehmer, J. (1998) Regio- and stereoselectivity in the metabolism of benzo[c]phenanthrene mediated by genetically engineered V79 Chinese hamster cells expressing rat and human cytochromes P450. Environ. Toxicol. Pharmacol. 5, 179196. Roberts-Thompson, S. J., McManus, M. E., Tukey, R. H., Gonzalez, F. J., and Holder, G. M. (1995) Metabolism of polycyclic azaaromatic carcinogens catalyzed by four expressed human cytochromes P450. Cancer Res. 55, 1052-1059. Christou, M., Savas, U., Spink, D. C., Gierthy, J. F., and Jefcoate, C. R. (1994) Coexpression of CYP 1A1 and a human analogue of cytochrome P450-EF in response to 2,3,7,8-tetrachloro-dibenzop-dioxin in the human mammary carcinoma-derived MCF-7 cells. Carcinogenesis 15, 725-732. Kumar, S. (1985) Synthesis of trans-10,11-dihydroxy-10,11-dihydrodibenz[a,h]acridine and its diastereomeric epoxides: possible carcinogenic metabolites of dibenz[a]acridine. J. Org. Chem. 50, 3070-3073. Kumar, S., and Agarawal, N. L. (1986) Synthesis of dihydrodiols and diol epoxides of dibenz[a,h]acridine. J. Org. Chem. 51, 24452449. Van Cantfort, J., Degraeve, J., and Gielen, J. E. (1977) Radioactive assay for aryl hydrocarbon hydroxlyase. Improved method and biological importance. Biochem. Biophys. Res. Commun. 79, 505512. Kumar, S., Kole, P. L., Balani, S. K., and Jerina, D. M. (1992) Synthesis of enantiomerically pure bay-region 10,11-diol-8,9-

678

Chem. Res. Toxicol., Vol. 17, No. 5, 2004

epoxide diastereomers of carcinogen dibenz[a,h]acridine. J. Org. Chem. 50, 2784-2787. (42) Lehr, R. E., Kumar, S., Shirai, N., and Jerina, D. M. (1985) Synthesis of enatiomerically pure bay-region 3,4-diol-1,2-epoxide diastereomers and other derivatives of potent carcinogen dibenz[c,h]acridine. J. Org. Chem. 50, 98-107. (43) Yagi, H., Vyar, K. P., Tada, M., Thakker, D. R., and Jerina, D. M. (1982) Synthesis of the enantiomeric bay-region diol epoxides of benz[a]anthracene and chrysene. J. Org. Chem. 47, 1110-1117. (44) Yagi, H., Thakker, D. R., Ittah, Y., Croisy-Delcey, M., and Jerina,

Yuan et al. D. M. (1983) Synthesis and assignment of absolute configuration to the trans-3,4-diol-1,2-epoxides of benzo[c]phenanthrene. Tetrahedron Lett. 24, 1349-1352. (45) Thakker, D. R., Yagi, H., Levin, W., Wood, A. W., Conney, A. H., and Jerina, D. M. (1985) Polycyclic aromatic hydrocarbons: metabolic activation to ultimate carcinogens. In Bioactivation of Foreign Compounds (Anders, M. W., Ed.) pp 177-242, Academic Press, New York.

TX049979I