Use of Genetically Engineered Salmonella typhimuriumOY1002/1A2 Strain Coexpressing Human Cytochrome

P450 1A2 and NADPH-Cytochrome P450 Reductase andBacterial O-Acetyltransferase in SOS/umu Assay

Pramod Aryal,1,2 Takao Terashita,2 F. Peter Guengerich,3

Tsutomu Shimada,1 and Yoshimitsu Oda1*1Osaka Prefectural Institute of Public Health, Osaka, Japan

2Laboratory of Food Microbiology, Faculty of Agriculture, Kinki University,Nara, Japan

3Department of Biochemistry and Center in Molecular Toxicology,Vanderbilt University School of Medicine, Nashville, Tennessee

The major pathway of bioactivation of procarcino-genic heterocyclic aromatic amines (HCAs) iscytochrome P450 1A2 (CYP1A2)–catalyzedN-hydroxylation and subsequent esterification byO-acetyltransferase (O-AT). We have previouslyreported that an umu tester strain, Salmonella ty-phimurium OY1001/1A2, endogenously coex-pressing human CYP1A2 and NADPH-P450 reduc-tase (reductase), is able to detect the genotoxicityof some aromatic amines [Aryal et al., 1999,Mutat Res 442:113–120]. To further enhance thesensitivity of the strain toward HCAs, we devel-oped S. typhimurium OY1002/1A2 by introduc-ing pCW0/1A2:hNPR (a bicistronic construct co-expressing human P450 1A2 and the reductase)and pOA102 (constructed by subcloning the Sal-monella O-AT gene in the pOA101-expressingumuC0lacZ gene) in S. typhimurium TA1535. Inaddition, as an O-AT–deficient strain, we devel-oped the OY1003/1A2 strain by introducingpCW0/1A2:hNPR and pOA101 into O-AT–defi-cient S. typhimurium TA1535/1,8-DNP. StrainsOY1001/1A2, OY1002/1A2, and OY1003/1A2 expressed, respectively, about 150, 120, and

140 nmol CYP1A2/l culture (in whole cells), andrespective cytosolic preparations acetylated 15, 125,and ^0 nmol isoniazid/min/mg protein as the O-ATactivities of cytosolic preparations, respectively. Wecompared the induction of umuC gene expression asa measure of genotoxicity and observed that theOY1002/1A2 strain was more sensitive thanOY1001/1A2 strain toward the genotoxicity of2-amino-1,4-dimethylimidazo[4,5-f]quinoline (MeIQ),2-amino-3-methylimidazo[4,5-f]quinoline (IQ),2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline(MeIQx), 2-aminoanthracene, 2-amino-6-methyl-dipyrido[1,2-a::3,29-d]imidazole, 3-amino-1,4-dimethyl-5H-pyrido[4,3-b]indole, and 3-amino-1-methyl-5H-pyrido[4,3-a]indole. However, thegenotoxicity of MeIQ, IQ, and MeIQx was notdetected with the OY1003/1A2 strain. Theseresults indicate that the newly developed strainOY1002/1A2 can be employed in detectingpotential genotoxic aromatic amines requiringbioactivation by CYP1A2 and O-acetyltrans-ferase. Environ. Mol. Mutagen. 36:121–126,2000. © 2000 Wiley-Liss, Inc.

Key words: umu test; human P450 1A2; heterocyclic aromatic amines; O-acetyltransferase;Salmonella typhimurium OY1002/1A2; genotoxicity

INTRODUCTION

Heterocyclic aromatic amines (HCAs), which are chem-ical mutagens and carcinogens in rodents [Ohgaki et al.,1991; Wakabayashi et al., 1992], are well known to bepresent in a variety of cooked foods [Sugimura and Sato,1983; Felton et al., 1986; Wakabayashi et al., 1992]. Sincehuman tissues are known to activate HCAs at appreciablelevels [Wakabayashi et al., 1997], they are of great concernfrom the view of health risks to humans. HCAs must bebioactivated by cytochrome P450 1A2 (CYP1A2) toN-hydroxy derivatives [Conney, 1982; Guengerich, 1988;Guengerich and Shimada, 1991, 1998] and acetylated byO-acetyltransferase (O-AT) [Aeschbacher and Turesky,

1991] to confer their mutagenic and carcinogenic proper-ties.

In recent years, different genotoxic assays using bacteriaand mammalian cells, coexpressing human CYP1A2 and

Grant sponsor: Ministry of Education, Science, and Culture of Japan; Grantsponsor: Ministry of Health and Welfare of Japan; Grant sponsor: UnitedStates Public Health Service; Grant numbers: R35 CS44353, P30 ES00267.

*Correspondence to: Yoshimitsu Oda, Osaka Prefectural Institute of PublicHealth, 3-69 Higashinari-ku 1-chome, Osaka 537-0025, Japan. E-mail:ysoda@iph.pref.osaka.jp

Received 5 January 2000; provisionally accepted 15 March 2000; and infinal form 14 April 2000

Environmental and Molecular Mutagenesis 36:121–126 (2000)

© 2000 Wiley-Liss, Inc.

NADPH-P450 reductase (reductase), have been developedto more sensitively detect promutagens and procarcinogens[Davies et al., 1989; Yanagawa et al., 1994; Wu et al., 1997;Josephy et al., 1998; Suzuki et al., 1998; Kranendonk et al.,1999]. Using induction ofumuCgene expression as the SOSresponse to DNA damage, we established anumu test sys-tem withSalmonella typhimuriumOY1001/1A2 coexpress-ing human CYP1A2 and the reductase, and found that theassay system was capable of detecting the genotoxicity ofsome carcinogenic HCAs, without the addition of a meta-bolic activation system, within a short period [Aryal et al.,1999].

In this study, we report the development of a newumutester strain,S. typhimuriumOY1002/1A2, coexpressinghuman CYP1A2, reductase, and overexpressingO-AT. Inaddition, we developed anO-AT–deficient strain,S. typhi-murium OY1003/1A2, coexpressing human CYP1A2 andreductase. The sensitivity of the newly developed strainOY1002/1A2 toward genotoxicity of HCAs was comparedwith the previously established strain OY1001/1A2 and theO-AT–deficient OY1003/1A2 strain.

MATERIALS AND METHODS

Chemicals and Enzymes

Acetyl coenzyme A (Ac-Co A); 2-aminoanthracene (2-AA); 3-amino-1,4-dimethyl-5H-pyrido[4,3-b]-indole (Trp-P-1); 2-amino-3,4-dimethy-limidazo[4,5-f]quinoline (MeIQ); 2-amino-3,8-dimethylimidazo[4,5-f]-quinoxaline (MeIQx); 2-amino-6-methyl-dipyrido[1,2-a::3,29-d]imidazole(Glu-P-1); 2-amino-3-methylimidazo[4,5-f]quinoline (IQ); 3-amino-1-methyl-5H-pyrido[4,3-b]indole (Trp-P-2); isopropylb-D-thiogalactoside (IPTG); di-methylsulfoxide (DMSO); 2-nitrophenylb-D-galactopyranoside (ONPG); anddithiothreitol (DTT) were purchased from Wako Pure Chemicals (Osaka,

Japan).d-Aminolevulinic acid (d-ALA) and isoniazid (INH) were from SigmaChemicals (St. Louis, MO). Restriction enzymes (DraI, PvuII, ScaI, andStuI)were from New England Biolabs (Beverly, MA). T4 DNA polymerase,EcoRI, and ligation kit were from Takara Biomedicals (Shiga, Japan). Allother chemicals and reagents used were of the highest purity grade avail-able.

Bacterial Strains and Plasmids

The bacterial strains and the plasmids employed in the present study areillustrated in Table I.Escherichia coliDH5a, S. typhimuriumTA1535[Ames et al., 1973], TA1535/1,8-DNP [Watanabe et al., 1990], andSJ10002 were used as host strains for transformation. Plasmids pACYC184[Chang and Cohen, 1978] and pCW [Shet et al., 1993] were used as acloning vector.

Construction of plasmids pCW0/1A2:hNPR [Parikh et al., 1997] andpOA101 [Aryal et al., 1999] have been described. Plasmid pOA102 (Fig.1) was constructed by subcloning the DNA fragment (;1.5 kb), carryingthe Salmonella O-AT gene along with the promoter, obtained fromPvuIIdigestion of pYG221 [Oda et al., 1993].

Development ofS. typhimuriumOY1001/1A2 has been described byAryal et al. [1999]. S. typhimuriumOY1002/1A2 andS. typhimurium

Fig. 1. Structure of plasmid pOA102. Tc, tetracycline-resistant gene; Ori,DNA replication origin;O-AT, Salmonella O-acetyltransferase gene;umu-C0lacZ, umuC0lacZ fusion gene under control ofumuoperon.

TABLE I. Bacterial Strains and Plasmids

Bacteria or plasmid Description Source or reference

E. coli DH5a F2, F80dlacZDM15, D(lacZY A-argF)U169, deoR, recA1, endA1, hsdR17(rk2,mk1),pho A, supE44,l2, thi-1, gyrA96, relA1

F.P. Guengerich

S. typhimuriumSJ10002 metA, metE, ilv, xyl, strA, hspLT, hspS, galE K. KutsukakeTA1535 hisG46, ga1,D(chl, uvrB, bio), rfa Ames et al., 1973TA1535/1,8-DNP As TA1535, but deficient inO-acetyltransferase Watanabe et al., 1990TA1535/pSK1002 As TA1535, but harbors pSK1002 Oda et al., 1985OY1001/1A2 As TA1535, but harbors pCW0/1A2:hNPR and pOA101 Aryal et al., 1999OY1002/1A2 As TA1535, but harbors pCW0/1A2:hNPR and pOA102 This studyOY1003/1A2 As TA1535/1,8-DNP, but harbors pCW0/1A2:hNPR and pOA101 This study

PlasmidpSK1002 umuD1-umuC0lacZ, Ampr Shinagawa et al., 1983pYG221 As bluescript KSM13, but has 1.35-kb fragment betweenEcoRV site (6.5-kb) andBamHI

(7.65-kb) of pYG122 containingSalmonella O-acetyltransferase geneOda et al., 1993

pACYC184 Tetr, Cmr Chang and Cohen, 1978pOA101 As pACYC184, but hasumuC0lacZ gene containingEcoRI-StuI-digested 9.91-kb fragment

of pSK1002 subcloned atDraI site in the chloramphenicol gene, TetrAryal et al., 1999

pOA102 As pOA101, but has bacterialO-acetyltransferase gene subcloned atDraI site andScaIsite, respectively, in the chloramphenicol gene, Tetr

This study

pCW Ampr Shet et al., 1973pCW0/1A2:hNPR As pCW, but has human P450 1A2 and reductase in a bicistronic construct, Ampr Parikh et al., 1997

122 Aryal et al.

OY1003/1A2 were developed by introducing pCW0/1A2:hNPR andpOA102, and pCW0/1A2:hNPR and pOA101 intoS. typhimuriumTA1535andS. typhimuriumTA1535/1,8-DNP strains, respectively. The method oftransformation of strains was as described by Maniatis et al. [1989].

Determination of CYP1A2 and O-AT Activitiesin S. typhimurium Strains

Expression of CYP1A2 inS. typhimuriumstrains was carried out asdescribed [Sandhu et al., 1994; Aryal et al., 1999] with slight modification.Induction of the doubletac promoter was initiated by addition of 0.2 mMIPTG and the level of expression was measured as described by Omura andSato [1964], after an 18-hr incubation at 28°C with shaking (200 rpm).

Because it has been reported that bacterial acetyltransferase exhibitsN-or O-acetyltransferase activity [Watanabe et al., 1990], expression ofO-ATwas determined by measuringN-acetyltransferase activity by the cytosolicpreparations ofS. typhimuriumstrains as described by Hein et al. [1982].Protein content was measured as described by Lowry et al. [1951].

umu Assay Using S. typhimurium Tester Strains

We used the method described by Aryal et al. [1999] with slightmodification. The overnight culture of tester strains was diluted 100-fold inTGlyT medium (1% Bacto tryptone, 0.5% NaCl, 0.2% glycerol, and 1mgtetracycline/ml, 1.0 mM IPTG, 0.5 mMd-ALA, and 250 ml of traceelements/l [Sandhu et al., 1994]). Induction of theumuCgene was deter-mined by measuring cellularb-galactosidase activity according to Oda etal. [1985].

RESULTS

Development of umu Tester Strains

To overexpressSalmonella O-acetyltransferase (O-AT),we developed the plasmid pOA102 (Fig. 1) by subcloningthe O-AT gene at theScaI site of pOA101. After recoveryof pOA102 and pCW0/1A2:hNPR plasmids fromE. coliDH5a, they were modified by transforming intoS. typhi-murium SJ10002 (restriction2, modification1), and thenwere introduced intoS. typhimuriumTA1535. Tetracycline/ampicillin-resistant transformants thus obtained were desig-nated asS. typhimuriumOY1002/1A2. In a similar manner, theOY1003/1A2 strain was constructed by transforming theO-AT–deficient S. typhimuriumTA1535/1,8-DNP strainwith pOA101 and followed by pCW0/1A2:hNPR.

Expression of CYP1A2 and O-AT Proteinsin S. typhimurium Strains

Expression of CYP1A2 protein inS. typhimuriumstrainswas determined in whole cells by the FezCO differencespectra. The expression level of CYP1A2 was found to beapproximately 150, 120, and 140 nmol/l cell culture ofOY1001/1A2, OY1002/1A2, and OY1003/1A2, respec-tively. CYP1A2 and reductase activity of membrane prep-arations from the three strains did not show any significantdifference from earlier reports for OY1001/1A2 [Aryal etal., 1999] (data not shown).

Expression ofO-acetyltransferase in different strains wasdetermined by measuring the INH acetylation by cytosolicpreparation. Strains OY1001/1A2 and OY1002/1A2 werefound to acetylate 15 and 125 nmol INH/min/mg protein,respectively. However, OY1003/1A2 was not found to acet-ylate INH in any detectable amount.

Detection of Cytotoxicity and Genotoxicity ofProcarcinogens in S. typhimurium Strains

To examine the potency and sensitivity of newly estab-lished S. typhimuriumstrains toward the cytotoxicity andgenotoxicity of procarcinogenic HCAs, different concentra-tions of MeIQ, IQ, and MeIQx were incubated with Salmo-nella strains and induction ofumuCgene expression wasdetermined. A dose-dependent increase in the induction ofumuCgene expression was observed in OY1001/1A2 andOY1002/1A2 strains. More enhanced induction ofumuCgene expression was observed in OY1002/1A2 than inOY1001/1A2 at equimolar concentrations of the com-pounds (Fig. 2, lower panel). However, cytotoxicity towardOY1002/1A2 was observed on increasing the concentrationof the test compounds (Fig. 2, upper panel). In contrast withOY1001/1A2 and OY1002/1A2, induction ofumuC geneexpression inO-AT–deficient strain OY1003/1A2 was ob-served to be within the range of background level at thedoses used (Fig. 2, lower panel).

To further examine whether the newly establishedO-AT–overexpressing OY1002/1A2 strain is highly sensitive to-ward the genotoxicity of some aromatic amines, we com-pared the induction ofumuCgene expression by Glu-P-1,Trp-P-1, Trp-P-2, and 2-AA in the OY1001/1A2 andOY1002/1A2 strains (Fig. 3). Dose-dependent increases ininduction of umuC gene expression by Glu-P-1, Trp-P-2,and 2-AA were observed in OY1002/1A2, although nosignificant induction was observed in OY1001/1A2 (induc-tion of b-galactosidase was about 1.5 times the backgroundlevel at the highest dose). However, a dose-dependent in-crease in induction ofb-galactosidase activity by Trp-P-1was observed in both strains, although induction ofb-ga-lactosidase activity in strain OY1002/1A2 was about two-fold greater than in strain OY1001/1A2 at high dose. Cy-totoxicity toward both strains at doses used was notobserved, although slight cytotoxicity (10%) was observedwith 40 mM 2-AA (data not shown). The genotoxic poten-tial of heterocyclic aromatic amines and 2-AA used in thepresent study was found to be in the order of MeIQ. IQ .MeIQx . Glu-P-1. 2-AA . Trp-P-1. Trp-P-2.

DISCUSSION

To increase the sensitivity of tester strain, in this study weintroduced two plasmids, one containing human CYP1A2and the reductase genes, and the other containingSalmo-nella O-AT and umuC0lacZ fusion genes, into theS. typhi-

Coexpression of Human P450 and O-AT 123

murium TA1535 strain and established a new tester strainOY1002/1A2. We examined the induction ofumuCgeneexpression to see whether the newly established tester strainwould exhibit a more enhanced sensitivity toward procar-cinogenic HCAs than either OY1001/1A2, expressing hu-man CYP1A2 and the reductase [Aryal et al., 1999], or theO-AT–deficient strain OY1003/1A2. The results showedthat the new strain OY1002/1A2 is very sensitive to HCAs.Thus, our study indicates that the genotoxicity of HCAsdepends not only on the CYP1A2 enzyme but also on theO-AT enzyme. To our knowledge, this is the first instance inwhich an SOS/genotoxicity assay, with endogenous expres-sion of human P450 andO-AT enzymes, has been employedfor the detection of promutagenic HCAs.

We compared the CYP1A2 contents andO-AT activity inthree strains. The CYP1A2 contents inS. typhimuriumOY1001/1A2, OY1002/1A2, and OY1003/1A2 strains werealmost similar and the present result is in good agreementwith that of Josephy et al. [1995], in which expression ofCYP1A2 inS. typhimuriumDJ4501A2 is reported to be 170nmol/liter. Expression ofO-AT varied among the strains.The cytosol ofO-AT–overexpressing strain OY1002/1A2exhibited about eightfold higher INH-N-acetylation thanthat of the OY1001/1A2 strain. These results indicate thatCYP1A2 andO-AT are sufficiently expressed in OY1002/1A2.

As expected, strain OY1002/1A2 showed higher sensi-tivities toward the genotoxicity of MeIQ, IQ, and MeIQxthan did strain OY1001/1A2 (Fig. 2). In comparing thesensitivities (the concentrations of chemicals that induceb-galactosidase activity by twofold over the backgroundlevel), the OY1002/1A2 strain exhibited six- to 21-foldhigher sensitivities toward MeIQ, IQ, and MeIQx than didthe OY1001/1A2 strain. Similarly, OY1002/1A2 exhibitedenhanced sensitivity toward the genotoxicity of Trp-P-1. Inaddition, the genotoxicity of 2-AA, Glu-P-1, and Trp-P-2was detected only in the OY1002/1A2 strain (Fig. 3). Theseresults clearly support the view that the OY1002/1A2 straincan metabolically activate aromatic amines to DNA-dam-aging products that induce the SOS response to a greatextent. The genotoxic potencies of HCAs detected in thepresent study exhibited a pattern similar to the results ofJosephy et al. [1998] forE. coli DJ4309 and Suzuki et al.[1998] for S. typhimuriumTA1538/AR strains, expressinghuman CYP1A2 and the reductase, andO-AT. However,the sensitivity of the OY1002/1A2 strain to these chemicalswas considerably lower than that observed with theO-AT–overexpressing strain NM2009, exogenously supplementedwith rat S9 fraction [Oda et al., 1995]. This may be ex-plained by the difference in the amount of P450 present inthe S9 mix and expressed in the OY1002/1A2 strain.

Most aromatic amines are known to undergoO-acetyla-

Fig. 2. Effect of MeIQ (A), IQ (B), and MeIQx (C) on bacterial growth (upper panel) and induction ofumuCgene expression inS. typhimuriumOY1001/1A2 (f), OY 1002/1A2 (F), and OY1003/1A2 (Œ). Bars indicate SD (means of triplicate independent experiments).

124 Aryal et al.

tion by recombinant humanN-acetyltransferase enzymes(NAT1 and NAT2) [Grant, 1993]. These enzymes are re-sponsible for “Phase II” activation of procarcinogenic aro-matic amines to exert genotoxicity in bacteria and mamma-lian cells [Grant et al., 1992; Watanabe et al., 1994;Yanagawa et al., 1994; Wild et al., 1995; Wu et al., 1997].We previously reported that the NM6001 and NM6002strains, overexpressing human NAT1 or NAT2, detect thegenotoxicity of aromatic amines with higher sensitivities[Oda et al., 1999]. Therefore, the present test system couldbe further developed to understand possible mechanisms ofbioactivation of procarcinogenic HCAs by human enzymesin interpreting the risk to humans by introducing humanNAT cDNA(s) in combination with human CYP1A2 andthe reductase.

The mutagenicity tests that employ bacterial strains, ex-pressing human CYP1A2 and the reductase, and bacterialO-AT [Josephy et al., 1998; Suzuki et al., 1998; Kranen-donk et al., 1999], and our present test system can beemployed as biological tools to detect genotoxic HCAs. Inaddition, ourumusystem is very simple and could be usedto examine the genotoxicity of potential HCAs in a rela-tively shorter period with fewer complications than in mu-tagenicity assays, depending on the load of sample to beexamined. Furthermore, we believe that thisumu assaycould be applied to the expression of other forms of humanP450 and the detection of other classes of promutagens andprocarcinogens.

In conclusion, we showed here that the genetically engi-neeredumutester strain OY1002/1A2, coexpressing humanCYP1A2 and NADPH-P450 reductase, provides an excel-lent test system not only for detecting potential promuta-genic aromatic amines but also for analyzing the catalyticspecificity and metabolic activation of putative carcinogeniccompounds.

REFERENCES

Aeschbacher HU, Turesky RJ. 1991. Mammalian cell mutagenicity andmetabolism of heterocyclic aromatic amines. Mutat Res 259:235–250.

Ames BN, Lee FD, Durston WE. 1973. An improved bacterial test systemfor the detection and classification of mutagens and carcinogens.Proc Natl Acad Sci USA 70:782–786.

Aryal P, Yoshikawa K, Terashita T, Guengerich FP, Shimada T, Oda Y.1999. Development of a new genotoxicity test system withSalmo-nella typhimuriumOY1001/1A2 expressing human CYP1A2 andNADPH-P450 reductase. Mutat Res 442:113–120.

Chang ACY, Cohen SN. 1978. Construction and characterization of am-plifiable multicopy DNA cloning vehicles derived from the p15Acryptic miniplasmid. J Bacteriol 134:1141–1156.

Conney AH. 1982. Induction of microsomal enzymes by foreign chemicalsand carcinogenesis by polycyclic aromatic hydrocarbons [G.H.A.Clowes lectures]. Cancer Res 42:4875–4917.

Davies R, Crespi CL, Rudo K, Turner TR, Langenbach R. 1989. Devel-opment of a human cell line by selection and drug-metabolizinggene transfection with increased capacity to activate promutagens.Carcinogenesis 10:885–891.

Felton JS, Knize MG, Shen NH, Anderson BD, Bjeldones LF, Hatch FT.

Fig. 3. Induction ofumuCgene expression inS. typhimuriumOY1001/1A2 (f) and OY1002/1A2 (F) by Glu-P-1 (A), 2-AA (B), Trp-P-1 (C), andTrp-P-2 (D). Bars indicate SD (means of triplicate independent experiments).

Coexpression of Human P450 and O-AT 125

1986. Identification of the mutagens in cooked beef. Environ HealthPerspect 67:17–24.

Grant DM. 1993. Molecular genetics ofN-acetyltransferases. Pharmaco-genetics 3:45–50.

Grant DM, Josephy PD, Lord HL, Morrison LD. 1992.Salmonella typhi-murium strains expressing human arylamineN-acetyltransferasemetabolism and mutagenic activation of arylamines. Cancer Res52:3961–3964.

Guengerich FP. 1988. Roles of cytochrome P450 enzymes in chemicalcarcinogenesis and cancer chemotherapy. Cancer Res 48:2946–2954.

Guengerich FP, Shimada T. 1991. Oxidation of toxic and carcinogenicchemicals by human cytochrome P450 enzymes. Chem Res Toxi-col 4:391–407.

Guengerich FP, Shimada T. 1998. Activation of procarcinogens by humancytochrome P450 enzymes. Mutat Res 400:201–213.

Hein DW, Hirata M, Glowinski IB, Weber WW. 1982. Biochemicalevidence for the coexistence of monomorphic and polymorphicN-acetyltransferase activities on a common protein in rat liver.J. Pharmacol Exp Ther 220:1–7.

Josephy PD, DeBruin LS, Lord HL, Oak JN, Evans DH, Guo Z, Dong MS,Guengerich FP. 1995. Bioactivation of aromatic amines by recom-binant human cytochrome P450 1A2 expressed in Ames testerstrain bacteria: a substitute for activation by mammalian tissuepreparation. Cancer Res 55:799–802.

Josephy PD, Evans DH, Parikh A, Guengerich FP. 1998. Metabolic acti-vation of aromatic amine mutagen by simultaneous expression ofhuman P450 1A2, NADPH-cytochrome P450 reductase, andN-acetyltransferase inEscherichia coli.Chem Res Toxicol 11:70–74.

Kranendonk M, Fisher CW, Roda R, Carriera F, Theisen P, Laires A, RueffJ, Vermeulen NPE, Estabrook RW. 1999.Escherichia coliMTC, aNADPH-cytochrome P450 reductase competent mutagenicity testerstrain for the expression of human cytochrome P450: comparison ofthree types of expression systems. Mutat Res 439:287–300.

Lowry OH, Rosenbrough NJ, Farr AL, Randal RJ. 1951. Protein measure-ment with the Folin phenol reagent. J Biol Chem 193:265–275.

Maniatis T, Fritsch EF, Sambrook J. 1989. Molecular cloning: a laboratorymanual. Cold Spring Harbor, NY: Cold Spring Harbor LaboratoryPress.

Oda Y, Nakamura S, Oki I, Kato T, Shinagawa H. 1985. Evalution of anew test system (umu-test) for the detection of environmentalmutagens and carcinogens. Mutat Res 47:219–229.

Oda Y, Yamazaki H, Watanabe M, Nohmi T, Shimada T. 1993. Highlysensitiveumu test system for the detection of nitroarenes inSal-monella typhimuriumNM3009 having highO-acetyltransferase andnitroreductase activities. Environ Mol Mutagen 21:357–364.

Oda Y, Yamazaki H, Watanabe M, Nohmi T, Shimada T. 1995. Develop-ment of high sensitiveumu test system: rapid detection of geno-toxicity of promutagenic aromatic amines bySalmonella typhi-murium strain NM2009 possessing highO-acetyltransferaseactivity. Mutat Res 334:145–156.

Oda Y, Yamazaki H, Shimada T. 1999. Role of humanN-acetyltrans-ferases, NAT1 or NAT2, in genotoxicity of nitroarenes and aro-matic amines inSalmonella typhimuriumNM6001 and NM6002.Carcinogenesis 20:1079–1083.

Ohgaki H, Takayama S, Sugimura T. 1991. Carcinogenicities of hetero-cyclic amines in cooked food. Mutat Res 259:399–410.

Omura T, Sato R. 1964. The carbon monoxide-binding pigment of livermicrosome. II. Solubilization, purification, and properties. J BiolChem 239:2379–2387.

Parikh A, Gillam EMJ, Guengerich FP. 1997. Drug metabolism byEsch-erichia coli expressing human cytochromes P450. Nat Biotechnol15:784–788.

Sandhu P, Guo Z, Baba T, Martin MV, Tukey RH, Guengerich FP. 1994.Expression of modified human cytochrome P450 1A2 inEsche-richia coli: stabilization, purification, spectral characterization, andcatalytic activities of the enzyme. Arch Biochem Biophys 309:168–177.

Shet MS, Fisher CW, Holmans PL, Estabrook RW. 1993. Human cyto-chrome P450 3A4: enzymatic properties of a purified recombinantfusion protein containing NADPH-P450 reductase. Proc Natl AcadSci USA 90:11748–11752.

Shinagawa H, Kato T, Ise T, Makino K, Nakata A. 1983. Cloning andcharacterization ofumuoperon responsible for inducible mutagen-esis inEscherichia coli.Gene 23:167–174.

Sugimura T, Sato S. 1983. Mutagens—carcinogens in foods. Cancer Res43(Suppl):2415s–2421s.

Suzuki A, Kushida H, Iwat H, Watanabe M, Nohmi T, Fujita K, GonzalezFJ, Kamataki T. 1998. Establishment of aSalmonellatester strainhighly sensitive to mutagenic heterocyclic amines. Cancer Res58:1833–1838.

Wakabayashi K, Nagao M, Esumi H, Sugimura T. 1992. Food-derivedmutagens and carcinogens. Cancer Res 52(Suppl):2092s–2098s.

Wakabayashi K, Totsuka Y, Fukutome K, Oguri A, Ushiyama H, SugimuraT. 1997. Human exposure to mutagenic/carcinogenic heterocyclicamines and comutagenic betacarbolines. Mutat Res 376:253–259.

Watanabe M, Ishidate M Jr, Nohmi T. 1990. Sensitive method for thedetection of mutagenic nitroarenes and aromatic amines: new de-rivatives of Salmonella typhimuriumstrains possessing elevatedO-acetyltransferase levels. Mutat Res 234:337–348.

Watanabe M, Matsuoka A, Yamazaki N, Hayashi M, Deguchi T, Nohmi T,Sofuni T. 1994. New sublines of Chinese hamster CHL stablyexpressing human NAT1 or NAT2N-acetyltransferases orSalmo-nella typhimuriumO-acetyltransferase: comparison of the sensitiv-ities to nitroarenes and aromatic amines using thein vitro micro-nucleus test. Cancer Res 54:1672–1677.

Wild D, Feser W, Michel S, Lord HL, Josephy PD. 1995. Metabolicactivation of heterocyclic amines catalyzed by human arylamineN-acetyltransferase isozymes (NAT1 and NAT2) expressed inSal-monella typhimurium.Carcinogenesis 16:643–648.

Wu RW, Tucker JD, Sorensen KJ, Thompson LH, Felton JS. 1997.Differential effect of acetyltransferase expression on the genotox-icity of heterocyclic amines in CHO cells. Mutat Res 390:93–103.

Yanagawa Y, Sawada M, Deguchi T, Gonzalez F, Kamataki T. 1994.Stable expression of human CYP1A2 andN-acetyltransferases inChinese hamster CHL cells: mutagenic activation of 2-amino-3-methylimidazo[4,5-f ]quinoline and 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline. Cancer Res 54:3422–3427.

Accepted by—M. Prival

126 Aryal et al.