Clin Genet 2000: 57: 270–277Printed in Ireland. All rights reser6ed

Original Article

Gender-specific effects of NAT2 andGSTM1 in bladder cancer

Schnakenberg E, Lustig M, Breuer R, Werdin R, Hubotter R,Dreikorn K, Schloot W. Gender-specific effects of NAT2 and GSTM1in bladder cancer.Clin Genet 2000: 57: 270–277. © Munksgaard, 2000

One approach for risk assessment of cancer is the evaluation of poly-morphic enzymes involved in cancer using molecular tools. Phase IIenzymes are involved in the detoxification of several drugs, environ-mental substances and carcinogenic compounds. Here, we analyzed en-zymes for their putative relevance in urinary bladder cancer. Thehereditable enzyme polymorphism of arylamine N-acetyltransferase 2(NAT2) and glutathione S-transferase M1 (GSTM1) and T1 (GSTT1)was studied in 157 hospital-based patients and in 223 control subjects.Slow acetylation was not observed to be a significant risk factor ofdeveloping bladder cancer (OR: 1.33; 95% CI 0.85–2.09). One genotyperesponsible for slow acetylation (NAT2*5B/*6A) was observed signifi-cantly more frequently in bladder cancer patients compared with con-trol subjects (OR: 1.63; 95% CI 1.03–2.58).Gender-specific effects were observed when patients were divided intosubgroups. In male patients, slow acetylators were identified as carryinga significant increased risk of developing bladder cancer, in particularwhen the genotype NAT2*5B/*6A was combined with the GSTM1 nullgenotype (OR: 4.39; 95% CI 1.98–9.74). By contrast, the same geno-type combination significantly protected female patients from bladdercancer (OR: 0.21; 95% CI 0.06–0.80).

E Schnakenberga, M Lustiga,R Breuera, R Werdinb,R Hubotterb, K Dreikornb andW Schloota

a Center for Human Genetics and GeneticCounselling, University of Bremen,b Urological Division, St. Jurgen Hospital ofBremen, Germany

Corresponding author: Dr Eckart Schnaken-berg, Center for Human Genetics and Ge-netic Counselling, University of Bremen,Leobener Street ZHG, D-28359 Bremen,Germany

Received 4 November 1999, revised andaccepted for publication 31 December1999

N-acetyltransferase (NAT) polymorphism is a ge-netic trait related to the metabolism of severaldrugs and xenobiotics and is associated with thedevelopment of certain malignancies such as blad-der or lung cancer (1–3). NAT belongs to thephase II enzymes which are involved in the detox-ification of exogenous substances. Since the genestructures for human NAT1 and NAT2 were de-scribed (4), several studies have occurred concern-ing allelic variants and their influences in cancersand other diseases such as Parkinson’s disease (5),extracolonic manifestations of familial adenoma-tous polyposis (6) and allergic diseases in children(7). At least 17 alleles have been characterized inthe NAT2 gene (8) and nine alleles have beenreported in the NAT1 gene (9, 10).

Slow acetylation is said to be one risk factor fordeveloping bladder cancer and is determined byNAT2. Several studies with subjects occupationallyexposed to arylamine carcinogens, showed in-creased risks ranging from 2- to 17-fold (11–13).

Other studies could not confirm these associations(14–17).

Sex-specific differences in bladder cancer havebeen known for many years. In Germany, bladdercancer occurs with an incidence of 47/100000 sub-jects each year and male patients are affected 2.6times more frequently than female patients (18).

There is conflicting evidence regarding the roleof glutathione S-transferases (GSTs) in bladdercancer. In German bladder cancer patients,GSTM1 null genotypes are significantly more fre-quent than in the general population (19), whileGSTT1 was associated with cancer risk only innon-smokers (3, 20). The increased frequency ofhomozygous deletions of the GSTM1 gene in pa-tients with urothelial cancer was also observed inJapanese (21) and Egyptian bladder cancer pa-tients (22). On the other hand, a large multi-popu-lation study revealed no statistical evidence thatthe GSTM1 null genotype significantly increasesbladder cancer susceptibility (23). Therefore, we

270

Gender-specific effects (of NAT2 and GSTM1)

decided to analyze the GST polymorphism sepa-rately and in combination with NAT genotypes.

Here, we report our results from a study ofphase II enzyme polymophisms in bladder cancerpatients that were not selected according to anyoccupational criteria. To our knowledge, it is thefirst investigation that reports on gender-specificeffects concerning the development of bladder can-cer by N-acetyltransferase and glutathione S-trans-ferase genotypes.

Materials and methods

Study subjects

Blood samples from 157 unrelated bladder cancerpatients and 223 control subjects were obtainedfrom the northern part of Germany between 1995and 1997. Patients suffering from primary bladdercancer were hospitalized at the ‘Division of Urol-ogy of the St. Jurgen Hospital’, Bremen. None ofthe patients refused to participate. The case groupcomprised 110 male (mean age 68.99910.90 years)and 47 female (mean age 75.32910.41 years) pa-tients. Blood samples for the control group werederived from healthy blood donors of the St. Jur-gen hospital and students and could therefore notbe matched concerning age with bladder cancercases. The study was approved by the local ethicscommission.

All patients gave their informed consent andwere interviewed concerning risk factors like to-bacco smoking, occupational environment andprevious infections of the urinary tract. Patientsand control subjects were Caucasians and perma-nent residents of the northern part of Germany.Tumor stages and grading were evaluated fromprimary bladder cancer tissues according to theWorld Health Organization of International Clas-sification of Diseases in Oncology (24). In total,about 50% of patients with bladder cancer of thisarea could be analyzed. Tumor sampling was notperformed if bladder tumor specimens were toosmall for additional applications other than diag-nostic analysis. At least 10 mg of tumor tissuefrom each patient could be prepared.

The control group consisted of 223 individualswith no diseases reported at the time of theircontribution to NAT2 genotyping. There were 132male (mean age 37.60913.38 years) and 91 female(mean age 37.61914.23 years) subjects. The con-trol subjects had no history of malignant diseases.No data were evaluated concerning smoking oroccupation.

Genotyping using PCR

DNA was prepared from frozen blood samplescollected in tubes containing 100 ml EDTA (15%).After thawing, DNA was isolated as described byLahiri and Nurnberger (25). Briefly, 500 ng ofisolated DNA was added to a PCR reactionmixture.

NAT2 genotyping was carried out as describedby Schnakenberg et al. (26) and nomenclature wastaken from Vatsis et al. (27).

The method used to detect the GSTM1 andGSTT1 genotypes was adopted from Arand et al.(28). This method was completed by primer se-quences specific to the interferon gene as an inter-nal standard reaction as described by Brockmolleret al. (3). For multiplex PCR, 0.5 mmol/l of eachprimer, 200 mmol/l of each dNTP and 2.5 U Taqpolymerase were applied to a total volume of 50 ml.PCR was performed consisting of 30 cycles. Afteran initial melting temperature of 95°C (4 min),amplification was started by melting (95°C, 1 min),annealing (65°C, 1 min) and extension (72°C, 2min). A final 72°C extension step for 10 min wasperformed. The PCR products from coamplifica-tion of GSTM1, GSTT1 and the interferon genewere then electrophoresed on 3% Biozym smallDNA agarose (Biozym, Hess. Oldendorf, Ger-many) and visualized by ethidium bromide.

Statistical analysis

Odds ratio (OR) and 95% confidence intervals (CI)were used to describe the strength of association tothe disease using SPSS 8.0. The x2 test was used toprove statistical significance.

Results

Of 157 patients suffering from bladder cancer,72.6% were slow acetylators compared with 66.4%of the control group. Eight NAT2 alleles(NAT2*4; *13; *5A; *5B; *5C; *6A; *6B and *7B)were identified, giving rise to 36 different geno-types (Table 1). The slightly increased number ofslow acetylators do not indicate that slow acetyla-tion is linked to increased bladder cancer (OR:1.33; 95% CI 0.85–2.09). When the risk calculationwas performed for each genotype separately, onegenotype of slow acetylation was associated with asignificantly increased risk of developing bladdercancer. For the genotype NAT2*5B/2*6A, we cal-culated an OR of 1.63 (95% CI 1.03–2.58). Thedata for all calculated NAT2 alleles and genotypesare presented in Table 2.

271

Schnakenberg et al.

In the control group, the GSTM1 null genotypeoccurred with an incidence of 57.8% compared to59.2% in bladder cancer cases. Statistical compari-sons in male subjects showed no difference in theoccurrence of the GSTM1 null genotype betweentumor patients and control subjects. In female sub-jects, the GSTM1 null genotype occurred less fre-quently in both patients and control subjects(Table 3). The GSTT1 null genotype occurred in21.5% of control subjects and 17.8% of patients. Inmale patients, the GSTT1 null genotype was ob-served with a similar frequency as in control sub-jects. In female patients, the GSTT1 null genotypewas less frequent than in control subjects. This

difference did not reach significance. A typicalfragment pattern of the deletion of GSTM1 and/orGSTT1 gene by multiplex PCR is given in Fig. 1.

The risk assessment of the glutathione S-trans-ferases GSTM1 and GSTT1 revealed no increasedrisk of the occurrence of the GSTM1 or GSTT1null genotypes in bladder cancer patients (Table 3).Furthermore, the combination of GSTM1 orGSTT1 null genotypes with carriers of GSTM1 orGSTT1 genes (i.e. GSTM1 0*0 and GSTT1 1*1 orGSTM1 1*1 and GSTT1 0*0) showed no increasedrisks for the development of bladder cancer. Incontrast, there was a trend for null genotypes ofGSTM1 and GSTT1 (GSTM1 0*0+GSTT1 0*0)

Table 1. Distribution of 36 NAT2 genotypes in bladder cancer patients and control subjects

Controls femaleTumors femaleControls maleTumors maleControls totalTumors total

Genotype/rapid3 6 3 72*4/2*4 6 13

2*4/2*13 115 32*4/2*5A 8

159201135202*4/2*5B222*4/2*5C

52*4/2*6A 8 6 311 112*4/2*6B 2 1 2 1

112*4/2*7B112*13/2*13

2*13/2*5A1 12*13/2*5B 1 1

2*13/2*5C2 22*13/2*6A 1 2 1

2*13/2*6B2*13/2*7B

31204423Total/Rapid 7543

Genotype/slow12*5A/2*5A 12 1 1

1 72*5A/2*5B 2 23 91 22*5A/2*5C 1 2

22222*5A/2*6A12*5A/2*6B 1

2*5A/2*7B2*5B/2*5B 17 32 11 22 6 10

52*5B/2*5C 6 5 312512*5B/2*6A 25651 2645

334 142*5B/2*6B 132*5B/2*7B 1 3 1

1 12*5C/2*5C 1 13 7 2 32*5C/2*6A 5 10

2121422*5C/2*6B1 12*5C/2*7B 1 1

2*6A/2*6A 10 17 5 9 5 811112*6A/2*6B

2*6A/2*7B 5 3 4 2 1 12*6B/2*6B 2 1 212*6B/2*7B2*7B/2*7B

60278887148Total/Slow 114Total/Rapid 43 75 23 44 20 31

157 223 110 132 47Total 91

272

Gender-specific effects (of NAT2 and GSTM1)

Table 2. Risk calculations (OR; 95% CI) of male and female bladder cancer patients for NAT2 alleles and genotypes in combination with GSTM1 and GSTT1genotypes

GSTM1 0*0 GSTM1 1*0/1*1 GSTT1 0*0 GSTT1 1*0/1*11.06 (0.70–1.60) 0.94 (0.62–1.43) 0.79 (0.44–1.33) 1.26 (0.75–2.12)

NAT21.33 (0.85–2.09) 1.49 (0.80–2.75) 1.18 (0.61–2.30) 1.27 (0.47–3.39) 1.35 (0.82–2.23)Slow0.75 (0.48–1.17) 0.67 (0.36–1.24) 0.84 (0.44–1.64)Rapid 0.79 (0.29–2.11) 0.74 (0.45–1.23)

NAT2*4 0.71 (0.48–1.06) 0.69 (0.40–1.18) 0.74 (0.42–1.32) 0.61 (0.26–1.43) 0.75 (0.48–1.18)0.99 (0.74–1.34) 1.23 (0.84–1.81) 0.71 (0.45–1.14)NAT2*5B 1.28 (0.64–2.58) 0.92 (0.66–1.27)

NAT2*6A 1.31 (0.95–1.80) 1.15 (0.76–1.73) 1.57 (0.94–2.62) 0.88 (0.43–1.81) 1.45 (1.01–2.08)*1.63 (1.03–2.58)* 1.66 (0.92–2.99) 1.56 (0.75–3.24) 1.08 (0.38–3.04)NAT2*5B/*6A 1.80 (1.08–3.01)*0.79 (0.44–1.42) 1.20 (0.51–2.80) 0.54 (0.23–1.25)NAT2*4/*5B 0.72 (0.20–2.61) 0.87 (0.44–1.69)

NAT2*6A/*6A 0.83 (0.37–1.86) 0.97 (0.35–2.65) 0.61 (0.15–2.46) n.c.‡ 1.14 (0.48–2.73)0.73 (0.39–1.36) 1.00 (0.46–2.16)NAT2*5B/*5B 0.38 (0.12–1.22) 2.76 (0.43–17.63) 0.59 (0.30–1.16)1.46 (0.62–3.45) 0.86 (0.27–2.72) 3.14 (0.76–13.04) 2.76 (0.43–17.63) 1.22 (0.46–3.25)NAT2*4/*6A

* pB0.05. ‡ not calculated because control subjects n=0.

to play a protective role. On the other hand, thecombination of GSTT1 1*0/1*1 and the alleleNAT2*6A led to a significantly increased risk forthe development of bladder cancer (OR: 1.45; 95%CI 1.01–2.08). When the GSTT1 1*1 genotype wascombined with NAT2*5B/*6A, the risk rose to anOR of 1.80 (95% CI 1.08–3.01), as shown in Table2.

Gender-specific effects were observed when pa-tients were divided into subgroups (Fig. 2). Therisk of developing bladder cancer is significantlyincreased for male slow acetylators (OR: 1.87; 95%CI 1.04–3.36), while female slow acetylators carry-ing this genotype seemed to be protected againstbladder cancer, but this did not reach statisticalsignificance (OR: 0.70; 95% CI 0.34–1.44). Thehighest risk was observed for male patients withthe slow genotype NAT2*5B/*6A (OR: 2.85; 95%CI 1.61–5.05). On the other hand, the same geno-type showed a decreased risk of developing bladdercancer in females (OR: 0.39; 95% CI 0.15–1.02).Furthermore, when male slow acetylators also car-ried the GSTM1 0*0 genotype, the risk increasedsignificantly to an OR of 3.38 (95% CI 1.39–8.22).This was especially true for male slow acetylatorswith the genotype NAT2*5B/*6A where the riskrose to an OR of 4.39 (95% CI 1.98–9.74). Incontrast, the same combination of genotypesshowed a decreased risk in females with an OR of0.49 (95% CI 0.20–1.25 female slow acetylatorsand GSTM1 0*0) and a significantly decreased riskwith an OR of 0.21 (95% CI 0.06–0.80) for femaleswith the genotypes NAT2*5B/*6A and GSTM10*0. The opposite tendency was observed in rapidacetylators. While male rapid acetylators showed asignificantly decreased risk of developing bladdercancer, female rapid acetylators have an elevatedrisk, especially when the GSTM1 null genotypeoccurs in combination with NAT2*4/*5B. In addi-

tion, the rapid acetylator genotype NAT2*4/*6Aincreased the risk significantly for the developmentof bladder cancer in female patients, while in malepatients, the opposite trend was observed. Theseresults indicate that the assessment of risk of devel-oping bladder cancer must take into account gen-der-specific influences and that GSTM1 nullgenotypes can at least modify the risk of bladdercancer.

Discussion

This study of polymorphisms of NAT2, GSTM1and GSTT1 is the first one to characterize gender-specific effects in bladder cancer patients by geno-typing. Since it is possible to divide individuals intorapid or slow acetylators, several studies have beenperformed characterizing the acetylator status, es-pecially in bladder cancer patients. Now, molecularmethodologies are available to predict the N-acetyltransferase phenotype accurately (\99.2%)by genotyping (8) which could reduce expensivedrug-loading tests. In addition, genotypingmethodologies are a powerful tool for the assess-ment of the risk of developing cancer and otherdiseases.

Most of the published studies have identifiedslow acetylation as a susceptibility factor in thedevelopment of bladder cancer. Male subjectscarry an increased risk (2–3-fold) and the occur-rence of bladder cancer in males is 34/100000 peryear compared with 13/100000 in females (8). Thesame ratio could be observed in our study (72%male vs. 28% female patients), but the reasons forincreased susceptibility of male subjects cannot beexplained. Occupational and lifestyle factors havebeen discussed as increasing the risk of bladdercancer (30, 31), but cannot explain gender-specificeffects.

273

Schnakenberg et al.

In our study, patients were not selected accord-ing to occupational criteria and represented abroad range of professions. Furthermore, in riskassessment, genotyping methods carry the advan-tage of not being influenced by exogenous factors.Therefore, our results represent gender-specific ef-fects not selected on the basis of occupational andlifestyle factors.

In this study, six point mutations of the NAT2gene (C282T, T341C, C481T, G590A, A803G andG857A) were analyzed, leading to eight differentalleles and 36 genotypes. According to a largecohort study of 844 German individuals (29), thesepoint mutations occurred in our investigationswith a frequency of \1% and represented thegenetic polymorphism of the NAT2 gene in Cau-casians of this region. In contrast to the observa-tions of Cascorbi et al. (29), who reported 58.9% ofthe control subjects to be slow acetylators, wedetermined that 66.4% were slow acetylators.These differences might be explained by deviatedclassification of the NAT2*13 allele. To ourknowledge, this allele is characterized by one pointmutation at nt 282 and is not associated withanamino acid alteration. Vatsis et al. (27) reportedwild-type enzyme activity for this mutation. There-fore, we classified NAT2*13 under rapidacetylation.

The results of our investigations indicate that theglutathione S-transferases, GSTM1 and GSTT1,do not enhance the risk of developing of bladdercancer. The GSTM1 or GSTT1 genotype alone orin combination cannot be used as a predictivefactor in the risk assessment of bladder cancer butcan modify the risk of bladder cancer develop-ment, especially in male slow acetylator bladdercancer patients. The allelic distribution of theGSTM1 null genotype in our control group isslightly higher (57.8%) compared with Brockmolleret al. [50.7%, (19)] but does not differ significantlyfrom our bladder cancer patients (59.2%). These

Fig. 1. Multiplex PCR of GSTM1 and GSTT1. An intrna-tional standard (INF: interferon) was used to prove the PCRreaction. (M: marker, 1: without deletions, 2: deletion ofGSTM1, 3: deletion of GSTT1, 4: deletion of GSTM1 andGSTT1.)

274

Tabl

e3.

Odd

sra

tioca

lcul

atio

nof

GST

M1,

GST

T1an

dco

mbi

ned

geno

type

sof

blad

der

canc

erpa

tient

s

Male

Tota

lFe

male

n(%

)n

(%)

n(%

)

Case

sCo

ntro

lsO

R(9

5%CI

)Ca

ses

Cont

rols

OR

(95%

CI)

Case

sCo

ntro

lsO

R(9

5%CI

)

0.96

(0.5

8-1.

59)

GST

M1

(1*0

/1*1

)16

(10.

2)35

(15.

7)0.

84(0

.40–

1.72

)64

(40.

8)94

(42.

2)0.

94(0

.62-

1.43

)48

(30.

6)59

(26.

5)72

(32.

3)1.

08(0

.65-

1.79

)31

(19.

7)57

(25.

5)1.

19(0

.55–

2.42

)93

(59.

2)G

STM

10*

012

9(5

7.8)

1.06

(0.7

0-1.

60)

62(3

9.5)

109

(48.

9)1.

02(0

.52-

2.02

)38

(24.

2)66

(29.

5)0.

78(0

.49–

1.24

)G

STT1

(1*0

/1*1

)12

9(8

2.2)

175

(78.

5)1.

26(0

.75-

2.12

)91

(57.

9)0.

68(0

.29–

1.56

)26

(11.

7)10

(6.4

)0.

98(0

.49-

1.93

)G

STT1

0*0

22(9

.9)

18(1

1.5)

0.79

(0.4

4-1.

33)

48(2

1.5)

28(1

7.8)

51(2

2.9)

1.23

(0.7

4-2.

04)

11(7

.0)

26(1

1.7)

1.72

(0.8

5–3.

51)

GST

M1

(1*0

/1*1

)+G

STT1

(1*0

/1*1

)48

(30.

6)77

(34.

5)1.

36(0

.90-

2.05

)37

(23.

6)G

STM

1(1

*0/1

*1)+

GST

T10*

016

(10.

2)17

(7.6

)1.

38(0

.67-

2.81

)11

(7.0

)8

(3.6

)1.

72(0

.67-

4.45

)5

(3.2

)9

(4.0

)1.

09(0

.34–

3.44

)0.

76(0

.34–

1.72

)40

(17.

9)27

(17.

2)0.

80(0

.47-

1.37

)G

STM

10*

0+G

STT1

(1*0

/1*1

)58

(26.

0)54

(34.

4)0.

84(0

.54-

1.29

)98

(44.

0)81

(51.

6)G

STM

10*

0+G

STT1

0*0

12(7

.6)

31(1

3.9)

0.51

(0.2

5-1.

03)

8(5

.1)

14(6

.3)

0.66

(0.2

7-1.

64)

4(2

.5)

17(7

.6)

0.41

(0.1

3–1.

28)

Gender-specific effects (of NAT2 and GSTM1)

Fig. 2. Risk calculation (OR) of NAT2 genotypes and in combination with GSTM1 0*0 in male and female patients with bladdercancer (c pB0.05; § p\0.05). Female: CIs of the OR for NAT2*4/*5B+GSTM1 0*0 genotype were 1.26–16.85; NAT2*4/*6A:1.02–18.02. Male: CIs of the OR for slow+GSTM1 0*0 genotype were 1.39–8.22; NAT2*5B/*6A+GSTM1 0*0: 1.98–9.74 (datashown incompletely).

differences might be explained by the compositionof both our bladder cancer and the control group.Bladder cancer patients were not selected accord-ing to occupation and included a fairly high fre-quency of ‘spontaneous’ bladder cancer patients.On the other hand, our control group consisted ofyoung students and blood donors with an ageobviously lower than that of bladder cancer pa-tients. When we compared the frequency ofGSTM1 null genotypes of our control subjects

with other investigations of Caucasian origin, weobserved similar results to this study (23, 32).

The frequency of the GSTT1 null genotypeamong European populations varied between 10and 30%. In a Swedish population, a relatively lowfrequency (10%) of GSTT1 null genotypes wasobserved (33) compared with Great Britain [40%,(34)] and Germany [30–40%, (35)]. In this study,we observed a frequency of 21.5%. The relevanceof the GSTT1 null genotype in the susceptibility to

275

Schnakenberg et al.

cancer is inconclusive. In lung, oral, gastric andcolorectal cancers, no significant results wereidentified (36). In bladder cancer cases, an in-creased risk associated with the GSTT1 nullgenotypes was observed only in a group of non-smokers (20).

It is difficult to explain gender-specific differ-ences in phase II polymorphisms influencing therisk for developing bladder cancer. Enzyme poly-morphisms may influence the metabolism of sex-specific factors, i.e. hormones, and modify therisk of development of bladder cancer. Sex-spe-cific differences in lifestyle could also contributeto sex-specific effects in tumor disposition. More-over, environmental factors, such as the workingplace, could affect the disposition to bladder can-cer. Furthermore, sex-specific gene regulationcould enhance gene expression leading to differ-ences concerning the enzyme concentration. Inaddition, both exogenous and endogenous factorscould influence sex-specific tumor disposition.

References

1. Risch A, Wallace DMA, Bathers S, Sim E. Slow N-acetylation genotype is a susceptibility factor in occupa-tional and smoking related bladder cancer. Hum MolGenet 1995: 4: 231–236.

2. Cascorbi I, Brockmoller J, Mrozikiewicz PM, Bauer S,Loddenkemper R, Roots I. Homozygous rapid arylamineN-acetyltransferase (NAT2) genotype as susceptibilityfactor for lung cancer. Cancer Res 1996: 56: 3961–3966.

3. Brockmoller J, Cascorbi I, Kerb R, Roots I. Combinedanalysis of inherited polymorphisms in arylamine N-acetyltransferase 2, glutathione S-transferases M1 andT1, microsomal epoxide hydrolase, and cytochrome P450enzymes as modulators of bladder cancer risk. CancerRes 1996: 56: 3915–3925.

4. Blum M, Grant DM, McBride OW, Heim M, MeyerUA. Human arylamine N-acetyltransferase genes: isola-tion, chromosomal localization and functional expres-sion. DNA Cell Biol 1990: 9: 193–203.

5. Bandmann O, Vaughan P, Holmans P, Marsden CD,Wood NW. Association of slow acetylator genotype forN-acetyltransferase 2 with familial Parkinson’s disease.Lancet 1997: 350: 1136–1139.

6. Scott RJ, Taeschner W, Heinimann K et al. Associationof extracolonic manifestations of familial adenomatouspolyposis with acetylation phenotype in a large FAP kin-dred. Eur J Hum Genet 1997: 5: 43–49.

7. Zielinska E, Niewiarowski W, Bodalski J, Stanczyk A,Bolanowski W, Rebowski G. Arylamine N-acetyltrans-ferase (NAT2) gene mutations in children with allergicdiseases. Clin Pharmacol Ther 1997: 62: 635–642.

8. Agundez JAG, Olivera M, Martınez C, Ladero JM, Ben-ıtez J. Identification and prevalence study of 17 allelicvariants of the human NAT2 gene in a white population.Pharmacogenetics 1996: 6: 423–428.

9. Deitz AC, Doll MA, Hein DW. A restriction fragmentlength polymorphism assay that differentiates human N-acetyltransferase-1 (NAT1) alleles. Anal Biochem 1997:253: 219–224.

10. Butcher NJ, Ilett KF, Minchin RF. Functional polymor-phism of the human arylamine N-acetyltransferase type 1gene caused by C190T and G560A mutations. Pharmaco-genetics 1998: 8: 67–72.

11. Cartwright RA, Glasham RW, Rogers HJ et al. The roleof N-acetyltransferase phenotypes in bladder carcinogen-esis: a pharmacogenetic epidemiological approach tobladder cancer. Lancet 1982: 2: 842–845.

12. Weber WW, Hein DW, Litwin A, Lower GM. Relation-ship of acetylator status to isoniazid toxicity, lupus ery-thematosus and bladder cancer. FASEB J 1983: 42:3086–3097.

13. Hanke J, Krajewska B. Acetylation phenotype and blad-der cancer. J Occup Med 1990: 32: 917–918.

14. Woodhouse KW, Adams PC, Clothier A, Mucklow JC,Rawlins MC. N-acetylation phenotype in bladder cancer.Hum Toxicol 1982: 1: 443–445.

15. Ladero JM, Kwok CK, Jara C, Fernandez L, Silmi AM,Tapia D, Uson AC. Hepatic acetylator phenotype inbladder cancer patients. Ann Clin Res 1985: 17: 96–99.

16. Karakaya AE, Cok I, Sardas S, Gogus O, Sardas OS.N-acetyltransferase phenotype of patients with bladdercancer. Hum Toxicol 1986: 5: 333–335.

17. Hayes RB, Bi W, Rothman N et al. N-acetylation phe-notype and genotype and risk of bladder cancer in ben-zidine exposed workers. Carcinogenesis 1993: 14:675–678.

18. Golka, K., Bandel, T., Urfer, W. and Bolt, H.M. (1995)Berufliche Risikofaktoren fur Tumoren der Harnblaseund der Prostata. Bundesanstalt fur Arbeitsschutz, Dort-mund.

19. Brockmoller J, Kerb R, Drakoulis N, Staffeldt B, RootsI. Glutathione S-transferase M1 and its variants A and Bas host factors of bladder cancer susceptibility: a case-control study. Cancer Res 1994: 54: 4103–4111.

20. Kempkes M, Golka K, Reich S, Reckwitz T, Bolt HM.Glutathione S-transferase GSTM1 and GSTT1 nullgenotypes as potential risk factors for urothelial cancerof the bladder. Arch Toxicol 1996: 71: 123–126.

21. Katoh T, Inatomi H, Nagaoka A, Sugita A. CytochromeP4501A1 gene polymorphism and homozygous deletionof the glutathion S-transferase M1 gene in urothelialcancer patients. Carcinogenesis 1995: 16: 655–657.

22. Anwar WA, Abdel-Rahmann SZ, El-Zein RA, MostafaHM, Au WW. Genetic polymorphism of GSTM1,CYP2E1 and CYP2D6 in Egyptian bladder cancer pa-tients. Carcinogenesis 1996: 17: 1923–1929.

23. Lin HJ, Han CY, Bernstein DA, Hsiao W, Lin BK,Hardy S. Ethnic distribution of the glutathione trans-ferase Mu 1-1 (GSTM1) null genotype in 1473 individu-als and application to bladder cancer susceptibility.Carcinogenesis 1994: 15: 1077–1081.

24. Hermanek P, Scheibe O, Spiessl B, Wagner G. TNM-Klassifikation Maligner Tumoren, 4. Berlin: AkademieVerlag, 1989.

25. Lahiri DK, Nurnberger JI. A rapid non-enzymaticmethod for the preparation of HMW DNA from bloodfor RFLP studies. Nucleic Acids Res 1991: 19: 5444.

26. Schnakenberg E, Ehlers C, Feyerabend W et al. Geno-typing of the polymorphic N-acetyltransferase (NAT2)and loss of heterozygosity in bladder cancer patients.Clin Genet 1998: 53: 396–402.

27. Vatsis KP, Weber WW, Bell DA et al. Nomenclature forN-acetyltransferases. Pharmacogenetics 1995: 5: 1–17.

276

Gender-specific effects (of NAT2 and GSTM1)

28. Arand M, Muhlbauer R, Hengstler J et al. A multiplexpolymerase chain reaction protocol for the simultaneousanalysis of the glutathion S-transferase GSTM1 andGSTT1 polymorphism. Anal Biochem 1996: 236: 184–186.

29. Cascorbi I, Drakoulis N, Brockmoller J, Maurer A, Sper-ling K, Roots I. Arylamine N-acetyltransferase (NAT2)mutations and their allelic linkage in unrelated Caucasianindividuals: correlation with phenotypic activity. Am JHum Genet 1995: 57: 581–592.

30. Kunze E, Chang-Claude J, Frentzel-Beyme R. Life styleand occupational risk factors for bladder cancer in Ger-many. Cancer 1992: 69: 1776–1790.

31. Hartge P, Harvey EB, Linehan WM et al. Unexplainedexcess risk of bladder cancer in men. J Natl Cancer Inst1990: 82: 1636–1640.

32. Delomenie C, Mathelier-Fusade P, Longuemaux S et al.Glutathione S-transferase (GSTM1) null genotype and

sulphonamide intolerance in acquired immunodeficiencysyndrome. Pharmacogenetics 1997: 5: 519–520.

33. Warholm M, Rane A, Alexandrie A-K, Monaghan G,Rannug A. Genotypic and phenotypic determination ofpolymorphic glutathione transferase T1 in a Swedish pop-ulation. Pharmacogenetics 1995: 5: 252–254.

34. Pemble S, Schroder KR, Spencer SR et al. Human glu-tathione S-transferase theta (GSTT1): cDNA cloning andthe characterization of a genetic polymorphism. Biochem J1994: 300: 271–276.

35. Peter H, Deutschmann S, Reichel C, Hallier E.Metabolism of methyl chloride by human erythrocytes.Arch Toxicol 1989: 63: 351–355.

36. Deakin M, Elder J, Hendrickse C et al. Glutathione S-transferase GSTT1 genotypes and susceptibility to cancer:studies of interactions with GSTM1 in lung, oral, gastricand colorectal cancers. Carcinogenesis 1996: 17: 881–884.

277

.