- Email: [email protected]
Gene polymorphisms in detoxification enzymes as susceptibility factor for head and neck cancer?
Gene polymorphisms in detoxification enzymes as susceptibility factor for head and neck cancer? S. GRONAU,
MD,
D. KOENIG-GREGER, M. JERG, and H. RIECHELMANN,
OBJECTIVE: Impaired detoxification of carcinogens found in tobacco smoke appears to increase the risk for tobacco associated cancer. The objective of this study was to investigate concomitant polymorphisms in genes encoding for various detoxification enzymes in patients with head and neck squamous cell carcinoma (HNSCC). METHODS: In 187 patients with HNSCC and in 139 healthy control subjects, the polymorphisms of cytochrome P450 1A1 (CYP1A1), cytochrome P450 2D6 (CYP2D6), and glutathione S-transferase 1 and (GSTM1, GSTT1) were detected by polymerase chain reaction. RESULTS: No significant association were identified between CYP1A1 and CYP2D6 gene polymorphisms and HNSCC. Patients with laryngeal cancer revealed the GSTM1 null genotype more frequently than did the control subjects (P < 0.05). The coincidence of GSTM1 and GSTT1 null genotype was found twice as great in patients as in control subjects (P < 0.05). CONCLUSIONS: It is assumed that detoxification enzymes are functionally redundant and only the simultaneous deficiency of several detoxification enzymes increase the risk for HNSCC in alcohol- and tobacco-exposed individuals. (Otolaryngol Head Neck Surg 2003;128:674-80.)
C ytochrome
P450 1A1 (CYP1A1) and 2D6 (CYP2D6) are important phase I detoxification enzymes, catalyzing the oxidation of polycyclic aromatic hydrocarbons and aromatic amines to epoxides. A T-to-C transition downstream of exon 7 of the CYP1A1gene at the position 6235 results in an MSP I restriction site.1 A replacement of
From the Department of Otorhinolaryngology, University Ulm. This work was supported by a grant from ZAKF University Clinics of Ulm. Reprint requests: Silke Gronau, Department of Otorhinolaryngology, University Ulm, Prittwitzstr 45, D-89075, Ulm, Germany; e-mail, [email protected]. Copyright © 2003 by the American Academy of Otolaryngology–Head and Neck Surgery Foundation, Inc. 0194-5998/2003/$30.00 ⫹ 0 doi:10.1016/S0194-5998(03)00176-1 674
MD,
Ulm, Germany
adenine to guanine in exon 7 of the CYP1A1 gene results in an amino acid substitution (Ile to Val). Both mutated enzymes reveal enhanced enzyme activity and were overrepresented in Japanese patients with lung cancer.2 The family of CYP2D6 genes are involved in the detoxification of the tobacco-specific nitrosamine 4-(methylnitrosamino)-1-butanone.3 Three functional relevant gene-inactivating mutations have been described.4 Based on these polymorphisms, poor, intermediate, and extensive metabolizers have been differentiated in the study of drug pharmacokinetics. Glutathione S-transferase 1 and 1 are phase II detoxification enzymes conjugating reduced glutathione to electrophilic groups. Homozygous deletion (null genotype) in these genes results in missing gene product.5 The GSTM gene family include five isoenzymes (M1 to M5) with high genetic similarity.6 Deficient expression of glutathione S-transferase 1 prevails in ⯝50% of a white population.7 Deficient expression of glutathione S-transferase 1 prevails in ⯝20% of a European caucasian population.8 It seems likely that alterations of detoxification enzymes influence the risk of tobacco smoke- and alcohol-associated cancers in exposed individuals.9 Although phenotyping of detoxification enzymes is often affected by methodologic problems such as measurement of metabolites or lack of specificity, the determination of the phenotype influencing genotypes depends on knowledge of the relevant mutations. The simplicity of identifying probable genetic markers of cancer risk by polymerase chain reaction (PCR) technique released in a lot of population association studies and thus the functional consequences of a genetic polymorphism are not clear. However, current data on the association of cytochrome P450 oxidase and glutathione S-transferases with tobacco smoke-related cancer, including head and neck squamous cell carcinoma (HNSCC), are conflicting.10-12 Several factors might contribute to these different results, including the choice of control
Otolaryngology– Head and Neck Surgery Volume 128 Number 5
group. In several studies, an association between smoking and drinking habit and the genotype of several detoxification enzymes was reported.13,14 It thus seems appropriate to match patient and control groups with respect to tobacco smoke and alcohol consumption. Moreover, the functional redundancy of several detoxification enzymes may differ at different sites. This means that the lack of activity might be easily compensated, such as in the liver or peripheral blood lymphocytes, but eventually not in other tissues such as oropharyngeal or laryngeal mucosa. Therefore, we investigated the frequency of single and of combined genetic polymorphisms of relevant detoxification enzymes in patients with HNSCC and in control subjects. In addition, it was of interest to determine whether the frequency of detoxification enzyme polymorphisms was in particular associated with a certain tumor site. MATERIALS AND METHODS Blood samples were obtained from 187 white patients with squamous cell carcinoma of the upper aerodigestive tract treated at the Department of Otorhinolaryngology of the University of Ulm from October 1998 to October 2000. One hundred thirty-nine healthy control subjects, with matched age, gender, and smoking and drinking habits, were recruited through the use of a detailed questionnaire. After informed consent was obtained, each patient and control subject signed a tissue transfer contract. The study was performed in agreement with the local ethic committee. Whole blood samples were used for DNA extraction with an extraction kit (Qiagen, Hilden, Germany). After proteinase K lysis of the leukocytes, the sample solution was loaded onto spin columns. The DNA was absorbed onto a silica gel membrane and was washed out in a second step. The gene profiles were assessed by PCR followed by digestion appropriate for the specific restriction sites. Genotyping CYP1A1 A restriction fragment length polymorphism analysis (RFLP) was performed to detect polymorphisms in the CYP1A1 gene. The following primers were used in a standard PCR protocol
GRONAU et al
675
containing 60 ng DNA, 1 U Hot-Star-Polymerase (Quiagen) with buffer provided by the manufacturer, and 200 mol dNTP mix (Promega, Madison, WI). To detect the MSP I cleavage site in the CYP1A1 gene, 25 pmol of the primer CYP1A11, 5⬘-AAG AGG TGT AGC CGC TGC ACT-3⬘, and the primer CYP1A12, 5⬘-TAG GAG TCT TGT CTC ATG CCT-3⬘, were used. After a denaturation at 95°C for 15 minutes, 34 cycles were used at 94°C for 1 minute, 65°C for 1 minute, 72°C for 1 minute, and a final step at 72°C for 5 minutes. The amplified fragments were digested with 2 U MSP I (Boehringer Mannheim, Mannheim, Germany) overnight at 37°C (Fig 1). To detect the polymorphism in exon 7 we used 3 primers. primer 1A11, 5⬘-CCT CAA TGC AGG CTA GAA TAG AAG G-3⬘, primer 1A12, 5⬘GAA GTG TAT CGG TGA GAC CA-3⬘, and primer 1A13, 5⬘-GAA GTG TAT CGG TGA GAC CG-3⬘. The least both primers are only different in the last underlined nucleotide to detect the replacement. After a initial denaturation at 95°C for 15 minutes, 21 cycles at 94°C for 1 minute, 65°C for 1 minute, and 72°C for 1 minute followed. A final step at 72°C for 10 minutes terminated the PCR. The genotypes were visualized by gel electrophoresis stained with ethidium bromide in a 2.5% agarose gel. Genotyping of CYP2D6 The three gene-inactivating mutations in the CYP2D6 gene contain a G-to-A transition at the junction of intron 3 and exon 4, a base pair deletion in exon 5, and a total gene deletion. In the normal CYP2D6 gene, there is an ScrFI restriction site (CCNGG) that is lost in individuals with the G-to-A transition. Therefore, this mutation cannot be directly applied. To detect the G-to-A transition in the CYP2D6 gene, we used the following designed mismatched oligonucleotides for PCR to introduce a DraIII restriction site 4: primer A, 5⬘-TGC CGC CTT CGC CAA CCA CT-3⬘, and primer B, 5⬘-GGC TGG GTC CGA GGT CAC CC-3⬘ (the mismatched base pairs are underlined). As an internal positive control, the primer C, 5⬘CGG CCC AGC CAC TCT CGT GT-3⬘, and primer D, 5⬘-AAC AGG GTC CCA GCT GAG GAG-3⬘, were included.4 Using 50 L probe volume, the PCR contained 50 TO 100 ng DNA, 0.25
676 GRONAU et al
Otolaryngology– Head and Neck Surgery May 2003
Fig 1. CYP1A1 MSP I RFLP. Lanes 1 and 6, standard. Lane 2, wt/wt. Lane 3, wt/wt. Lane 4, m/m.
M of each primer (A, B, C, and D), 200 mol dNTP mix (Promega), and 5 U Taq polymerase (Boehringer Mannheim). After an initial denaturation at 94°C for 3 minutes, 30 cycles of PCR were carried out at 94°C for 1 minute, 60°C for 1 minute, and 72°C for 1 minute. A final polymerization step at 72°C for 4 minutes was added. The PCR products were digested with 3 U DraIII (Boehringer Mannheim) and 2 U ScrFI with buffer at 37°C overnight. The PCR products were stained with ethidium bromide following gel electrophoresis in a 2.5% agarose gel. Genotyping of GSTM1 and GSTT1 To detect GSTM1 and GSTT1 polymorphisms, a multiplex PCR was performed. Using Hot-StarPolymerase (Quiagen), standard PCR conditions were chosen as recommended by the manufacturer. The primers GSTM1A, 5⬘-GAA CTC CCT GAA AAG CTA AAG C-3⬘, and GSTM1B, 5⬘GTT GGG CTC AAA TAT ACG GTG G-3⬘, were used for amplification of a 273-bp fragment. To generate a 480-bp fragment, the primers GSTT1, 5⬘-GAA CTC CCT GAA AAG CTA AAG C-3⬘, and GSTT2, 5⬘-GTT GGG CTC AAA TAT ACG GTG G-3⬘, were used.15 As an internal control, we
coamplified a -interferon gene with following primers, IFN3, 5⬘-GGC ACA ACA GGT AGT AGG CG-3⬘, and IFN5, 5⬘-GCC ACA GGA GCT TCT GAC ACC-3⬘.16 The PCR products were visualized by gel electrophoresis stained with ethidium bromide. In this PCR assay, the absence of a 273-bp or a 480-bp fragment indicates the GSTM1 or GSTT1 null genotype. The 180-bp fragment of -interferon was used as a positive control. Statistical Analysis Statistical analysis was performed with WinStat Version 3.1 (Kalmi Inc, Cambridge, MA). The CYP1A1 and CYP2D6 genotypes were grouped into categories of poor, medium, and high metabolizers. The GSTM1 and GSTT1 genotypes were categorized at wild-type and null genotype. To evaluate a possible association of genotype and smoking behavior, patients and control subjects were divided in 3 groups (nonsmoker, ⬍10 packyears, and ⬎10 pack-years). For alcohol consumption, patients and control subjects were grouped into those who did not drink alcohol daily, those who drink ⬍50 g alcohol daily, and those who drink ⬎50 g alcohol daily. To test the association
Otolaryngology– Head and Neck Surgery Volume 128 Number 5
GRONAU et al
677
Table 1. Demographic variables and risk factors of head and neck squamous cell carcinoma in patients with tumors and control subjects Variable
Patients with tumors
Control subjects
Mean age (y) Male (%) No smoking (%) ⬍10 pack/y (%) ⬎10 pack/y (%) No alcohol (%) ⬍50 g alcohol/day (%) ⬎50 g alcohol/day (%)
58.1 ⫾ 11.6 86.6 24.0 19.8 56.2 35.3 29.4 35.3
55.9 ⫾ 13.1 86.3 34.4 15.6 50.0 37.0 35.2 27.8
between the different genotypes and the patients or control subjects, we used the Fisher‘s exact P test and the 2 test. Cases with missing data were excluded from the statistical analysis. RESULTS One hundred eighty-seven white patients (162 men and 25 women) with squamous cell carcinoma of the upper aerodigestive tract and 139 healthy white control subjects (120 men and 19 women) were consecutively enclosed. The mean age was 58.1 ⫾ 11.6 years in the patient group and 55.9 ⫾ 13.1 years in the control group. Of the patients with cancer, 56.2% smoked ⬎10 packyears, 19.8% smoked ⬍10 pack-years, and 24.0% did not smoke. In the control group, 50.0% smoked ⬎10 pack-years, 15.6% smoked ⬍10 pack-years, and 34.4% did not smoke. Of the patients with cancer, 35.3% drank ⬎50 g alcohol daily, 29.4% drank ⬍50 g alcohol daily, and 35.3% said they did not drink alcohol daily. Of the healthy control subjects, 27.8% drank ⬎50 g alcohol daily, 35.2% drank ⬍50 g alcohol daily, and 27.8% said they did not drink alcohol daily (Table 1). Most patients had cancer of the mouth and oropharynx. Specifically, 53 (28.3%) of 187 patients had laryngeal cancer, 32 (17.1%) had hypopharyngeal cancer, and 85 (45.5%) had oropharyngeal cancer or cancer of the oral cavity, and in 17 (9.1%) cases, the primary site was unknown or other localization (Table 2). As outlined in Table 3, no patient showed the homozygous mutated genotype (m/m) for the CYP1A1 gene in the MSP I restriction site or in exon 7. In the control subjects, the homozygous
Table 2. Site of cancer Tumor site
No. of patients (%)
Larynx Hypopharynx Oropharynx, oral cavity Unknown primary other site
53 (28.3) 32 (17.1) 85 (45.5) 17 (9.1)
genotype was found twice. One mutated chromosome (wt/m) was seen in 45 (24.1%) of 187 patients and in 24 (17.9%) of 139 control subjects. The normal gene (wt/wt) was found in 142 (75.9%) of 187 patients and in 113 (81.3%) of 139 control subjects (P ⬎ 0.05). Moreover, in comparing only individuals smoking ⬎20 pack-years in the patient and control groups, no remarkable differences in the CYP1A1 genotypes were found. The investigated polymorphisms in the CYP2D6 gene are summarized in Table 4. One hundred fourteen (61.0%) of 187 patients and 79 (56.8%) of 139 control subjects were extensive metabolizers. The intermediate type was found in 60 (32.1%) of 187 patients and in 53 (38.1%) of 139 control subjects. Of the patients, 13 (6.9%) were poor metabolizers, whereas 7 (5.1%) of the control subjects were poor metabolizers (P ⬍ 0.05). The GSTM1 null genotype was found in 42.8% (80 of 187) of patients and in 48.9% (68 of 139) of control subjects (P ⬎ 0.05), and 27 (15.8%) of 171 patients and 19 (14.4%) of 131 control subjects revealed the GSTT1 null genotype (P ⬎ 0.05). Coincident null genotypes in the GSTM1 and GSTT1 genes were found in 13.5% of the patients and in 6.1% of the healthy control subjects (P ⬍ 0.05) (Table 5).
Otolaryngology– Head and Neck Surgery May 2003
678 GRONAU et al
Table 3. Polymorphisms in CYP1A1 gene MSP 1 restriction site and in exon 7 in patients and healthy control subjects
CYP 1A1 MSP I site (n) Patients with tumors Control subjects CYP 1A1 exon 7 (n) Patients with tumors Control subjects
wt/wt
wt/m
m/m
142 (75.9%) 113 (81.3%)
45 (24.1%) 24 (17.9%)
0 (0%) 2 (1.4%)
155 (82.9%) 120 (86.4%)
32 (17.1%) 17 (12.2%)
0 (0%) 2 (1.4%)
wt/wt, homozygous wild-type; wt/m, heterozygous wild-type/mutation; m/m, homozygous mutation.
Table 4. CYP2D6 polymorphisms classified as extensive, intermediate, and poor metabolizer CYP 2D6
Extensive metabolizer
Intermediate metabolizer
Poor metabolizer
Patients with tumors Control subjects
114 (61.0%) 79 (56.8%)
60 (32.1%) 53 (38.1%)
13 (6.9%) 7 (5.1%)
To determine the distribution of the GSTT1 and GSTM1 genotypes in patients and control subjects who were heavy smokers and drinkers (⬎20 packyears and ⬎50 g alcohol per day), we further subdivided the groups. We analyzed 49 patients and 28 healthy control subjects who were heavy smokers and drinkers for the contribution of both of these genotypes. The GSTT1 or GSTM1 null genotype was not found more frequently in the patient group than in the control group (P ⬎ 0.5). Coincident null genotypes in the GSTM1 and GSTT1 genes were significantly more frequent in patients than in healthy control subjects (P ⬍ 0.01). The genotype grouped by the site of cancer is outlined in Table 6. The GSTM1 null genotype was particularly frequent in patients with laryngeal cancer (P ⬍ 0.05). With this exception, site of cancer was not correlated with the genotype of GSTM1 or GSTT1 (Table 6). DISCUSSION Polymorphisms in genes encoding for detoxification enzymes may influence enzyme activity.17 It is assumed that individual differences in the detoxification of chemical carcinogens are important factors in the susceptibility to various types of human cancer. The cytochrome P450 family consists of enzymes of the first phase detoxification. They catalyze the formation of active metabolites, allowing electrophilic attack for further metabo-
lization. Some xenobiotics exert their carcinogenic activity only following activation by cytochrome P450 enzymes. Benzo(a)pyrene is a paradigmatic carcinogen found in tobacco smoke. After activation by cytochrome P450, an epoxide arises, which forms DNA adducts, possibly initiating carcinogenesis. Enhanced enzyme activity may result in increased epoxide formation and consequently increased DNA damage. The investigated polymorphisms in the CYP gene result in enhanced enzyme activity.2 The CYP1A1 (MSP I and/or exon 7) homozygous mutations are found in ⯝7% to 10% of the white population and in up to 33% of a Japanese population. In some studies, an increased susceptibility to squamous cell carcinoma of the lung or the upper aerodigestive tract was associated with these homozygous (m/m) polymorphisms.11,18 In other investigations, no such association was found.12,19 In this investigation, the homozygous mutation, as well in the MSP I restriction site as in exon 7, was found only once in the control group. No patient revealed this genotype. The reason for the infrequent homozygous polymorphism in our population is not clear. Methodologic problems are unlikely, because we performed every PCR twice and accepted only consistent results. Most likely, the low frequency of CYP1A1 polymorphism is a regional variation.
Otolaryngology– Head and Neck Surgery Volume 128 Number 5
GRONAU et al
679
Table 5. Polymorphism in GSTM1 and GSTT genes in patients with tumors and control subjects
GSTM1 positive GSTM1 negative GSTT1 positive GSTT1 negative GSTM1 and/or GSTT1 positive GSTM1 and GSTT1 negative
Patients with tumors
Control subjects
107 (57.2%) 80 (42.8%) 157 (83.9%) 30 (16.1%) 160 (85.6%) 27 (14.4%)
71 (51.1%) 68 (48.9%) 118 (84.9%) 21 (15.1%) 130 (93.5%) 9 (6.5%)
Table 6. GSTM1 polymorphism and site of cancer
Tumor site
Larynx
Hypopharynx
Oropharynx, oral cavity
GSTM1 positive GSTM1 negative
14 (26.4%) 39 (73.6%)
18 (56.3%) 14 (43.7%)
38 (44.7%) 47 (55.3)
In the CYP2D6 gene, we did not find relevant differences between patients and control subjects. The allelic frequencies of extensive, intermediate, and poor metabolizer genotypes in our patient or control population are consistent with previous reports.3,4 The supergene family of glutathione S-transferases includes enzymes of phase II detoxification, which can be grouped into 4 classes: ␣, , , and . Except for , each class contains several genes. They catalyze the conjugation of reduced glutathione with a large number of electrophils. This results in increased water solubility, allowing renal excretion of various carcinogens or epoxides formed during phase I detoxification. Five isoenzymes of GSTM have been identified.6 The glutathione S-transferase gene consist of 2 classes: the GSTT1 and GSTT2 genes. Homozygous deletions in the GSTM1 or GSTT1 gene result in missing enzyme activity due to the lack of gene product.5,8 Association studies of GSTM1 polymorphisms and lung cancer or cancer of the upper aerodigestive tract yielded contradictory results. This did not change if the lifetime smoking history was taken into account.10,12,15,19,20 In this investigation, no association was found between the frequency of GSTM1 null genotype alone or the GSTT1 null genotype alone and head and neck cancer. In analyzing only heavy smokers and drinkers, we also found no differences in the distribution of the genotypes. However, the coin-
Unknown primary, other site
P value
9 (52.9%) 8 (47.1%)
⬎0.05 ⬍0.05
cidence of GSTM1 and GSTT1 null genotype was more frequent in patients than in control subjects and especially in patients who were heavy smokers and drinkers (P ⬍ 0.01). In patients and control subjects who were nonsmokers or who had a smoke history of ⬍20 pack-years and did not drink alcohol daily, no association of the gene polymorphisms alone or in coincidence was found (data not shown). The simultaneous absence of the 2 enzymes GSTT1 and GSTM1 could result in the accumulation of electrophil intermediates because of reduced phase II detoxification capacity. These carcinogens lead to the formation of DNA-adducts and finally to DNA damage. Accumulation of carcinogenic intermediate metabolites in tobacco smoke-exposed tissues might result in tumor growth. Concomitant alcohol and tobacco abuse is a multiplicative risk factor for squamous cell carcinomas of the upper aerodigestive tract. The investigated polymorphisms are phase I and phase II enzymes that are in particular responsible for the detoxification of polycyclic aromatic hydrocarbons and aromatic amines found in tobacco smoke. Interestingly, an association of the GSTM1 null genotype with laryngeal carcinoma was observed. The larynx is that part of the upper aerodigestive tract that is most exposed to tobacco smoke and not directly to alcohol. On the one hand, it is possible that in laryngeal mucosa the
Otolaryngology– Head and Neck Surgery May 2003
680 GRONAU et al
deficient GSTM1 enzyme cannot be compensated for by other enzymes. On the other hand, it could be that in the larynx, the effect of the GSTM1 null polymorphism is not biased by the interference of alcohol. Because of a possible association between smoking and drinking habits and the genotype of several detoxification enzymes,13,14 we matched the patient and the control groups. The 2 groups were similar in smoking and drinking habits. Also, the gender and the age of patients and control subjects were similar. In conclusion, the lack of a single detoxification enzyme was not associated with head and neck cancer in this investigation. The simultaneous null genotype of GSTM1 and GSTT1 was more common in cancer patients than in control subjects. Moreover, the GSTM1 null genotype was overrepresented in patients with laryngeal cancer. Additional investigations on the tissue-specific concentrations of detoxification enzymes are necessary to evaluate possible sequelae of a single deficient enzyme in certain tissues such as laryngeal or oropharyngeal mucosa. Therefore, it is also necessary to determine the enzyme concentration and enzyme activity in tissue to acquire information about the phenotype of genes. REFERENCES
1. Bale A, Nebert D, McBride O. Subchromosomal localisation of the dioxin-inducible P-450 locus (CYP1) and description of two RFLPs detected with a 3⬘ P-450 cDNA probe. Cytogenet Cell Genet 1987;46:574-5. 2. Nakachi K, Imai K, Hayashi S, et al. Genetic susceptibility to squamous cell carcinoma of the lung in relation to cigarette smoking dose. Cancer Res 1991;51:5177-80. 3. Rannung A, Alexandrie AK, Persson I, et al. Genetic polymorphism of cytochromes 1A1, 2D6 and 2E1: regulation and toxicological significance. J Occup Environ Med 1995;37:25-36. 4. Anwar WA, Abdel-Rahman SZ, El-Zein RA, et al. Genetic polymorphism of GSTM1, CYP2E1 and CYP2D6 in Egyptian bladder cancer patients. Carcinogenesis 1996;17:1923-9. 5. Roots I, Drakoulis N, Brockmo¨ ller J. Polymorphic enzymes and cancer risk: concepts, methodology and data review. In: Kalow W, editor. Pharmacogenetics of Drug
6.
7. 8.
9.
10. 11. 12.
13.
14. 15. 16.
17. 18.
19.
20.
Metabolism. New York: Pergamon Presse; 1992. p. 81541. Hayes JD, Pulford DJ. The glutathione S-transferase supergene family: regulation of GST and the contribution of the isoenzymes to cancer chemoprotection and drug resistance. Crit Rev Biochem Mol Biol 1995;30:445-600. Board P. Biochemical genetics of glutathione S-transferase in man. Am J Hum Genet 1981;33:36-43. Pemble S, Schroeder KR, Spencer SR, et al. Human glutathione S-transferase theta (GSTT1): cDNA cloning and the characterisation of a genetic polymorphism. Biochem J 1994;300:267-71. Tang DL, Rundle A, Warburton D, et al. Association between both genetic and environmental biomarkers and lung cancer: evidence of a great risk of lung cancer in women smoker. Carcinogenesis 1998;19:1949-53. Ford JG, Li Y, O’Sullivan MM, et al. Glutathione Stransferase M1 polymorphism and lung cancer risk in African-Americans. Carcinogenesis 2000;21:1971-5. Sato M, Sato T, Izumo T, et al. Genetic polymorphism of drug-metabolizing enzymes and susceptibility to oral cancer. Carcinogenesis 1999;20:1927-31. Oude Ophius MB, van Lieshout EM, Roelofs HM, et al. Glutathione S-transferase M1 and T1 and cytochrome P4501A1 polymorphism in relation to the risk for benign and malignant head and neck lesions. Cancer 1999;82: 936-43. Saarikoski ST, Sata F, Husgafvel-Pursiainen K, et al. CYP2D6 ultrarapid metabolizer genotype as a potential modifier of smoking behaviour. Pharmacogenetics 2000; 10:5-10. Boustead C, Taber H, Idle JR, et al. CYP2D6 genotype and smoking behaviour in cigarette smokers. Pharmacogenetics 1997;7:411-4. Cheng L, Sturgis EM, Eicher SA, et al. Glutathione S-transferase polymorphisms and risk of squamous cell carcinoma of head and neck. Int J Cancer 1999;84:220-4. Madhusudan G, Prasann K, Kamala K. Human leucocyte glutathion S-transferase isoenzyme (class mu) and susceptibility to smoking-related cancers. J Toxicol Environ Health 1995;46:1-8. Smith G, Stanley LA, Sim E, et al. Metabolic polymorphisms and cancer susceptibility. Cancer Surv 1995;25: 27-75. Drakoulis N, Cascorbi I, Brockmo¨ ller J, et al. Polymorphisms in human lung cancer: exon 7 mutation (4889 A to G) and a T to C mutation in the 3⬘-flanking region. Clin Invest 1994;72:240-8. Bennett W, Alavanja MCR, Blomeke B, et al. Environmental tobacco smoke, genetic susceptibility and risk of lung cancer in never smoking women. J Natl Cancer Inst 1999;91:2009-14. Cascorbi I, Brockmo¨ ller J, Roots I. A C4887A polymorphism in exon 7 of human CYP1A1: population frequency, mutation linkages and impact on lung cancer susceptibility. Cancer Res 1996;56:4965-9.
MD,
D. KOENIG-GREGER, M. JERG, and H. RIECHELMANN,
OBJECTIVE: Impaired detoxification of carcinogens found in tobacco smoke appears to increase the risk for tobacco associated cancer. The objective of this study was to investigate concomitant polymorphisms in genes encoding for various detoxification enzymes in patients with head and neck squamous cell carcinoma (HNSCC). METHODS: In 187 patients with HNSCC and in 139 healthy control subjects, the polymorphisms of cytochrome P450 1A1 (CYP1A1), cytochrome P450 2D6 (CYP2D6), and glutathione S-transferase 1 and (GSTM1, GSTT1) were detected by polymerase chain reaction. RESULTS: No significant association were identified between CYP1A1 and CYP2D6 gene polymorphisms and HNSCC. Patients with laryngeal cancer revealed the GSTM1 null genotype more frequently than did the control subjects (P < 0.05). The coincidence of GSTM1 and GSTT1 null genotype was found twice as great in patients as in control subjects (P < 0.05). CONCLUSIONS: It is assumed that detoxification enzymes are functionally redundant and only the simultaneous deficiency of several detoxification enzymes increase the risk for HNSCC in alcohol- and tobacco-exposed individuals. (Otolaryngol Head Neck Surg 2003;128:674-80.)
C ytochrome
P450 1A1 (CYP1A1) and 2D6 (CYP2D6) are important phase I detoxification enzymes, catalyzing the oxidation of polycyclic aromatic hydrocarbons and aromatic amines to epoxides. A T-to-C transition downstream of exon 7 of the CYP1A1gene at the position 6235 results in an MSP I restriction site.1 A replacement of
From the Department of Otorhinolaryngology, University Ulm. This work was supported by a grant from ZAKF University Clinics of Ulm. Reprint requests: Silke Gronau, Department of Otorhinolaryngology, University Ulm, Prittwitzstr 45, D-89075, Ulm, Germany; e-mail, [email protected]. Copyright © 2003 by the American Academy of Otolaryngology–Head and Neck Surgery Foundation, Inc. 0194-5998/2003/$30.00 ⫹ 0 doi:10.1016/S0194-5998(03)00176-1 674
MD,
Ulm, Germany
adenine to guanine in exon 7 of the CYP1A1 gene results in an amino acid substitution (Ile to Val). Both mutated enzymes reveal enhanced enzyme activity and were overrepresented in Japanese patients with lung cancer.2 The family of CYP2D6 genes are involved in the detoxification of the tobacco-specific nitrosamine 4-(methylnitrosamino)-1-butanone.3 Three functional relevant gene-inactivating mutations have been described.4 Based on these polymorphisms, poor, intermediate, and extensive metabolizers have been differentiated in the study of drug pharmacokinetics. Glutathione S-transferase 1 and 1 are phase II detoxification enzymes conjugating reduced glutathione to electrophilic groups. Homozygous deletion (null genotype) in these genes results in missing gene product.5 The GSTM gene family include five isoenzymes (M1 to M5) with high genetic similarity.6 Deficient expression of glutathione S-transferase 1 prevails in ⯝50% of a white population.7 Deficient expression of glutathione S-transferase 1 prevails in ⯝20% of a European caucasian population.8 It seems likely that alterations of detoxification enzymes influence the risk of tobacco smoke- and alcohol-associated cancers in exposed individuals.9 Although phenotyping of detoxification enzymes is often affected by methodologic problems such as measurement of metabolites or lack of specificity, the determination of the phenotype influencing genotypes depends on knowledge of the relevant mutations. The simplicity of identifying probable genetic markers of cancer risk by polymerase chain reaction (PCR) technique released in a lot of population association studies and thus the functional consequences of a genetic polymorphism are not clear. However, current data on the association of cytochrome P450 oxidase and glutathione S-transferases with tobacco smoke-related cancer, including head and neck squamous cell carcinoma (HNSCC), are conflicting.10-12 Several factors might contribute to these different results, including the choice of control
Otolaryngology– Head and Neck Surgery Volume 128 Number 5
group. In several studies, an association between smoking and drinking habit and the genotype of several detoxification enzymes was reported.13,14 It thus seems appropriate to match patient and control groups with respect to tobacco smoke and alcohol consumption. Moreover, the functional redundancy of several detoxification enzymes may differ at different sites. This means that the lack of activity might be easily compensated, such as in the liver or peripheral blood lymphocytes, but eventually not in other tissues such as oropharyngeal or laryngeal mucosa. Therefore, we investigated the frequency of single and of combined genetic polymorphisms of relevant detoxification enzymes in patients with HNSCC and in control subjects. In addition, it was of interest to determine whether the frequency of detoxification enzyme polymorphisms was in particular associated with a certain tumor site. MATERIALS AND METHODS Blood samples were obtained from 187 white patients with squamous cell carcinoma of the upper aerodigestive tract treated at the Department of Otorhinolaryngology of the University of Ulm from October 1998 to October 2000. One hundred thirty-nine healthy control subjects, with matched age, gender, and smoking and drinking habits, were recruited through the use of a detailed questionnaire. After informed consent was obtained, each patient and control subject signed a tissue transfer contract. The study was performed in agreement with the local ethic committee. Whole blood samples were used for DNA extraction with an extraction kit (Qiagen, Hilden, Germany). After proteinase K lysis of the leukocytes, the sample solution was loaded onto spin columns. The DNA was absorbed onto a silica gel membrane and was washed out in a second step. The gene profiles were assessed by PCR followed by digestion appropriate for the specific restriction sites. Genotyping CYP1A1 A restriction fragment length polymorphism analysis (RFLP) was performed to detect polymorphisms in the CYP1A1 gene. The following primers were used in a standard PCR protocol
GRONAU et al
675
containing 60 ng DNA, 1 U Hot-Star-Polymerase (Quiagen) with buffer provided by the manufacturer, and 200 mol dNTP mix (Promega, Madison, WI). To detect the MSP I cleavage site in the CYP1A1 gene, 25 pmol of the primer CYP1A11, 5⬘-AAG AGG TGT AGC CGC TGC ACT-3⬘, and the primer CYP1A12, 5⬘-TAG GAG TCT TGT CTC ATG CCT-3⬘, were used. After a denaturation at 95°C for 15 minutes, 34 cycles were used at 94°C for 1 minute, 65°C for 1 minute, 72°C for 1 minute, and a final step at 72°C for 5 minutes. The amplified fragments were digested with 2 U MSP I (Boehringer Mannheim, Mannheim, Germany) overnight at 37°C (Fig 1). To detect the polymorphism in exon 7 we used 3 primers. primer 1A11, 5⬘-CCT CAA TGC AGG CTA GAA TAG AAG G-3⬘, primer 1A12, 5⬘GAA GTG TAT CGG TGA GAC CA-3⬘, and primer 1A13, 5⬘-GAA GTG TAT CGG TGA GAC CG-3⬘. The least both primers are only different in the last underlined nucleotide to detect the replacement. After a initial denaturation at 95°C for 15 minutes, 21 cycles at 94°C for 1 minute, 65°C for 1 minute, and 72°C for 1 minute followed. A final step at 72°C for 10 minutes terminated the PCR. The genotypes were visualized by gel electrophoresis stained with ethidium bromide in a 2.5% agarose gel. Genotyping of CYP2D6 The three gene-inactivating mutations in the CYP2D6 gene contain a G-to-A transition at the junction of intron 3 and exon 4, a base pair deletion in exon 5, and a total gene deletion. In the normal CYP2D6 gene, there is an ScrFI restriction site (CCNGG) that is lost in individuals with the G-to-A transition. Therefore, this mutation cannot be directly applied. To detect the G-to-A transition in the CYP2D6 gene, we used the following designed mismatched oligonucleotides for PCR to introduce a DraIII restriction site 4: primer A, 5⬘-TGC CGC CTT CGC CAA CCA CT-3⬘, and primer B, 5⬘-GGC TGG GTC CGA GGT CAC CC-3⬘ (the mismatched base pairs are underlined). As an internal positive control, the primer C, 5⬘CGG CCC AGC CAC TCT CGT GT-3⬘, and primer D, 5⬘-AAC AGG GTC CCA GCT GAG GAG-3⬘, were included.4 Using 50 L probe volume, the PCR contained 50 TO 100 ng DNA, 0.25
676 GRONAU et al
Otolaryngology– Head and Neck Surgery May 2003
Fig 1. CYP1A1 MSP I RFLP. Lanes 1 and 6, standard. Lane 2, wt/wt. Lane 3, wt/wt. Lane 4, m/m.
M of each primer (A, B, C, and D), 200 mol dNTP mix (Promega), and 5 U Taq polymerase (Boehringer Mannheim). After an initial denaturation at 94°C for 3 minutes, 30 cycles of PCR were carried out at 94°C for 1 minute, 60°C for 1 minute, and 72°C for 1 minute. A final polymerization step at 72°C for 4 minutes was added. The PCR products were digested with 3 U DraIII (Boehringer Mannheim) and 2 U ScrFI with buffer at 37°C overnight. The PCR products were stained with ethidium bromide following gel electrophoresis in a 2.5% agarose gel. Genotyping of GSTM1 and GSTT1 To detect GSTM1 and GSTT1 polymorphisms, a multiplex PCR was performed. Using Hot-StarPolymerase (Quiagen), standard PCR conditions were chosen as recommended by the manufacturer. The primers GSTM1A, 5⬘-GAA CTC CCT GAA AAG CTA AAG C-3⬘, and GSTM1B, 5⬘GTT GGG CTC AAA TAT ACG GTG G-3⬘, were used for amplification of a 273-bp fragment. To generate a 480-bp fragment, the primers GSTT1, 5⬘-GAA CTC CCT GAA AAG CTA AAG C-3⬘, and GSTT2, 5⬘-GTT GGG CTC AAA TAT ACG GTG G-3⬘, were used.15 As an internal control, we
coamplified a -interferon gene with following primers, IFN3, 5⬘-GGC ACA ACA GGT AGT AGG CG-3⬘, and IFN5, 5⬘-GCC ACA GGA GCT TCT GAC ACC-3⬘.16 The PCR products were visualized by gel electrophoresis stained with ethidium bromide. In this PCR assay, the absence of a 273-bp or a 480-bp fragment indicates the GSTM1 or GSTT1 null genotype. The 180-bp fragment of -interferon was used as a positive control. Statistical Analysis Statistical analysis was performed with WinStat Version 3.1 (Kalmi Inc, Cambridge, MA). The CYP1A1 and CYP2D6 genotypes were grouped into categories of poor, medium, and high metabolizers. The GSTM1 and GSTT1 genotypes were categorized at wild-type and null genotype. To evaluate a possible association of genotype and smoking behavior, patients and control subjects were divided in 3 groups (nonsmoker, ⬍10 packyears, and ⬎10 pack-years). For alcohol consumption, patients and control subjects were grouped into those who did not drink alcohol daily, those who drink ⬍50 g alcohol daily, and those who drink ⬎50 g alcohol daily. To test the association
Otolaryngology– Head and Neck Surgery Volume 128 Number 5
GRONAU et al
677
Table 1. Demographic variables and risk factors of head and neck squamous cell carcinoma in patients with tumors and control subjects Variable
Patients with tumors
Control subjects
Mean age (y) Male (%) No smoking (%) ⬍10 pack/y (%) ⬎10 pack/y (%) No alcohol (%) ⬍50 g alcohol/day (%) ⬎50 g alcohol/day (%)
58.1 ⫾ 11.6 86.6 24.0 19.8 56.2 35.3 29.4 35.3
55.9 ⫾ 13.1 86.3 34.4 15.6 50.0 37.0 35.2 27.8
between the different genotypes and the patients or control subjects, we used the Fisher‘s exact P test and the 2 test. Cases with missing data were excluded from the statistical analysis. RESULTS One hundred eighty-seven white patients (162 men and 25 women) with squamous cell carcinoma of the upper aerodigestive tract and 139 healthy white control subjects (120 men and 19 women) were consecutively enclosed. The mean age was 58.1 ⫾ 11.6 years in the patient group and 55.9 ⫾ 13.1 years in the control group. Of the patients with cancer, 56.2% smoked ⬎10 packyears, 19.8% smoked ⬍10 pack-years, and 24.0% did not smoke. In the control group, 50.0% smoked ⬎10 pack-years, 15.6% smoked ⬍10 pack-years, and 34.4% did not smoke. Of the patients with cancer, 35.3% drank ⬎50 g alcohol daily, 29.4% drank ⬍50 g alcohol daily, and 35.3% said they did not drink alcohol daily. Of the healthy control subjects, 27.8% drank ⬎50 g alcohol daily, 35.2% drank ⬍50 g alcohol daily, and 27.8% said they did not drink alcohol daily (Table 1). Most patients had cancer of the mouth and oropharynx. Specifically, 53 (28.3%) of 187 patients had laryngeal cancer, 32 (17.1%) had hypopharyngeal cancer, and 85 (45.5%) had oropharyngeal cancer or cancer of the oral cavity, and in 17 (9.1%) cases, the primary site was unknown or other localization (Table 2). As outlined in Table 3, no patient showed the homozygous mutated genotype (m/m) for the CYP1A1 gene in the MSP I restriction site or in exon 7. In the control subjects, the homozygous
Table 2. Site of cancer Tumor site
No. of patients (%)
Larynx Hypopharynx Oropharynx, oral cavity Unknown primary other site
53 (28.3) 32 (17.1) 85 (45.5) 17 (9.1)
genotype was found twice. One mutated chromosome (wt/m) was seen in 45 (24.1%) of 187 patients and in 24 (17.9%) of 139 control subjects. The normal gene (wt/wt) was found in 142 (75.9%) of 187 patients and in 113 (81.3%) of 139 control subjects (P ⬎ 0.05). Moreover, in comparing only individuals smoking ⬎20 pack-years in the patient and control groups, no remarkable differences in the CYP1A1 genotypes were found. The investigated polymorphisms in the CYP2D6 gene are summarized in Table 4. One hundred fourteen (61.0%) of 187 patients and 79 (56.8%) of 139 control subjects were extensive metabolizers. The intermediate type was found in 60 (32.1%) of 187 patients and in 53 (38.1%) of 139 control subjects. Of the patients, 13 (6.9%) were poor metabolizers, whereas 7 (5.1%) of the control subjects were poor metabolizers (P ⬍ 0.05). The GSTM1 null genotype was found in 42.8% (80 of 187) of patients and in 48.9% (68 of 139) of control subjects (P ⬎ 0.05), and 27 (15.8%) of 171 patients and 19 (14.4%) of 131 control subjects revealed the GSTT1 null genotype (P ⬎ 0.05). Coincident null genotypes in the GSTM1 and GSTT1 genes were found in 13.5% of the patients and in 6.1% of the healthy control subjects (P ⬍ 0.05) (Table 5).
Otolaryngology– Head and Neck Surgery May 2003
678 GRONAU et al
Table 3. Polymorphisms in CYP1A1 gene MSP 1 restriction site and in exon 7 in patients and healthy control subjects
CYP 1A1 MSP I site (n) Patients with tumors Control subjects CYP 1A1 exon 7 (n) Patients with tumors Control subjects
wt/wt
wt/m
m/m
142 (75.9%) 113 (81.3%)
45 (24.1%) 24 (17.9%)
0 (0%) 2 (1.4%)
155 (82.9%) 120 (86.4%)
32 (17.1%) 17 (12.2%)
0 (0%) 2 (1.4%)
wt/wt, homozygous wild-type; wt/m, heterozygous wild-type/mutation; m/m, homozygous mutation.
Table 4. CYP2D6 polymorphisms classified as extensive, intermediate, and poor metabolizer CYP 2D6
Extensive metabolizer
Intermediate metabolizer
Poor metabolizer
Patients with tumors Control subjects
114 (61.0%) 79 (56.8%)
60 (32.1%) 53 (38.1%)
13 (6.9%) 7 (5.1%)
To determine the distribution of the GSTT1 and GSTM1 genotypes in patients and control subjects who were heavy smokers and drinkers (⬎20 packyears and ⬎50 g alcohol per day), we further subdivided the groups. We analyzed 49 patients and 28 healthy control subjects who were heavy smokers and drinkers for the contribution of both of these genotypes. The GSTT1 or GSTM1 null genotype was not found more frequently in the patient group than in the control group (P ⬎ 0.5). Coincident null genotypes in the GSTM1 and GSTT1 genes were significantly more frequent in patients than in healthy control subjects (P ⬍ 0.01). The genotype grouped by the site of cancer is outlined in Table 6. The GSTM1 null genotype was particularly frequent in patients with laryngeal cancer (P ⬍ 0.05). With this exception, site of cancer was not correlated with the genotype of GSTM1 or GSTT1 (Table 6). DISCUSSION Polymorphisms in genes encoding for detoxification enzymes may influence enzyme activity.17 It is assumed that individual differences in the detoxification of chemical carcinogens are important factors in the susceptibility to various types of human cancer. The cytochrome P450 family consists of enzymes of the first phase detoxification. They catalyze the formation of active metabolites, allowing electrophilic attack for further metabo-
lization. Some xenobiotics exert their carcinogenic activity only following activation by cytochrome P450 enzymes. Benzo(a)pyrene is a paradigmatic carcinogen found in tobacco smoke. After activation by cytochrome P450, an epoxide arises, which forms DNA adducts, possibly initiating carcinogenesis. Enhanced enzyme activity may result in increased epoxide formation and consequently increased DNA damage. The investigated polymorphisms in the CYP gene result in enhanced enzyme activity.2 The CYP1A1 (MSP I and/or exon 7) homozygous mutations are found in ⯝7% to 10% of the white population and in up to 33% of a Japanese population. In some studies, an increased susceptibility to squamous cell carcinoma of the lung or the upper aerodigestive tract was associated with these homozygous (m/m) polymorphisms.11,18 In other investigations, no such association was found.12,19 In this investigation, the homozygous mutation, as well in the MSP I restriction site as in exon 7, was found only once in the control group. No patient revealed this genotype. The reason for the infrequent homozygous polymorphism in our population is not clear. Methodologic problems are unlikely, because we performed every PCR twice and accepted only consistent results. Most likely, the low frequency of CYP1A1 polymorphism is a regional variation.
Otolaryngology– Head and Neck Surgery Volume 128 Number 5
GRONAU et al
679
Table 5. Polymorphism in GSTM1 and GSTT genes in patients with tumors and control subjects
GSTM1 positive GSTM1 negative GSTT1 positive GSTT1 negative GSTM1 and/or GSTT1 positive GSTM1 and GSTT1 negative
Patients with tumors
Control subjects
107 (57.2%) 80 (42.8%) 157 (83.9%) 30 (16.1%) 160 (85.6%) 27 (14.4%)
71 (51.1%) 68 (48.9%) 118 (84.9%) 21 (15.1%) 130 (93.5%) 9 (6.5%)
Table 6. GSTM1 polymorphism and site of cancer
Tumor site
Larynx
Hypopharynx
Oropharynx, oral cavity
GSTM1 positive GSTM1 negative
14 (26.4%) 39 (73.6%)
18 (56.3%) 14 (43.7%)
38 (44.7%) 47 (55.3)
In the CYP2D6 gene, we did not find relevant differences between patients and control subjects. The allelic frequencies of extensive, intermediate, and poor metabolizer genotypes in our patient or control population are consistent with previous reports.3,4 The supergene family of glutathione S-transferases includes enzymes of phase II detoxification, which can be grouped into 4 classes: ␣, , , and . Except for , each class contains several genes. They catalyze the conjugation of reduced glutathione with a large number of electrophils. This results in increased water solubility, allowing renal excretion of various carcinogens or epoxides formed during phase I detoxification. Five isoenzymes of GSTM have been identified.6 The glutathione S-transferase gene consist of 2 classes: the GSTT1 and GSTT2 genes. Homozygous deletions in the GSTM1 or GSTT1 gene result in missing enzyme activity due to the lack of gene product.5,8 Association studies of GSTM1 polymorphisms and lung cancer or cancer of the upper aerodigestive tract yielded contradictory results. This did not change if the lifetime smoking history was taken into account.10,12,15,19,20 In this investigation, no association was found between the frequency of GSTM1 null genotype alone or the GSTT1 null genotype alone and head and neck cancer. In analyzing only heavy smokers and drinkers, we also found no differences in the distribution of the genotypes. However, the coin-
Unknown primary, other site
P value
9 (52.9%) 8 (47.1%)
⬎0.05 ⬍0.05
cidence of GSTM1 and GSTT1 null genotype was more frequent in patients than in control subjects and especially in patients who were heavy smokers and drinkers (P ⬍ 0.01). In patients and control subjects who were nonsmokers or who had a smoke history of ⬍20 pack-years and did not drink alcohol daily, no association of the gene polymorphisms alone or in coincidence was found (data not shown). The simultaneous absence of the 2 enzymes GSTT1 and GSTM1 could result in the accumulation of electrophil intermediates because of reduced phase II detoxification capacity. These carcinogens lead to the formation of DNA-adducts and finally to DNA damage. Accumulation of carcinogenic intermediate metabolites in tobacco smoke-exposed tissues might result in tumor growth. Concomitant alcohol and tobacco abuse is a multiplicative risk factor for squamous cell carcinomas of the upper aerodigestive tract. The investigated polymorphisms are phase I and phase II enzymes that are in particular responsible for the detoxification of polycyclic aromatic hydrocarbons and aromatic amines found in tobacco smoke. Interestingly, an association of the GSTM1 null genotype with laryngeal carcinoma was observed. The larynx is that part of the upper aerodigestive tract that is most exposed to tobacco smoke and not directly to alcohol. On the one hand, it is possible that in laryngeal mucosa the
Otolaryngology– Head and Neck Surgery May 2003
680 GRONAU et al
deficient GSTM1 enzyme cannot be compensated for by other enzymes. On the other hand, it could be that in the larynx, the effect of the GSTM1 null polymorphism is not biased by the interference of alcohol. Because of a possible association between smoking and drinking habits and the genotype of several detoxification enzymes,13,14 we matched the patient and the control groups. The 2 groups were similar in smoking and drinking habits. Also, the gender and the age of patients and control subjects were similar. In conclusion, the lack of a single detoxification enzyme was not associated with head and neck cancer in this investigation. The simultaneous null genotype of GSTM1 and GSTT1 was more common in cancer patients than in control subjects. Moreover, the GSTM1 null genotype was overrepresented in patients with laryngeal cancer. Additional investigations on the tissue-specific concentrations of detoxification enzymes are necessary to evaluate possible sequelae of a single deficient enzyme in certain tissues such as laryngeal or oropharyngeal mucosa. Therefore, it is also necessary to determine the enzyme concentration and enzyme activity in tissue to acquire information about the phenotype of genes. REFERENCES
1. Bale A, Nebert D, McBride O. Subchromosomal localisation of the dioxin-inducible P-450 locus (CYP1) and description of two RFLPs detected with a 3⬘ P-450 cDNA probe. Cytogenet Cell Genet 1987;46:574-5. 2. Nakachi K, Imai K, Hayashi S, et al. Genetic susceptibility to squamous cell carcinoma of the lung in relation to cigarette smoking dose. Cancer Res 1991;51:5177-80. 3. Rannung A, Alexandrie AK, Persson I, et al. Genetic polymorphism of cytochromes 1A1, 2D6 and 2E1: regulation and toxicological significance. J Occup Environ Med 1995;37:25-36. 4. Anwar WA, Abdel-Rahman SZ, El-Zein RA, et al. Genetic polymorphism of GSTM1, CYP2E1 and CYP2D6 in Egyptian bladder cancer patients. Carcinogenesis 1996;17:1923-9. 5. Roots I, Drakoulis N, Brockmo¨ ller J. Polymorphic enzymes and cancer risk: concepts, methodology and data review. In: Kalow W, editor. Pharmacogenetics of Drug
6.
7. 8.
9.
10. 11. 12.
13.
14. 15. 16.
17. 18.
19.
20.
Metabolism. New York: Pergamon Presse; 1992. p. 81541. Hayes JD, Pulford DJ. The glutathione S-transferase supergene family: regulation of GST and the contribution of the isoenzymes to cancer chemoprotection and drug resistance. Crit Rev Biochem Mol Biol 1995;30:445-600. Board P. Biochemical genetics of glutathione S-transferase in man. Am J Hum Genet 1981;33:36-43. Pemble S, Schroeder KR, Spencer SR, et al. Human glutathione S-transferase theta (GSTT1): cDNA cloning and the characterisation of a genetic polymorphism. Biochem J 1994;300:267-71. Tang DL, Rundle A, Warburton D, et al. Association between both genetic and environmental biomarkers and lung cancer: evidence of a great risk of lung cancer in women smoker. Carcinogenesis 1998;19:1949-53. Ford JG, Li Y, O’Sullivan MM, et al. Glutathione Stransferase M1 polymorphism and lung cancer risk in African-Americans. Carcinogenesis 2000;21:1971-5. Sato M, Sato T, Izumo T, et al. Genetic polymorphism of drug-metabolizing enzymes and susceptibility to oral cancer. Carcinogenesis 1999;20:1927-31. Oude Ophius MB, van Lieshout EM, Roelofs HM, et al. Glutathione S-transferase M1 and T1 and cytochrome P4501A1 polymorphism in relation to the risk for benign and malignant head and neck lesions. Cancer 1999;82: 936-43. Saarikoski ST, Sata F, Husgafvel-Pursiainen K, et al. CYP2D6 ultrarapid metabolizer genotype as a potential modifier of smoking behaviour. Pharmacogenetics 2000; 10:5-10. Boustead C, Taber H, Idle JR, et al. CYP2D6 genotype and smoking behaviour in cigarette smokers. Pharmacogenetics 1997;7:411-4. Cheng L, Sturgis EM, Eicher SA, et al. Glutathione S-transferase polymorphisms and risk of squamous cell carcinoma of head and neck. Int J Cancer 1999;84:220-4. Madhusudan G, Prasann K, Kamala K. Human leucocyte glutathion S-transferase isoenzyme (class mu) and susceptibility to smoking-related cancers. J Toxicol Environ Health 1995;46:1-8. Smith G, Stanley LA, Sim E, et al. Metabolic polymorphisms and cancer susceptibility. Cancer Surv 1995;25: 27-75. Drakoulis N, Cascorbi I, Brockmo¨ ller J, et al. Polymorphisms in human lung cancer: exon 7 mutation (4889 A to G) and a T to C mutation in the 3⬘-flanking region. Clin Invest 1994;72:240-8. Bennett W, Alavanja MCR, Blomeke B, et al. Environmental tobacco smoke, genetic susceptibility and risk of lung cancer in never smoking women. J Natl Cancer Inst 1999;91:2009-14. Cascorbi I, Brockmo¨ ller J, Roots I. A C4887A polymorphism in exon 7 of human CYP1A1: population frequency, mutation linkages and impact on lung cancer susceptibility. Cancer Res 1996;56:4965-9.