Tobacco Documents Online

HUTAT RE$-FUND MOL tq 97 ~- tO~ELSEVZF..~ SCZENCE 8V NE ~ Fundamental and Molecular Mechanisms of Mutagenesis ELSEVIER Mutation Research 376 (1997) 135-142 Polymorphisms of CYP1A1 and GSTM1 influence the in vivo function of CYPIA2 Stewart MacLeod a, Rashmi Sinha b, Fred F. Kadlubar c, Nicholas P. Lang d,* University of Arkansas for Medical Sciences. Little Rock, AR 72205, USA b National Cancer Institute, NIH, Bethesda, MD, USA = National Center for Toxicological Research, Jefferson, AR, USA a JLM Veterans Affairs Medical Center and Unicersity of Arkansas for Medical Sciences, Little Rock, AR 72205, USA Accepted 18 June 1996 Abstract - Differences in human cancer susceptibility have been attributed to polymorphisms of carcinogen metabolizing enzymes. Our efforts have focused on the systems responsible for metabolism of aromatic and heterocyclic amines found in cigarette smoke and in cooked foods. Cytochrome P4501A2 (CYP1A2), which catalyzes aromatic and heterocyclic amine N-oxida- tion, has been implicated as a risk factor in both urinary bladder and colorectal cancer. In the present study we used the results of caffeine to the effects of cigarette smoke and in meat phenotyping experiments measure compounds present cooked at high temperature on CYP1A2 activity. Subjects in the smoking cessation study had mean CYP1A2 activity of 17.8 (expressed as the urinary molar ratio of [17X + 17U]/137X) while smoking; however, this activity decreased to 10.9 three weeks after cessation of smoking. Subjects in the cooked meat feeding study had mean CYP1A2 activity of 9.01 after 1 week of consuming meat cooked at low temperature, but this value increased to 12.7 after I week of consuming meat cooked at high temperature. Because no association has been identified between differences in CYP1A2 activity and variations in the CYP1A2 structural gene, we sought to determine whether the activities of other carcinogen metabolizing enzymes are involved in the regulation of CYP1A2 activity. CYP1A2 activity was higher in individuals who express the GSTM 1 null allele compared to those expressing the GSTMI*A,B allele, 10.2 vs. 8.5 for unexposed conditions and 15.0 vs. 12.3 for exposed conditions. CYPIA1 genotyping demonstrated that individuals possessing the Ile/Ile CYP1A1 genotype had greater mean CYP1A2 activity than those who had the heterozygous Ile/Val alletic variant of the CYP1A1 However, to gene. upon exposure cigarette smgke or high-temperature cooked meat, individuals possessing the heterozygous form of the CYPIA1 gene had significantly increased CYP1A2 activity (18.1) compared to those with the more common Ile/Ile CYP1A1 genotype (I 3.3). These results indicate that CYPIA2, CYP1A1, and GSTMI gene-gene interactions could be important confounders in the interpretation of molecular epidemiology studies. Keywords: Cytochrome P4501A1; Cytochrome P4501A2; Glutathione S-transferase M1; Polymorphism Abbreviations: CYP1A2: Cytochrome P4501A2; CYPIAI: Cytocbxome P4501A1; NAT2: N-acetyltransferase-2; 17X: 1,7-dimethyl xanthine; 17U: 1,7-dimethyl uracil; 137X: caffeine; GSTMI: Glutathione S-tr~nsferase M1; Ile/Val: Isoleucine/Valine; Ile/Ile: Isoleucine/Isoleucine; GSH: glutathione • Corresponding author. Tel.: + I 501-660-2038; Fax: + 1 501-671-2523; E-mail: smacleod@acre.uams.edu 0027-5107/97/$17.00 Copyright © 1997 Elsevier Science B.V. All rights reserved. PII S0097-5107(97)00036-5 THIS ARTICLE IS ~=OR IN - DIVIDUAL USE ONLY AND MAY NOT BE FURTHER REPRODUCED OR STORED ELECTRONICALLY NITHOUT I~R~'TTEN PERNISSION FROH THE COPYRIGHT HOLDER. • UNAUTHORIZED REPRODUCTION HAY RESULT IN FINANCIAL AND OTHER PENALTIES.

136 S. MacLeod et al. / Mutation Research 376 (1997) 135-142 1. Introduction Current evidence makes clear the inaccuracies of risk assessment following exposure to a chemical carcinogen without a concomitant evaluation of the metabolic pathways of activation and detoxification. These pathways exhibit both species and tissue specificity that may seriously weaken the role of extrapolation of data in one system to risk projec- tions in another tissue or species. There may also be interactions among these pathways that further com- plicate interpretation of risk data. Enzymes involved in the metabolism of carcinogens have been shown to exhibit a number of different polymorphisms that can affect their expression levels or catalytic activity. Nucleotide variations in the coding regiorr of a gene that lead to an amino acid substitution in the protein can alter enzymatic activity or substrate binding in a positive or negative manner (Gonzalez, 1989). Dele- tion of all or part of the coding region can lead to the production of an inactive enzyme (Blum et al., 1989) or can result in a total lack of protein synthesis (Seidegard et al., 1988). Polymorphisms in the non- coding region can affect transcriptional control ele- ments involved not only in basal enzyme expression (Nakachi et al., 1991) but can also disrupt the mech- anism of induction observed for many of these en- zymes (Gonzalez and Gelboin, 1993). Additionally, variations in the polyadenylation signal of a gene can result in changes in transcript polyadenylation thus affecting transcript half-life and indirectly affecting the quantity of enzyme produced (Bell et al., 1995). In this study we have focused on the enzymes which metabolize aromatic and heterocyclic amines found in cigarette smoke and in meat cooked at high temperatures.. Human exposure to these compounds has been implicated as a risk factor for colorectal cancer (Lang et al., 1994). Heterocyclic and aromatic amines are initially N-oxidized by the action of cytochrome P4501A2 (CYP1A2) in the liver (Ture- sky et al., 1991). As shown in Fig. 1, the resulting N-hydroxy heterocyclic amine can be O-acetylated in the liver, then detoxified by conjugation with glutathione (GSH) by the action of GSH transferases (GSTs) or glucuronidated by the action of UDP- glucuronosyltransferases. Alternatively, the N-hy- droxy heterocyclic amines can enter the bloodstream and be transported to target organs, where they can be O-acetylated in tissues containing N-acetyltrans- ferase-2 (NAT2). This N-acetoxy derivative can re- act with DNA to form covalent heterocyclic amine- DNA adducts (Flammang et al., 1987). The forma- tion of DNA adducts and the extent of tumor induc- tion is dependent on the amount of aromatic or heterocyclic amines which are converted to the reac- tive metabolite. The amount produced depends on Fig. I. Proposed metabolic pathway for metabolism of heterocyclic amines by humans. I--I

S. MacLeod et al. / M,~tation Research 376 (1997) 135-142 137 the quantity and activity of the enzymes involved in its formation, CYPIA2 and NAT2, and also the involved in detoxification, such as GSTs enzymes and UDP-glucuronosyltransferases. The activities of CYP1A2 and NAT2 both exhibit polymorphic distributions in human populations. These polymorphisms can be detected by phenotyp- ing assays that measure the ratio of caffeine metabo- lites and reflect the activities of CYP1A2 and NAT2 (Grant et al., 1984; Butler et al., 1992). Individuals are classified as slow or rapid N-acetylators or N- oxidizers, depending on their activity levels of NAT2 or CYP1A2, respectively. These assays have allowed the assessment of colorectal cancer risk by the deter- mination of NAT2/CYP1A2 phenotype in relation to dietary, exposure to heterocyclic amines from well-done cooked meat. The odds ratio for colorectal cancer ranges from 1.0 for slow metabolizers who eat rare or medium-cooked meat, to 6.5 for rapid metabolizers who eat well-done meat, These results support the hypothesis that an individual's cancer risk is related to individual carcinogen exposure and to polymorphisms in levels of carcinogen metaboliz- ing enzymes (Lang et al., 1994). Although differences in both constitutive and in- duced CYP1A2 function have been reported, no association has been identified between CYP1A2 activity and variations in the structural gene. Early evidence for CYPIA2 polymorphism involved inter- individual differences in the metabolism of phenacetin, an analgesic drug (Shahidi, 1968; AI- varez et al., 1979; Devonshire et al., 1983). Later studies using caffeine metabolizing phenotype assays demonstrated 40-60 fold variations in CYP1A2 ac- tivity between individuals (reviewed in Butler et al., 1989). Family studies of CYP1A2 activity indicate that variations in individual CYP1A2 activity appear to be under genetic control (Kadlubar, 1994). A number of studies have also demonstrated indi- vidual differences in the inducible component of CYP1A2 activity. Components of cigarette smoke (probably polycyclic aromatic hydrocarbons) induce CYP1A2 activity to a different extent in various smokers (Butler et a1., 1992). Consumption of pan- fried meat containing high levels of heterocyclic aromatic amines also increased CYPIA2 activities by varying degrees in non-smoking individuals (Sinha et al., 1994). A previous study by our laboratory found a marked racial difference in the effect of cigarette smoking on CYPIA2 activity. Caucasian smokers were found to have CYPIA2 levels nearly twice that of non-smokers while African-American smokers had CYP1A2 activity approximating that of African-American or Caucasian non-smokers (Lang et al., 1994). These results suggest genetically based differences in the response of CYPIA2 activity to xenobiotic exposure. Because a link between CYP1A2 activity and CYPIA2 gene variation has not been established (Nakajima et al., 1994), we initiated a study of the relationship between CYP1A2 activity and the activities of other polymorphic en- zymes that are involved in the metabolism of hetero- cyclic aromatic amines and polycyclic aromatic hy- drocarbons. We first examined the effects of cigarette smoking on CYP1A2 activity by measuring CYP1A2 function in smokers before and 3 weeks after cessa- tion of smoking. The second study measured the effects of consumption of well-done pan-fried meat on CYPIA2 activity in non-smokers. In addition, we examined the interaction between the CYP1A2 phe- notype and CYPIA1 genotype in these same individ- uals, since expression of these two enzymes is closely linked in all animal species examined and the Ile/Val CYPIA1 genotype in humans has been shown to result in higher inducibility of CYPIA1 activity (Cosma et al., I993; Crofts et al., I994). The GSTM1 genotype was also examined, since the absence of this gene has been previously associated with higher CYP1A2 activity in smokers, but not in non-smokers (Bartsch et al., 1995) and higher levels of aromatic amine-DNA adducts are found in smokers who pos- sess the GSTM1 null alleles (Yu et al., 1995). The GSTM1 genotype has also been reported to be asso- ciated with high inducibility of CYP1A1 expression in cultured human cells (Vaury et at., 1995). Thus, we evaluated the influence of the recently discovered polymorphisms in the CYPIA1 and GSTM1 genes on CYP1A2 activity in individuals in both the feeding and smoking cessation studies. 2. Materials and methods 2.1. Smoking cessation study Under a protocol approved by the University of Arkansas for Medical Sciences Human Research

138 S. MacLeod et al. / Mutation Research 376 (1997) 135-142 Committee, subjects were recruited from Stop Smok- ing programs in Little Rock, AR. Caffeine phenotyp- ing was performed at entry into the study while the subject was still actively smoking. After complete smoking cessation for 3 weeks, the subjects were phenotyped again. Ten ml of blood was drawn for genotyping studies at the time of phenotyping. The data in this report comes from the 20% (N = 22) of the entry subjects that were successful in completing the 3 weeks of smoking abstinence. 2.5. GSTM1 genotype GSTMI genotype was determined by the use of a PCR based assay as described by Bell et al. (1993). Since this assay results in the absence of a PCR product in the case of individuals with the GSTMI null phenotype, oligonucleotide primers which am- plify part of the 13-globin gene were employed in a multiplex PCR reaction as a positive control for DNA quality and PCR reaction conditions. 2.2. Pan-fried meat feeding study 2.6. PCR primers Study population and selection criteria were pre- viously described (Sinha et al., 1994). The feeding study consisted of two consecutive 7-day controlled dietary periods which differed only in the method of meat preparation. In the first dietary period, subjects consumed evening meals that included meat cooked at low (I00°C) temperature; while during the 2nd week, evening meals contained pan-fried meat cooked at high (250°C) temperature. Breakfast and lunch meals were provided so that the differences between week 1 and week 2 were limited to the consumption of meat cooked at low or high tempera- tures, respectively. 2.3. Phenotype determination The CYP1A2 phenotype of dietary study subjects was determined at the end of week I and again at the end of week 2 as described by Sinha et al. (1994). Subjects were administered 114 mg of caffeine in 3.6 g of instant coffee in 266 ml of water. Subjects voided after 4 h and urine was collected at 5 h. Caffeine and its metabolites were analyzed by the method described by Butler et al. (1992). CYP1A2 phenotype was calculated by comparing the ratio of (1,7-dimethylxanthine (17X) plus 1,7-dimethyluric acid (17U)) to 1,3,7-trimethylxanthine (137X). GSTM1 forward primer 5'GAACTCC- CTGAAAAGCTAAAGC 3', reverse primer 5'GT- TGGGCTCAAATATAGCGTGG 3'. 13-globin for- ward primer 5'CAACTI'CATCCACGTTCACC 3'. 13-globin reverse primer 5'GAAGAGCCAAGGA- CAGGTAC 3'. 2.7. CYP1A1 genotype determination The single base pair A/G polymorphism in exon 7 of the CYPIAI gene, which results in an lle/Val polymorphism in the protein, was detected by a modification of the method described by Hayashi et al. (1991a). Two different downstream primers con- taining the polymorphic A or G at their 3' terminus were used in separate reactions with a common upstream primer. When used in PCR reactions, these primers yield products only when the template DNA is complimentary to the terminal 3' base. Reaction conditions were as described by Hayashi et al. (1991b), except for the addition of TaqStart antibody (Clontech, Palo Alto, CA) to the reaction mix and an increase in annealing temperature to 70°C. These modifications served to eliminate false positive PCR products caused by primer annealing to non-compli- mentary templates. 2.4. DNA isolation from. lymphocytes 3. Statistical analysis Genomic DNA was isolated from lymphocytes by the proteinase K digestion and phenol-chltSroform extraction method described by Blin and Stafford (1976). Statistical calculations were performed using JMP 3.1 (SAS Institute Inc., Cary, NC) for the Macintosh (Apple Computer, Inc., Cupertino, CA). Means, standard deviations, standard error and t-test were

] i i i i i I i I i i 1 ] ] S. MacLeod et aL/ Mutation Research 376 (1997) 135-142 performed on CYPIA2 activity comparing unex- posed to exposed values for individuals who were GSTMI'A,B vs. GSTMI*0 genotype and individu- als who were CYP1A1 Val/Val versus those who were CYP1A1 Ile/Val genotype. 4. Results Caffeine phenotyping studies using the molar ra- tio of (l,7-dimethyl xanthine + 1,7-dimethyluric acid)/1,3,7-trimethyl xanthine (( 17X + 17 U)/137X) indicated that the average post-smoking CYPIA2 activity decreased to approximately 50% of levels noted during smoking (Fig. 2). In the Pan-Fried Meat feeding study that mea- sured CYP1A2 function (Sinha et al., 1994), it was found that 47 out 65 individuals (72%) had increased CYPIA2 activity after 1 week of consuming meat cooked at high temperatures, as compardd with the previous week of consuming an identical diet except that the meat was cooked at low temperatures (week 2 vs. week 1). However, 18 subjects or 28% of the test population did not exhibit increased CYP1A2 activity after the week of consumption of high-tem- perature cooked meat. The mean CYP1A2 activities were unchanged between those measured upon entry into the study (week 0, data not shown), and after 1 week of consuming low-temperature cooked meat (week 1 or unexposed). After 1 week of consuming Smoking Cessation Study (N-~) Post-Smoking Smoking Fig. 2. The influence of cigarette smoking on CYP1A2 activity. Mean (~--), median (~), 10th, 25th, 75th, 90th per- centile CYP1A2 activity as measured upon entry into the study (smoking) and after 3 weeks of smoking cessation (post-smoking). 35- 30- 2o~ .to- Pen-Fried Meat Feeding Study (N--66) ~ p---o.OOOl ~ 139 Low-Temp. High-Temp. Cooked Meat Cooked Meat Fig. 3. The influence of the consumption of meat cooked at high temperature on CYPIA2 activity. Mean (----), median (~), 10th, 25th. 75th, 90th percentile CYP1A2 activity as measured at the end of week 1 (low-temp. cooked meat) and at the end of week 2 (high-temp. cooked meat) of the pan-fried meat feeding study. high-temperature cooked meat (week 2 or exposed), mean CYP1A2 activity was increased to levels 50% higher than in the previous 2 weeks (Fig. 3). Fig. 4 shows that CYP1A2 activity was somewhat lower in the unexposed stages of the combined feed- ing study and smoking cessation study when the heterozygous Ile/Val form of CYP1A1 was present. Unexposed (Combined Groups) Exposed (Combined Groups Fig. 4. The influence of CYP1AI genotype on CYP1A2 activity. Since the Pan-Fried Meat feeding study results and the smoking cessation study results showed similar influences of CYP1A1, the two groups were analyzed together. Mean (~), median (.~), 10th, 25th, 75th, 90th percentile CYP1A2 activity is plotted. Exposed represents meat cooked at high temperature or cigarette smoke exposure, while unexposed represents meat cooked at low temperature or smoking cessation.

140 S. MacLeod et al. / Mutation Research 376 (1997) 135-142 However, when CYPIA2 activity was influenced by either cigarette smoke or high-temperature cooked meat, the presence of the heterozygous Ile/Val form of the CYP1A1 gene resulted in increased levels of CYPIA2 activity of 18.1 compared to 13.5 for indi- viduals possessing the lle/Ile form of the gene. The glutathione S-transferase I~ (GSTMI) gene is deleted in a homozygous manner in approximately 50% of the human population (GSTMI* 0). A ge- netic polymorphism in individuals expressing this gene consists of a single base change in codon 173 which produces two variant alleles designated GSTMI*A and GSTMI* B. Because there seems to be little difference between the activity of GSTM1 *A line periods of the studies (low-temperature cooked meat and non-smoking), mean CYPIA2 activity was 8.9 in individuals expressing GSTMI*A,B, while those expressing the null allele showed an increase in CYP1A2 activity to 10.5. When individuals were exposed to cigarette smoke or high-temperature cooked me~t, mean CYPIA2 activity for the group expressing GSTMI*A,B was 12.6. The lack of ex- pression of functional GSTMI resulted in an in- crease in CYP1A2 activity to 16.2. Gene-gene inter- actions were further evaluated by determining the CYPIA2 values of individuals with combinations of GSTM1 and CYPIA1 genotypes. Those individual.~ genotyped GSTMI*A,B and CYP1AI, Ile/Ile (N = and GSTMI*B alleles and between the levels of 35) had the lowest mean value at 11.8 while those GSTM 1 activity expressed whether one or two copies genotyped GSTMI" 0 and CYPIA1, Ile/Val ( N = of the gene are present (Bell et al., 1993), we have had the. highest value at 15.7. The genotype combined those individuals having one or two ,c,opies GSTM1 0 and CYP1A1, Ile/Ile (N = 40) had a of the gene into on~ group, designated GSTM1 A,B, value of. 14.8. Only 2 participants had the genotvpe for purposes of analysis. To determine whether the GSTM1 A,B and CYP1AI,Ile/Val. presence or absence of the GSTM1 gene product influenced CYP1A2 activity, individuals from the feeding study and the smoking cessation study were 5. Discuss|on genotyped for the GSTMI gene using a PCR-based assay. Fig. 5 show~ that CYP1A2 activity was higher In this study, we sought to identify factors that when the GSTMI gene was absent. During the base might influence both constitutive and induced levels of CYP1A2 expression. Wide variations in human CYP1A2 activity have been reported; however." the '~~ variant alleles identified in the structt~ral gene do not ~ ~l un~xpo~a [ F~pn,,a I explain the activity range reported to date (Nakajima ~ t (com~) ~ (com~,~a~ro.~)-[- [ et at., 1994). By studying a group of volunteers at ~ "] I [ I two levels of exposure (high- or low-temperature ~'*t [ --V ~"* [ [ cooked meat and cigarette smoking or non-smoking). ~ t ~- [ [ [---] [ it was possible to evaluate the interactions among ~" "] -I- ~o~, I I ~ I--I I CYP1A1, GSTMI and CYP1A2. ~, t lean ~ [ ~ I [ [ CYP1A1 has been shown to be genetically po.'- , '*] ~_~ ~ [ ~ t_~ [ morphic with an adenine to guanine mutation in ~ 't ~ ~-~ ] __[_ --~ ] exon 7 that results in an amino acid change from t -- I I isoleucine to valine in the protein. This amino acid °1 a~r~, '~ ] ~.~.~ ' ~r~,.~ ] substitution results in 2-4 times greater inducibility ................... of the enzyme than the more common Ile/Ile allele t'lg. 3. Ire inuuence or tJblNll genotype on UIt'taZ activity, z Since the Pan-Fried Meat feeding study results and the smoking t~osma et al., 1993; Idrotts et al., 1~94; vaury et al.. cessation study results showed similar influences of GSTMI, the 1995). CYP1A1 and CYP1A2 metabolize different two groups were analyzed together. Mean (~), median compounds in different organs: CYPIAI in lung. (~), 10th, 25th, 75th, 90th percentile CYP1A2 activity nlne~ntn nncl lvrnnhcw',uto~" nnrt ¢~ngDl~9 in tho llt, or- is plotted. Exposed represents consumption of meat ~ooked at ~.-,;-~-~']'-,~:'- ""'"~"'~.'~'~' ~".'~ ~----:" ": ...... ~ ""~"" . , . " t.iI"l/-kl IS responslote for tne actlvauon ol potv- h~gh temperature or mgarette smoke exposure, while unexposed ." represents consumption of meat cooked at low temperature or cyclic aromatic hydrocarbons while CYPIA2 is m- smoking cessation, volved in the activation of carcinogenic heterocyclic

S. MacLeod et al. / Mutation Research 376 (1997) 135-142 and aromatic amines. However, we reasoned that the action of CYPIA1 could potentially increase or de- crease circulating levels of compounds which influ- ence CYPIA2 activity. To test this hypothesis, we 141 highest CYPIA2 activity in univariate analysis (CYP1A1 Ile/Val and GSTMI'0) produced an in- termediate value when the two occurred together. This supports the hypothesis ~hat interaction between determined the CYP1AI genotype of the individuals the genes for activation and detoxification are com- in the pan-fried meat and the smoking cessation plex. studies and asked whether or not the polymorphic These results suggest that evaluating individual forms of CYP1A1 gene affected CYP1A2 pheno- cancer risk must take into account not only carcino- type. We found a correlation between the presence of gen exposure but also the activities of a complex the CYP1AI Ile/Val genotype and increased system of carcinogen metabolizing enzymes whose CYPIA2 activity levels under conditions of exposure catalytic products may influence the expression or to meat cooked at high temperature or cigarette induction of other enzymes which can either detoxify Since the CYP1A1 Ile/Val genotype results or activate potentially carcinogenic compounds. We smoke. in increased CYP1A1 activity, these results suggest have also shown that the levels of enzymes involved that a product of CYP1AI metabolism may act to in carcinogen metabolism can be manipulated in increase CYP1A2 activity. Alternatively, the human populations by dietary modification or by CYP1A1 and CYP1A2 genes may have been subject smoking cessation and that these changes in enzyme to evolutionary determinants that provided an advan- activity could potentially change the risk of environ- tage for gene-gene co-inducibility, mentally induced cancer. The quantity of carcinogenic metabolites formed in vivo are determined by levels of phase 1 enzymes responsible for the activation of procarcinogens as Acknowledgements well as the levels of phase 2 enzymes involved in the conjugation and detoxification. For this reason, we This work was supported in part by NCI grant examined whether or not the presence or absence of RO1 CA55751-03 and EPA grant R825280-01-0. expression of the GSTM1 gene affected CYP1A2 levels. Individuals in the feeding study and in the smoking cessation study were genotyped at the GSTM1 locus, and the results were used to evaluate References the influence of this genotype on the levels of Alvarez, A.P., A. Kappas, J.l_, Eisenman, K.E. Anderson, C.B. CYP1A2determined in caffeine phenotyping assays. Pantuck, E.J. Pantuck, K.C. Hsiao, W.A. Garland and A.H. Our study showed that exposure to meat cooked at high temperature and to cigarette smoke increased the activity of CYP1A2. However, expression of the null GSTM1 genotype also resulted in increased CYP1A2 levels at base line and with either exposure. Similar results have been reported for a GSTM1 effect on smoking induction of CYP1A2 activity by Bartsch et al. (1995). This data suggests that individ- uals who lack GSTM1 retain activity may higher levels of a compound which affects the activity of CYP1A2. This compound could be an inducer which is conjugated to GSH in the GSTMI*A,B individu- als, and is thereby not available to regulate CYPIA2 activity. Finally, the influence of combinations of CYP1A1 and GSTM1 were examined to determine whether the two pathways would have a synergistic effect on CYP1A2 activity. The genotypes giving the Conney (1979) Interindividual variation in drug metabolism, Clin. Pharmacol. Ther., 26, 407-419. Bartsch, H., M. Rojas, K. Alexander, A.-M. Camus, M. Casteg- naro, C. Malaveille, S. Antilla, A. Hirvonen, K. Husgafvel- Pursiainen, E. Hietanen and H. Vainio (1995) Metabolic poly- morphism affecting DNA binding and excretion of carcino- gens in humans, Pharmacogenetics, 5, $84-$90. Bell, D.A., J.A. Taylor, D.F. Paulson, C.N. Robertson, J.L. Mohler and G.W. Lucier (1993) Genetic risk and carcinogen exposure: a common inherited defect, of the carcinogen-metabolism gene glutathione S-transferase M1 (GST M1) that increases suscep- tibility to bladder cancer, J. Natl. Cancer Inst., 85, 1159-1164. Bell, D.A., A.F. Badawi, N.P. Lang, K.F. Ilett, F.F. Kadlubar and A. Hirvonen (1995) Polymorphism in the NAT1 polyadenyla- tion signal: association of NAT1* 10 allele with higher N- acetylation activity in bladder and colon tissue samples, Can- cer Res.,._55, 5226-5229. Blin, N. and D.W. Stafford (1976) Isolation of high-molecular weight DNA, Nucleic Acids Res., 3, 2303. Blum, M., D.M. Grant, A. Demierre and U.A. Meyer (1989)

142 S. MacLeod et al. / Mutation Research 376 (1997) 135-142 N-Acetylation pharmacogenetics: a gene deletion causes ab- sence of arylamine N-acetyltransferase in liver of slow acety- lator rabbits, Proc. Natl. Acad. Sci., USA, 86, 9554-9557. Butler, M.A., M. lwasaki, F.P. Guengerich and F.F. Kadlubar (1989) Human cytochrome P-450~,A (P-450IA2), the phenacetin O-deethylase, is primarily responsible for the hep- atic 3-demethylation of caffeine and N-oxidation of carcino- genic arylamines, Proc. Natl. Acad. Sci., USA, 86, 7696-7700. Butler, M., N. Lang, J. Young, N. Caporaso, P. Vineis, R. Hayes, C. Teitel, J. Massengill, M. Lawsen and F. Kadlubar (1992) Determination of CYP1A2 and acetylator phenotypes in sev- eral human populations by analysis of caffeine urinary metabolites, Pharmacogenetics, 2, 116-127. Cosma, G., F. Crofts, E. Taioli, P. Toniolo and S. Garte (1993) Relationship between genotype and function of the human CYP1AI gene, J. Toxicol. Environ. Health, 40, 309-319. Crofts, F., E. Taioli, J. Trachman, G.N. Cosma, D. Currie, P. Toniolo and S.J. Garte (I994) Functional significance of dif- ferent human CYP1AI genotypes, Carcinogenesis, 15, 2961- 2963. Devonshire, H.W., I. Kong, M. Cooper, T.P. Sloan, J.R. Idle and R.L. Smith (1983) The contribution of genetically determined oxidation status to intedndividual variation in phenacetin dis- position, Br. J. Clin. Pharmacol., 16, 157-166. Flammang, T.J., Y. Yamazoe, F.P. Guengerich and F.F. Kadlubar (1987) The S-acetyl coenzyme A-dependent metabolic activa- tion of the carcinogen N-hydroxy-2-aminofluorene by human liver cytosol and its relationship to the aromatic amine N- acetyltransferase phenotype, Carcinogenesis, 8, 1967-1970. Gonzalez, F.J. (1989) The molecular biology of cytochrome P450's, Pharmacol. Rev., 40, 244-288. Gonzalez, F.J. and H.V. Gelboin (1993) Role of human cy- tochrome P-450s in risk assessment and susceptibility to envi- ronmentally based disease, J. Toxicol. Environ. Health, 40, 289-308. Grant, D.M., B.K. Tang and W. Kalow (1984) A simple test for acetylator phenotype using caffeine, Br. J. Clin. Pharmacol., 17, 459-64. Hayashi, S.-I., J. Watanabe, K. Nakachi and K. Kawajiri (1991a) Genetic linkage of lung cancer-associated Msp I polymor- phism with amino acid replacement in the heine binding region of the human cytochrome P4501A1 gene, J. Biochem., 110, 407-411. Hayashi, S.-I., J. Watanabe, K. Nakachi and K. Kawajid (1991b) PCR detection of an A/G polymorphism within exon 7 of the CYP1A1 gene, Nucleic Acids Res., 18, 7194. Kadlubar, F.F. (1994) Biochemical individuality and its implica- tions for drug and carcinogen metabolism: recent insights from acetyltransferase and cytochrome P4501A2 phenotyping and genotyping in humans, Drug Metab. Rev., 26, 37-46. Lang, N.P., M.A. Butler, J. Massengill, M. Lawson, R.C. Stotts. M. Hauer-Jensen and F.F. Kadlubar (1994) Rapid metabolic phenotypes for acetyltransferase and cytochrome P4501A2 and putative exposure to food-borne heterocyclic amines in- crease the risk for colorectal cancer or polyps, Cancer Epi- demiol., Biomarkers Prey., 3, 675-682. Nakachi¢ K., K. lmai, S. Hayashi, J. Watanabe and K. Kawajiri (1991) Genetic susceptibility to squamous cell carcinoma of the lung in relation to cigarette smoking dose, Cancer Res., 5 l. 5177-5180. Nakajima, M., T. Yokoi, M. Mizutani, S. Shin, F.F. Kadlubar and T. Kamataki (1994) Phenotyping of CYPIA2 in Japane.~e populations by analysis of caffeine urinary metabolites: ab- sence of mutation prescribing the phenotype in the CYPIA2 gene, Cancer Epidemiol., Biomarkers Prey., 3, 413-421. Seidegard, J., W.R. Vosachek, R.W. Pero and W.R. Pearson (1988) Hereditary differences in the expression of the human glutathione transferase active on trans-stilbene oxide are due to a gene deletion, Proc. Natl. Acad., Sci. USA, 85, 7293- 7297. Shahidi, N.T. (1968) Acetophenetidin-induced methemoglobine- mia, Ann. N.Y. Acad. Sci., 151,822-832. Sinha, R., N. Rothman, E.D. Brown, S.D. Mark. R.N. Hoover. N.E. Caporaso, O.A. Lavender, M.G. Knize, N.P. Lang and F.F. Kadlubar (1994) Pan-fried meat containing high levels of heterocyclic aromatic amines but low levels of polycyclic aromatic hydrocarbons induces cytochrome P450IA2 activity, in humans, Cancer Res., 54, 6154-6159. Turesky, R.J., N.P. Lang, M.A. Butler, C.H. Teitel and F.F. Kadlubar (1991) Metabolic activation of carcinogenic hetero- cyclic aromatic amines by human liver and colon, Carcinogen- esis, 12, 1839-1845. Vaury, C., R. Laine', P. Noguiez, P. De Copper, C. Jaulin. F. Praz, D. Pompon and M. Amor-Gu6ret (1995) Human glu- tathione S-transferase M1 null genotype is associated with a high inducibility of cytochrome P450 1A1 gene transcription. Cancer Res., 55, 5520-5523. Yu, M.C., R.K. Ross, K.K. Chan, B.E. Henderson, P.L. Skippper. S.R. Tannenbaum and G.A. Coetzee (1995) Glutathione S- transferase M1 genotype affects aminobiphenyl-hemoglobin adduct Ievels in White, Black, and Asian smokers and non- smokers, Cancer Epidemiol., Biomarkers Prey., 4, 861-864.