Metabolic Activation of Aromatic Amine Mutagens ... - ACS Publications

Guelph-Waterloo Centre for Graduate Work in Chemistry, Department of ... and Department of Molecular Biology and Genetics, University of Guelph, Guelp...
0 downloads 0 Views 97KB Size
70

Chem. Res. Toxicol. 1998, 11, 70-74

Metabolic Activation of Aromatic Amine Mutagens by Simultaneous Expression of Human Cytochrome P450 1A2, NADPH-Cytochrome P450 Reductase, and N-Acetyltransferase in Escherichia coli P. David Josephy,* David H. Evans,† Asit Parikh,‡ and F. Peter Guengerich‡ Guelph-Waterloo Centre for Graduate Work in Chemistry, Department of Chemistry and Biochemistry and Department of Molecular Biology and Genetics, University of Guelph, Guelph, Ontario, Canada N1G 2W1, and Department of Biochemistry and Center in Molecular Toxicology, Vanderbilt University School of Medicine, Nashville, Tennessee 37232 Received September 17, 1997X

We describe the construction of a new strain of Escherichia coli designed to bioactivate aromatic amines and to detect their mutagenicity with high sensitivity. Strain DJ4309 bears two plasmids, a pACYC184-derived plasmid which expresses Salmonella typhimurium acetyl CoA:arylamine N-acetyltransferase (NAT) and a pBR322-derived plasmid which expresses human cytochrome P450 1A2 and NADPH-cytochrome P450 reductase. The combined actions of these enzymes convert aromatic amines into reactive, mutagenic N-acetoxy esters. The strain also carries a mutated copy of the lacZ gene (on an F′ factor) which reverts to the wildtype gene by a -(GpC) frameshift mutation. Strain DJ4309 expresses high levels of NAT and cytochrome P450 1A2 and is very sensitive to mutagenesis induced by representative aromatic amines. Mutagenicity of 2-aminoanthracene in strain DJ4309 is higher than can be obtained by rat liver homogenate 9000g supernatant (S9) activation in the parent strain lacking the P450 expression vector. Strain DJ4309 provides a useful system for detecting mutagenic aromatic amines and for studying their metabolism by human P450 1A2.

Introduction Despite years of success in the heterologous expression of individual recombinant enzymes (1, 2), the task of engineering bacteria to carry out multistep metabolic pathways remains challenging. Introduction of multiple expression vectors into a single bacterial cell is constrained by the need for plasmid replicon compatibility and by the limited number of selectable markers (e.g., antibiotic-resistance) and inducible promoters available in common cloning vectors. There are many possible applications for bacterial cells which can, for example, accomplish the stepwise synthesis of complex natural products or the biodegradation of persistent environmental pollutants. To realize this goal, multiple recombinant enzymes must be expressed simultaneously. Furthermore, each enzyme must be enzymatically active, supplied with all necessary cofactors and substrates, and nontoxic to the host cell. We have been pursuing the goal of replacing the requirement for the use of mammalian enzyme preparations, such as hepatic S91 (post-mitochondrial supernatant), for the activation of chemical mutagens and carcinogens in genotoxicity assays (3). This research is motivated by several considerations: Replacement of * Address correspondence to this author at Department of Chemistry and Biochemistry, University of Guelph. Phone: (519) 824-4120, ext. 3833. Fax: (519) 766-1499. E-mail: [email protected]. † Department of Molecular Biology and Genetics, University of Guelph. ‡ Vanderbilt University School of Medicine. X Abstract published in Advance ACS Abstracts, December 15, 1997.

hepatic S9 by enzymes expressed in culture would fulfill the promise of bacterial genotoxicity assays as animalfree tests and could also reduce the expense and increase the reproducibility of the assay. Rodents, the usual source of S9, may have very different patterns of drug and carcinogen metabolism compared to humans, and this problem weakens the validity of estimates of human health risk based on short-term assays. Mechanistic analysis of the biochemistry of S9-dependent metabolism is also difficult, since many enzyme activities (activating and/or detoxifying) are present in a complex mixture. For these reasons, several investigators have been exploring alternative approaches to in vitro bioactivation, especially the expression of recombinant enzymes in cultured cells. We believe that bacteria will continue to be favored organisms for expression of recombinant genes. The advantages of rapid growth, easy introduction and recovery of plasmid DNA, and high-level expression are very attractive features of bacterial systems (4, 5). We (6) and others (7) have recently demonstrated that the hepatic P450 monooxygenase complex, comprising a chosen form of P450 and its cognate flavoprotein reductase, NADPH-P450 reductase (8), can be replicated by expression of the recombinant enzymes in Escherichia coli. 1 Abbreviations: 2-AA, 2-aminoanthracene; 2-AF, 2-aminofluorene; δ-ALA, δ-aminolevulinic acid; IPTG, isopropyl β-D-thiogalactoside; IQ, 2-amino-3-methylimidazo[4,5-f]quinoline; LB, Luria broth; MeIQ, 2-amino-3,4-dimethylimidazo[4,5-f]quinoline; NAT, acetyl CoA:arylamine N-acetyltransferase; PhIP, 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine; S9, 9000g hepatic supernatant preparation.

S0893-228x(97)00171-9 CCC: $15.00 © 1998 American Chemical Society Published on Web 01/19/1998

Metabolic Activation of Aromatic Amine Mutagens

P450 1A2 activates mutagenic aromatic amines by N-hydroxylation (9). The resulting N-hydroxyarylamines can be further activated by N-acetyltransferase (NAT)catalyzed O-acetylation, which yields N-reactive acetoxy esters (10). Beginning with a derivative of the Ames test strain TA1538, we previously constructed Salmonella typhimurium strain DJ4501A2, which expresses human P450 1A2 and bacterial (S. typhimurium) NAT (11). Aromatic amines [2-aminoanthracene (2-AA), 2-aminofluorene (2-AF), and the food pyrolysis product mutagen MeIQ] were mutagenic in the absence of S9, in this strain. This publication demonstrated, for the first time, the principle of replicating a complete mutagen bioactivation pathway in a bacterial cell. However, there were some limitations to the approach used in that work. The strain carried two plasmids based on the vector pBR322, which are incompatible, since both carry the same replicon. Second, in the absence of expression of NADPH-P450 reductase, the system relied on the limited capacity of endogenous bacterial flavoproteins to reduce the recombinant P450 (12). In this paper, we describe a new mutagenicity assay strain, based on E. coli strain CC109. CC109 (13) carries an F′ factor which bears a +2 frameshift mutation in a sequence (GCGCGCGC) of the lacZ gene, which is similar to the hisD3052 “hotspot” site found in S. typhimurium Ames test strains such as TA1538 and TA98. We have previously demonstrated that this lacZ allele, carried in E. coli strains that are defective in DNA nucleotide excision repair (Uvr-), is a sensitive target for aromatic amineinduced mutagenesis (14). Our new strain, DJ4309, incorporates a null mutation in uvrA and carries two compatible plasmids: pNM12 (15), for constitutive expression of (S. typhimurium) NAT, and pCW1A2 reductase (6), for inducible expression of human P450 1A2 and human NADPH-P450 reductase. The recombinant bioactivation system expressed in strain DJ4309 activates typical aromatic amine mutagens.

Experimental Procedures IQ (2-amino-3-methylimidazo[4,5-f]quinoline) and MeIQ (2amino-3,4-dimethylimidazo[4,5-f]quinoline) were kindly donated by Toronto Research Chemicals, Inc. (Downsview, Ontario, Canada); 2-AA (practical grade) and δ-aminolevulinic acid (δALA) were purchased from Sigma Chemicals (St. Louis, MO). Solutions of mutagens and inhibitors were made in DMSO and stored at -15 °C; IPTG (isopropyl β-D-thiogalactoside) was purchased from Diagnostic Chemicals, Ltd., Charlottetown, PEI, Canada. β-D-Lactose was purchased from Kodak Chemicals, Rochester, NY. The strains and plasmids used in this study are listed in Table 1. E. coli strain DJ3109 (CC109 ∆uvrA109) was described previously (14). This strain encodes a lacZ frameshift mutation on an FN factor, along with a pro gene which complements a chromosomal ∆pro mutation. Selection for maintenance of the FN factor is maintained by isolation of strains on minimal glucose media. Plasmid pNM12 carries the S. typhimurium NAT gene, under control of its native promoter, subcloned into the tetR gene of plasmid pACYC184. This plasmid was obtained by first transforming E. coli strain DH5R with a mixed plasmid preparation extracted from S. typhimurium strain NM2009 (kind gift of Dr. Y. Oda) and selecting for chloramphenicolresistant colonies. Plasmid pNM12 was then purified from one such colony. Plasmid pCW1A2 reductase has been described in detail elsewhere (6). The plasmid is based on the expression vector pCW and carries the pBR322 replicon, lacIq, and the bla (ampicillin-resistance) marker. The bicistronic expression of

Chem. Res. Toxicol., Vol. 11, No. 1, 1998 71 Table 1. Plasmids and E. coli Strains entry

description

pNM12 pCW1A2 pCW1A2 reductase

Plasmid pACYC184 derivative bearing S. typhimurium nat gene pCW derivative bearing human P450 1A2 gene (11) pCW derivative bearing human P450 1A2 and P450 reductase genes (6)

Strain CC109 ∆uvrA109 (14) expresses S. typhimurium NAT; chlR expresses NAT, P450 1A2; chlR, ampR ) DJ3109 pNM12 pCW1A2 reductase, expresses NAT, P450 1A2, and P450 reductase; chlR, ampR P450 1A2 and NADPH-P450 reductase is controlled by tandem tac promoters. Plasmid pNM12 was transfected into strain DJ3109 by electroporation, to construct strain DJ3109 pNM12. This strain, expressing NAT but not P450, was used as a control strain for S9-dependent mutagenicity assays. Strains DJ3109 pNM12 pCW1A2 (expressing NAT and P450 1A2) and DJ4309 (DJ3109 pNM12 pCW1A2 reductase, expressing NAT, P450 1A2, and P450 reductase) were then constructed by transformation of DJ3109. All transformants were selected on LB medium with the appropriate antibiotics and then reisolated by streaking onto minimal glucose plates with antibiotics. A single colony was grown overnight in minimal glucose medium (16) supplemented with thiamine and the appropriate antibiotics (ampicillin, 50 µg/mL; chloramphenicol, 10 µg/mL). Frozen stock was prepared from this culture and stored at -70 °C. For both the mutagenicity and the enzyme assays, frozen stock was used to inoculate cultures, which were grown overnight in LB medium supplemented with antibiotics (concentrations as above), IPTG (1 mM), δ-ALA (0.5 mM), and trace element mix (17). (The supplements were omitted for DJ3109 pNM12.) For NAT assays, cultures were grown in LB medium at 37 °C. Acetyl coenzyme A-dependent NAT activity of sonicated extracts was assayed using a colorimetric assay for 2-aminofluorene, modified for use of a 96-well plate reader, as described previously (14). P450 expression was measured by reduced CO difference spectra (18) of whole bacterial cells. Mutagenicity assays were performed as follows. Cultures were grown at 30 °C overnight, with shaking. For the assay, the following were dispensed at room temperature into a 5-mL snap-cap tube: sodium phosphate buffer (0.1 M, pH 7.4), 0.5 mL; mutagen solution in DMSO (or DMSO alone); and bacterial culture, 0.1 mL. The volume of DMSO was e10 µL. After mixing, the tubes were incubated at 30 °C, with occasional shaking, for 30 min. Top agar (2 mL) was added, and the mixes were poured onto minimal lactose plates (16) and then incubated at 37 °C. Plates were counted after 40-48 h. Experiments to test R-naphthoflavone inhibition were performed as above, except that the buffer, cells, and inhibitor were incubated at 30 °C for 15 min before addition of mutagen. For the S9-dependent assays, rat liver S9 (Aroclor 1254induced male Sprague-Dawley rat) was purchased from Molecular Toxicology, Boone, NC, and stored at -70 °C. S9 mix was prepared as for the Ames test (19) at a concentration of 1 mg of S9 protein/plate (0.5 mL of S9 mix). Cultures of strain DJ3109 pNM12 were grown at 37 °C overnight in LB + chloramphenicol, and the preincubation (S9 mix, cells, mutagen) was performed at 37 °C. DJ3109 DJ3109 pNM12 DJ3109 pNM12 pCW1A2 DJ4309

Results and Discussion Strain Growth Properties. All of the constructs reported here grew well on either LB or minimal glucose

72 Chem. Res. Toxicol., Vol. 11, No. 1, 1998

medium. Expression of recombinant proteins sometimes hinders growth of bacterial strains, and we observed that overnight cultures of strains bearing pCW derivatives (e.g., DJ4309) did grow more slowly in the presence of IPTG. However, cultures reached an OD600 of about 0.8, suitable for use in mutagenicity assays, after about 15 h growth at 30 °C. After plating on minimal lactoseselective plates, revertant colonies were visible after 24 h and were counted after 48 h. NAT Activity. A recent report indicates that NAT activity is detectable in stationary-phase cultures of E. coli strain MX100 (20), and the coding sequence of the chromosomal NAT gene, determined by the E. coli genome sequencing project, appears to encode a functional enzyme (21). Nevertheless, NAT activity in CC109 was below the detection limit of our assay (14), and plasmid-mediated overexpression of NAT enzyme greatly enhanced sensitivity to aromatic amine mutagens in both S. typhimurium (22) and E. coli (14). Plasmids pYG213 and pYG219 contain the identical NAT gene insert (a 1.35-kb BamHI-EcoRV segment of the S. typhimurium chromosome) in plasmid pBR322, but the insert is located in the tetracycline-resistance gene (tetR) in pYG213 and in the β-lactamase gene (bla) in pYG219. Watanabe et al. (23) compared NAT expression from these two plasmids in S. typhimurium strain TA1538 1,8-DNP. Enzyme expression was reported to be about 4-fold higher with pYG213. Plasmid pNM12, like pYG213, carries the NAT gene subcloned into tetR, but the host plasmid is pACYC184 rather than pBR322. We measured NAT activities of ultrasonic lysates prepared from two strains bearing pNM12: NM2009 (a S. typhimurium strain) and DJ4309 (E. coli). The activities were 43 and 110 nmol of 2-AF acetylated/(min mg of protein), respectively. These values are severalfold higher than the levels we obtained previously with a CC109-derived strain bearing plasmid pYG219 (14). NAT expression levels will be influenced by bacterial species, strain background, plasmid copy number, and possibly growth phase, but it appears that the NAT insert in tetR, found in pYG219 and pNM12, gives superior expression. P450 Expression. Reduced CO difference spectra were conducted on whole bacterial cells, to assess the expression of P450 holoenzyme in strain DJ4309. Optimal expression was seen at 20 h and averaged 60 nmol of P450/L of culture (Figure 1). Mutagenesis. Spontaneous (no mutagen added) revertant yields for strain DJ3109 pNM12 (with S9 activation) and its P450-expressing derivatives DJ3109 pNM12 pCW1A2 and DJ4309 (DJ3109 pNM12 pCW1A2 reductase) are indicated by horizontal lines in the three panels of Figure 2. In the absence of S9, the three strains had similar levels of spontaneous mutagenicity (150-250 revertants/plate in each experiment, slightly lower for DJ3109 pNM12 pCW1A2). However, the spontaneous revertant yield in the presence of S9 (measured for DJ3109 pNM12 only) was higher, as shown (Figure 2B). Watanabe-Akanuma and Ohta (24) described the construction of uvrA and rfa (deep rough) derivatives of the lacZ frameshift strains constructed by Cupples et al. (13). They noted that uvrA rfa derivatives of CC109 gave high (400-700) spontaneous reversion levels in the presence of S9, similar to our present observation. Strain DJ4309 responds to aromatic amine mutagens in the absence of S9. In preliminary experiments with

Josephy et al.

Figure 1. Whole cell reduced CO difference spectrum of strain DJ4309 cells. Bacteria were pelleted and resuspended in an equal volume of assay buffer (50 mM MOPS, 100 mM KCl, pH 7.5) prior to recording the spectrum on an Aminco DW2/OLIS spectrophotometer (OLIS, Bogart, GA).

IQ and MeIQ (data not shown), we established the following characteristics of the mutagenic response: No response was seen in the parent strain, DJ3109 pNM12, in the absence of S9, and the response of strain DJ4309 is completely inhibited by preincubation with R-naphthoflavone (10 nmol/plate). These results verify that the response is dependent on P450 expression. Cultures grown without IPTG induction showed about one-half the mutagenic response of induced cultures. We attribute this effect to induction of P450 expression by lactose present as the carbon source on the selective plates. Indeed, addition of R-naphthoflavone after the preincubation period also reduced the mutagenic response, and a response can be seen without any preincubation, consistent with the hypothesis that mutagen activation occurs both during the preincubation and after plating. A reduced response was observed with cultures grown in the absence of δ-ALA, a metabolic precursor of heme. Many studies of recombinant P450 expression in E. coli have reported that δ-ALA supplementation improves enzyme expression. Dose-response curves for IQ and MeIQ with strain DJ4309 are shown in Figure 2A (no S9 was used in these experiments). Both IQ and MeIQ were activated by the expressed P450. For comparison, results for the parent strain DJ3109 pNM12 (with rat hepatic S9 activation) are given in Figure 2B. Dose-response curves for 2-AA are shown in Figure 3. In this figure, the spontaneous revertant yields for each strain were subtracted from the corresponding data points, to allow a direct comparison of induced mutagenicity in each strain. The expressed P450 + P450 reductase system of strain DJ4309 was more effective than rat liver S9 at activating 2-AA. DJ3109 pNM12 responds to S9-activated IQ or MeIQ at doses as low as 0.3 pmol/plate and to S9-activated 2-AA at doses of about 0.1-1 nmol/plate. Our strain is not quite as sensitive as the NAT-overexpressing S. typhimurium hisD3052 strains YG1024 and YG1019, which give detectable mutagenic responses at doses of 0.1 pmol of MeIQ (11) and 50 pmol of 2-AA (25). This may reflect differences between the lacZ and hisD DNA targets or uncharacterized metabolic differences between E. coli and S. typhimurium. DJ3109 pNM12 is about 10-fold more sensitive to IQ than is strain DJ3209 (14), the corresponding strain which carries the NAT gene on plasmid pYG219, and it is far more sensitive than comparable

Metabolic Activation of Aromatic Amine Mutagens

Chem. Res. Toxicol., Vol. 11, No. 1, 1998 73

Figure 2. Activation of IQ and MeIQ: A, strain DJ4309, without use of S9; B, strain DJ3109 pNM12, with rat hepatic S9 activation; C, strain DJ3109 pNM12 pCW1A2, without use of S9. Experiments were performed as described in the text. Symbols: MeIQ, open circles and solid line; IQ, solid circles and dotted line. The average spontaneous revertant yield for each strain is indicated by the horizontal solid line. (For strain DJ3109 pNM12, this corresponds to the spontaneous revertant yield in the presence of S9.) Data points represent the average ( SE of at least five plates from two or three independent experiments.

Figure 3. Activation of 2-AA. Data for this figure have been corrected by subtraction of the spontaneous background revertant yields for each strain. Symbols: strain DJ4309, open squares and solid line; strain DJ3109 pNM12 (with S9 activation), solid triangles and dashed line; strain DJ3109 pNM12 pCW1A2, open diamonds and dotted line.

strains which do not carry plasmid-borne copies of the NAT gene (25). Our data indicate that the relative activating abilities of the rat and recombinant human enzymes differ markedly between IQ/MeIQ and 2-AA. In other studies (R. Turesky and F. P. Guengerich, unpublished data) we have found that rat and human P450 1A2 enzymes are similar in their rates of metabolism of PhIP (2-amino-1methyl-6-phenylimidazo[4,5-b]pyridine) but that there are considerable differences between the two enzymes in their activation of some other heterocyclic amines. One should also bear in mind that enzymes other than P450 can contribute to aromatic amine activation by hepatic S9 preparation (26). Strain DJ3109 pNM12 pCW1A2 expresses NAT and P450 1A2, but not P450 reductase, and so is comparable to our previously described S. typhimurium strain DJ4501A2 (11). With strain DJ3109 pNM12 pCW1A2, IQ and MeIQ, tested over the same dose range as for strain DJ4309, gave almost no response (Figure 2C). 2-AA mutagenicity could be detected, but the response (Figure 3, dotted line) was much weaker than for strain DJ4309. The F′-borne lacZ assay in E. coli is a versatile and informative mutagenicity assay system and has the important advantage (compared to the Ames test strains)

of giving unambiguous identification of the sequence change responsible for reversion of each strain. Here, we show that the -(GpC) deletion target present in strain CC109 provides an excellent system for studying the recombinant human P450 1A2-dependent activation of aromatic amines. Sensitivity is enhanced by elimination of nucleotide excision repair of DNA, overexpression of NAT, and coexpression of human NADPH-P450 reductase. The yield of spontaneous revertants observed in strain DJ3109 (14) is not greatly increased by these additional strain construction steps. Also, the expression of P450 1A2 does not result in the large increase in spontaneous reversion associated with the use of S9. The new strains are phenotypically stable, and we have not encountered any evidence of rapid plasmid loss. We believe that this mutagenicity assay system can be adapted to the expression of other forms of human cytochrome P450 and detection of other classes of chemical mutagens and carcinogens. [During the preparation of this manuscript, a preliminary report was presented describing the expression of human cytochrome P450 1A2 and cytochrome P450 reductase in an E. coli strain bearing an argE ochre allele, which reverts primarily by base-substitution mutations (27).]

Acknowledgment. We thank the Natural Sciences and Engineering Research Council of Canada and the International Foundation for Ethical Research (P.D.J.), Medical Research Council of Canada (D.H.E.), and U.S. Public Health Service grants R35 CA44353, P30 ES00267 (F.P.G.), and T32 GM07347 (A.P., F.P.G.) for financial support and Ms. Crista Thompson for technical assistance. We also thank Laura Zajchowski and Tracey Henry, students in the P.D.J. lab, for preliminary studies which helped to guide this project and Dr. Robert Turesky, Nestle´ Research Centre, Lausanne, Switzerland, for helpful discussions.

References (1) Guengerich, F. P., Parikh, A., Johnson, E. F., Richardson, T. H., von Wachenfeldt, C., Cosme, J., Jung, F., Strassburg, C. P., Manns, M. P., Tukey, R. H., Pritchard, M., Fournel-Gigleux, S., and Burchell, B. (1997) Heterologous expression of human drugmetabolizing enzymes. Drug Metab. Dispos. 25, 1234-1241. (2) Guengerich, F. P., and Parikh, A. (1997) Expression of drug metabolizing enzymes. Curr. Opin. Biotechnol. 8, 623-628. (3) Josephy, P. D., Gruz, P., and Nohmi, T. (1997) Recent advances in the construction of bacterial genotoxicity assays. Mutat. Res. 386, 1-23.

74 Chem. Res. Toxicol., Vol. 11, No. 1, 1998 (4) Guengerich, F. P., Gillam, E. M. J., and Shimada, T. (1996) New applications of bacterial systems to problems in toxicology. Crit. Rev. Toxicol. 26, 551-583. (5) Guengerich, F. P., Martin, M. V., Guo, Z., and Chun, Y.-J. (1996) Purification of recombinant human cytochrome P450 enzymes expressed in bacteria. Methods Enzymol. 272, 35-44. (6) Parikh, A., Gillam, E. M. J., and Guengerich, F. P. (1997) Drug metabolism by Escherichia coli expressing human cytochromes P450. Nature Biotechnol. 15, 784-788. (7) Blake, J. A. R., Pritchard, M., Ding, S., Smith, G. C. M., Burchell, B., Wolf, C. R., and Friedberg, T. (1996) Coexpression of a human P450 (CYP3A4) and P450 reductase generates a highly functional monooxygenase system in Escherichia coli. FEBS Lett. 397, 210214. (8) Wang, M., Roberts, D. L., Paschke, R., Shea, T. M., Masters, B. S. S., and Kim, J. J. P. (1997) Three-dimensional structure of NADPH-cytochrome P450 reductase: Prototype for FMN- and FAD-containing enzymes. Proc. Natl. Acad. Sci. U.S.A. 94, 84118416. (9) Hammons, G. J., Milton, D., Stepps, K., Guengerich, F. P., Tukey, R. H., and Kadlubar, F. F. (1997) Metabolism of carcinogenic heterocyclic and aromatic amines by recombinant human cytochrome P450 enzymes. Carcinogenesis 18, 851-854. (10) Josephy, P. D. (1997) Molecular Toxicology, Oxford University Press, New York. (11) Josephy, P. D., DeBruin, L. S., Lord, H. L., Oak, J. N., Evans, D. H., Guo, Z., Dong, M.-S., and Guengerich, F. P. (1995) Bioactivation of aromatic amines by recombinant human cytochrome P450 1A2 expressed in Ames tester strain bacteria: A substitute for activation by mammalian tissue preparations. Cancer Res. 55, 799-802. (12) Jenkins, C. M., and Waterman, M. R. (1994) Flavodoxin and NADPH-flavodoxin reductase from Escherichia coli support bovine P450c17 hydroxylase activities. J. Biol. Chem. 269, 2740127408. (13) Cupples, C. G., Cabrera, M., Cruz, C., Miller, J. H. (1990) A set of lacZ mutations in Escherichia coli that allow rapid detection of specific frameshift mutations. Genetics 125, 275-280. (14) Marwood, T. M., Meyer, D., and Josephy, P. D. (1995) Escherichia coli lacZ strains engineered for detection of frameshift mutations induced by aromatic amines and nitroaromatic compounds. Carcinogenesis 16, 2037-2043. (15) Oda, Y., Yamazaki, H., Watanabe, M., Nohmi, T., and Shimada, T. (1993) Highly sensitive umu test system for the detection of mutagenic nitroarenes in Salmonella typhimurium NM3009 having high O-acetyltransferase and nitroreductase activities. Environ. Mol. Mutagen. 21, 357-364.

Josephy et al. (16) Miller, J. H. (1992) A Short Course in Bacterial Genetics, p 25.4, Cold Spring Harbor Laboratory Press, Plainview, NY. (17) Sandhu, P., Baba, T., and Guengerich, F. P. (1993) Expression of modified cytochrome P450 2C10 (2C9) in Escherichia coli, purification, and reconstitution of catalytic activity. Arch. Biochem. Biophys. 306, 443-450. (18) Omura, T., and Sato, R. (1964) The carbon monoxide-binding pigment of liver microsomes. I. Evidence for its hemoprotein nature. J. Biol. Chem. 239, 2370-2378. (19) Maron, D. M., and Ames, B. N. (1983) Revised methods for the Salmonella mutagenicity test. Mutat. Res. 113, 173-215. (20) Kranendonk, M., Commandeur, J. N., Laires, A., Rueff, J., and Vermeulen, N. P. (1997) Characterization of enzyme activities and cofactors involved in bioactivation and bioinactivation of chemical carcinogens in the tester strains Escherichia coli K12 MX100 and Salmonella typhimurium LT2 TA100. Mutagenesis 12, 245-254. (21) Blattner, F. R., Plunkett, G., III, Mayhew, G. F., Perna, N. T., and Glasner, F. D. GENBANK locus ECAE000243, accession AE000243 (direct submission, Jan. 16, 1997). (22) Watanabe, M., Ishidate, M. J., and Nohmi, T. (1990) Sensitive method for the detection of mutagenic nitroarenes and aromatic amines: New derivatives of Salmonella typhimurium tester strain possessing elevated O-acetyltransferase levels. Mutat. Res. 234, 337-348. (23) Watanabe, M., Igarashi, T., Kaminuma, T., Sofuni, T., and Nohmi, T. (1994) N-hydroxyarylamine O-acetyltransferase of Salmonella typhimurium: Proposal for a common catalytic mechanism of arylamine acetyltransferase enzymes. Environ. Health Perspect. 102 (Suppl. 6), 83-89. (24) Watanabe-Akanuma, M., and Ohta, T. (1994) Effects of DNA repair deficiency on the mutational specificity in the lacZ gene of Escherichia coli. Mutat. Res. 311, 295-304. (25) Watanabe, M., Sofuni, T., and Nohmi, T. (1993) Comparison of the sensitivity of Salmonella typhimurium strains YG1024 and YG1012 for detecting the mutagenicity of aromatic amines and nitroarenes. Mutat. Res. 301, 7-12. (26) Ayrton, A. D., Neville, S., and Ioannides, C. (1992) Cytosolic activation of 2-aminoanthracene: Implications in its use as diagnostic mutagen in the Ames test. Mutat. Res. 265, 1-8. (27) Kranendonk, M., Mesquita, P., Laires, A., Vermeulen, N. P. E., and Rueff, J. (1997) Expression of human cytochrome P450 1A2 in Escherichia coli: A new system for biotransformation and genotoxicity studies of chemical carcinogens. FASEB J. 11, A828.

TX970171Q