GSTA1, GSTM1, GSTP1, and GSTT1 polymorphisms and susceptibility to smoking-related bladder cancer: A case-control study

Urologic Oncology: Seminars and Original Investigations 31 (2013) 1184 –1192

Original article

GSTA1, GSTM1, GSTP1, and GSTT1 polymorphisms and susceptibility to smoking-related bladder cancer: A case-control study夡 Marija Matic, M.D., Ph.D.a,b, Tatjana Pekmezovic, M.D., Ph.D.a,d, Tatjana Djukic, M.D.a,b, Jasmina Mimic-Oka, M.D., Ph.D.a,b, Dejan Dragicevic, M.D., Ph.D.a,c, Biljana Krivic, M.D., M.S.c, Sonja Suvakov, M.D.a,b, Ana Savic-Radojevic, M.D., Ph.D.a,b, Marija Pljesa-Ercegovac, M.D., Ph.D.a,b, Cane Tulic, M.D., Ph.D.a,c, Vesna Coric, M.D.a,b, Tatjana Simic, M.D., Ph.D.a,b,* a

b

Faculty of Medicine, University of Belgrade, Belgrade, Serbia Institute of Medical and Clinical Biochemistry, Belgrade, Serbia c Clinic of Urology, Clinical Centre of Serbia, Belgrade, Serbia d Institute of Epidemiology, Belgrade, Serbia

Received 1 June 2011; received in revised form 5 August 2011; accepted 5 August 2011

Abstract Objectives: Glutathione S-transferases (GSTs) are a family of enzymes involved in detoxification. Genes encoding for GSTA1, GSTM1, GSTP1, and GSTT1 proteins are polymorphic, which can result in complete or partial loss of enzyme activity. Previous studies have associated polymorphisms of GSTA1, GSTM1, and GSTP1 genes with a higher risk of bladder cancer, but this is still controversial. Potential role of GSTA1 polymorphism in susceptibility to bladder cancer in Whites is lacking. We examined association between GSTA1, GSTM1, GSTP1, and GSTT1 gene variants and bladder cancer risk and evaluated whether they were modified by smoking. Materials and methods: A hospital-based case-control study recruited 201 incidence cases and 122 age-matched controls. Deletion polymorphism of GSTM1 and GSTT1 was identified by polymerase chain reaction method. Single nucleotide polymorphism of GSTA1 and GSTP1 was identified by restriction fragment length polymorphism method. Uniconditional multivariate logistic regression was applied to model association between genetic polymorphisms and bladder cancer risk, as well as effect modification by smoking. Results: No significant difference was observed in the distributions of GSTM1, GSTT1, GSTA1, and GSTP1 gene variants between patients and controls. None of the examined polymorphisms was significantly associated with bladder cancer risk independently. The results of gene–smoking interaction analyses indicated a significant combined effect of smoking and all common GST polymorphisms tested (P for trend ⫽ 0.001). However, the most significant effect on bladder cancer risk was observed in smokers carrying lower activity GSTA1-AB/BB and GSTM-null genotype (OR ⫽ 3.5, P ⬍ 0.05) compared with GSTA1-AA and GSTM1-active non-smokers. Overall, the risk observed did not significantly differ with respect to quantity of cigarettes smoked. However, heavy smokers with GSTM1-null genotype had 2 times higher risk of bladder cancer than GSTM1-null light smokers (OR ⫽ 4.8 vs. OR ⫽ 2.0) when GSTM1-active non-smokers served as reference group. Smokers carrying both GSTM1-null and GSTA1-AB ⫹ BB genotypes exhibited the highest risk of bladder cancer (OR ⫽ 2.00, P ⫽ 0.123). Conclusions: Null or low-activity genotypes of the GSTA1, GSTM1, GSTT1, and GSTP1 did not contribute independently towards the risk of bladder cancer in our patients. However, in association with smoking, both low activity GSTA1 and GSTM1-null genotype increase individual susceptibility to bladder cancer. © 2013 Published by Elsevier Inc. Keywords: Glutathione S-transferase (GST); Polymorphism; Bladder cancer; Smoking

1. Introduction 夡 This work was supported by grant 175052 from the Serbian Ministry of Education and Science. * Corresponding author. Tel.: ⫹381-11-2645-750; fax: ⫹381-11-2645750. E-mail address: [email protected] (T. Simic).

1078-1439/$ – see front matter © 2013 Published by Elsevier Inc. http://dx.doi.org/10.1016/j.urolonc.2011.08.005

Bladder cancer is the fourth most common cancer in men and was estimated to be the eighth leading cause of cancerrelated deaths in men in 2009 [1]. The incidence of bladder cancer varies 10-fold internationally. Western Balkans is a

M. Matic et al. / Urologic Oncology: Seminars and Original Investigations 31 (2013) 1184 –1192

medium risk area, with Serbia exhibiting incidence of 10/ 100,000 in males and 5/100,000 females [2]. There were 145,000 deaths related with bladder cancer, with populationbased 5-year survival rates ranging from 40% to 80% [3]. Tobacco is the main risk factor for bladder cancer, as approximately one-half of male urinary tract cancer and one-third of female urinary tract cancer are attributable to cigarette smoking [4], while occupational exposure, particularly to aromatic amines, may play role in perhaps 10% of bladder cancer [5]. Cigarette smoke is a rich source of free radicals, which are believed to be responsible for initiation of many tumors by inducing DNA damage that accumulates in cells. In addition to free radicals, more than 60 carcinogens have been found in cigarette smoke. Among these, sufficient evidence of carcinogenicity was found for polycyclic aromatic hydrocarbons (PAHs), such as benzo(a)pyrene and aromatic amines (such as 4-amino biphenyl) [6]. Carcinogenic PAHs require metabolic activation to elicit their detrimental effects. Both reactive PAH metabolites and arylamines [7–9] are detoxified by cytosolic members of glutathione S-transferase (GST) family of enzymes. The most well-characterized GST classes have been named ␣ (GSTA), ␮ (GSTM), ␲ (GSTP), and ␪ (GSTT). Appreciable GST activities are seen in bladder epithelium [10]. GST enzymes that belong to various classes have different, but sometimes overlapping, substrate specificities. Thus, the activity towards PAHs metabolites has been reported for GSTM1, GSTP1, and GSTA1 enzymes [11,12], while GSTM1 is involved in the metabolism of arilamines [13]. GSTT1 substrates include halogenated solvents formed endogenously from ethane, which is also present at high levels in cigarette smoke [14]. All above mentioned GST enzymes possess strong peroxidase activity and are key components in cellular defense against free radicals [15]. Several types of allelic variations have been identified within GST classes, with that in the GSTM1, GSTT1 and GSTP1 genes receiving the most attention in genetic epidemiologic studies [16 –22]. The role of GSTA1 polymorphism has emerged relatively recently in genetic epidemiologic studies. It is represented by 3 apparently linked, single nucleotide polymorphisms (SNPs): -567TOG, -69COT, -52GOA [23]. These substitutions result in differential expression with lower transcriptional activation of variant GSTA1*B (-567G, -69T, -52A) than common GSTA1*A allele (-567T, -69C,-52G) [23,24]. Due to the fact that inactivation of tobacco-related carcinogens and free radicals may be influenced by the activity of GSTs, numerous studies examined whether GSTM1, GSTT1 and GSTP1 polymorphisms modify the risk of bladder cancer associated with smoking. The GSTM1-null polymorphism has been consistently associated with an increased risk of bladder cancer in pooled and meta-analyses [11,25–27]. On the other hand, studies investigating the importance of GSTT1 and GSTP1 genetic polymorphisms in bladder carcinogenesis are more limited and inconsistent [12,27–36]. Previous studies have differed on whether GST

1185

polymorphisms modify the association between cigarette smoking and bladder cancer risk. Both smaller studies and meta-analyses have reported that smoking did not modify associations between GSTT1 [37], GSTM1 [11,25,27], or GSTP1 [38] polymorphisms and bladder cancer, whereas another study reported positive associations between bladder cancer risk and GSTM1-null [28,39,40] genotype as well as GSTP1Ile allele [41] in smokers. Interestingly, despite the fact that GSTA1 is also involved in detoxification of carcinogens present in tobacco smoke, the role of this polymorphism has been addressed in conjunction with risk of bladder cancer in only one study in Japan [42]. In this study, we examined the association between polymorphism within GSTA1, GSTM1, GSTP1, and GSTT1 genes and bladder cancer risk as well as the potential of effect modification by cigarette smoking in a case-control study in Serbian patients. Additionally, since the risk associated with each variant may be small, we studied potential joint and cumulative effects of GSTA1, GSTM1, GSTP1, and GSTT1 genotypes.

2. Materials and methods 2.1. Study subjects A hospital-based case-control study of urinary bladder carcinoma was carried out between September 2007 and January 2010. A total of 201 histologically confirmed urinary bladder carcinoma cases were recruited from the Clinics of Urology, Clinical center of Serbia, Belgrade. This is the national reference center for urology and nephrology, and the majority of bladder cancer cases from Serbia are diagnosed and treated at the Clinics. The control group consisted of 122 subjects who were recruited from individuals with nephrolithiasis admitted to the same hospital during the same period of time and had no history of any malignant disease. Urinary bladder carcinoma patients and corresponding controls did not differ with respect to mean age, gender, and educational level (Table 1). All the participants provided written informed consent. This study protocol was approved by the Ethical Committee of Medical Faculty, University of Belgrade and the research was carried out in compliance with the Helsinki Declaration. After informed consent was obtained, each subject was interviewed by well-trained interviewers using a standard questionnaire to collect information including demographic characteristics, history of cigarette smoking, and occupational exposure. Smoking status was categorized into never smokers, light smokers, and heavy smokers. The cut-off point between the last 2 categories was based on the median of cumulative smoking exposure among controls (20 pack-years). Study subjects who consumed more than 20 pack-years were defined as heavy smokers, while those who consumed 20 or less pack-years were defined as light smokers. The amount

1186

M. Matic et al. / Urologic Oncology: Seminars and Original Investigations 31 (2013) 1184 –1192

Table 1 Selected characteristics of patients with bladder cancer and controls Characteristic Gender Male Female Age (years) Educational level Elementary school Secondary school University Smoking habits Never smokers Current smokers No. of pack-years of smoking Occupational exposure No Yes

Cases n (%)

Controls n (%)

114 (57) 87 (43) 64.4 ⫾ 10.1

71 (58) 51 (42) 62.7 ⫾ 6.0

OR (95%CI)

P 0.885

62 (31) 110 (55) 29 (14)

0.078 0.646

31 (25) 73 (60) 18 (15)

52 (26) 149 (74) 31.6 ⫾ 30.1

56 (46) 66 (54) 21.1 ⫾ 28.0

1.00 2.43 (1.51–3.92)

127 (64) 74 (36)

105 (86) 17 (14)

1.00 3.48 (1.88–6.39)

0.001 0.002

0.001

CI ⫽ confidence interval; OR ⫽ odds ratio.

of pack-years was calculated using the following formula: pack-years ⫽ (cigarettes/day⫼20) ⫻ (smoked years). 2.2. DNA extraction and genotyping Five ml of blood was collected in citrate vials from cases and controls. Blood was collected when patients were admitted to the clinic. DNA was extracted from blood lymphocytes using the QIAGEN QIAmp (Qiagen, Inc., Chatsworth, CA) 96-spin blood protocol according to the manufacturer’s instructions. Analysis for GSTM1 polymorphism was done by polymerase chain reaction (PCR) method as described by Garcia-Closas [43]. Isolated DNA (100 –150 ng) was amplified in a total volume of 25 ␮l reaction mixture containing 20 pmol of each of the following primers. Exon 7 of CYP1A1 genes were co-amplified and used as an internal control using following primers: GSTM1: Forward, 5=-GAACTCCCTGAAAAGCTAAAGC-3= Reverse, 5=-GTTGGGCTCAAATATACGGTGG-3=; CYP1A1: Forward, 5=-GAACTGCCACTTCAGCTGTCT-3= Reverse, 5=-CAGCTGCATTTGGAAGTGCTC-3=. The reaction mixture was subjected to initial denaturation at 94°C for 4 minutes, followed by 33 cycles of 94° C for 2 minutes, 59°C for 1 minute, and 72°C for 1 minute. The final extension was done at 72°C for 10 minutes. The PCR products were eletrophoresed in 2% agarose gels, and visualized by ethidium bromide staining. The absence of the PCR product (215 bp bands) was indicative of the GSTM1null genotype. This assay does not distinguish between heterozygous or homozygous wild type genotypes so the presence of 215 bp bands was indicative for GSTM1 active genotype (both heterozygous and homozygous wild type

genotypes). Internal positive control (CYP1A1) PCR product corresponded to 312 bp. Analysis for GSTT1 polymorphism was done by PCR method as described by Pemble et al. [44]. Isolated DNA (100 –150 ng) was amplified in a total volume of 25 ␮l reaction mixture containing 20 pmol of each of the following primers. Exon 7 of CYP1A1 genes were co-amplified and used as an internal control. GSTT1: Forward, 5=-TTCCTTACTGGTCCTCACATCTC-3=; Reverse, 5=-TCACGGGATCATGGCCAG CA-3=. The reaction mixture was subjected to initial denaturation at 94°C for 4 minutes, followed by 35 cycles of 94°C for 2 minutes, 61°C for 1 minute, and 72°C for 1 minute. The final extension was done at 72°C for 10 minutes. The PCR products were electrophoresed in 2% agarose gels, and visualized by ethidium bromide staining. The absence of the PCR product (480 bp bands) was indicative of the GSTT1-null genotype. This assay does not distinguish between heterozygous or homozygous wild type genotypes so the presence of 480 bp bands was indicative for GSTT1 active genotype (both heterozygous and homozygous wild type genotypes). Internal positive control (CYP1A1) PCR product corresponded to 312 bp. GSTP1Ile105Val polymorphism was analyzed using a previously described polymerase chain reaction–restriction fragment length polymorphism (PCR-RFLP) method according to Harries et al. [45]. Briefly, amplification was conceded using primers: GSTP1Ile105Val Forward, 5=-ACCCCAGGGCTCTATGGGAA-3= Reverse, 5=-TGAGGGCACAAGAAGCCCCT-3=. Thirty nanograms of DNA was amplified in a total volume of 30 ␮l containing 16.7 pmol of each primer. The amplification was performed by denaturing at 95°C for 5

M. Matic et al. / Urologic Oncology: Seminars and Original Investigations 31 (2013) 1184 –1192

minutes, followed by 33 cycles at 94°C for 30 seconds, annealing at 55°C for 30 seconds and 72°C for 30 seconds. The final extension was done at 72°C for 5 minutes. The amplification 176bp products (20 ␮l) were digested by 10 U of restriction endonuclease Alw261 at 37°C over night and eletrophoresed on 3% agarose gel. Presence of restriction site resulting in only 2 fragments (91 bp and 85 bp) indicated mutant allele (G/G), and if A/G polymorphism incurred then it resulted in three fragments of 176 bp, 91 bp, and 85 bp. Polymorphism GSTA1 C-69T was determined by PCRRFLP according to Coles et al. [24]. The primers used in the PCR were: GSTA1 C-69T: Forward, 5=-TGT TGA TTG TTT GCC TGA AAT T-3= Reverse, 5=-GTT AAA CGC TGT CAC CCG TCC T-3= Thirty nanograms of DNA was amplified in a total volume of 30 ␮l containing 16.7 pmol of each primer. The amplification was performed by denaturing at 95°C for 1 minute, followed by 32 cycles at 94°C for 60 seconds, annealing at 62°C for 60 seconds and 72°C for 60 seconds. The final extension was done at 72°C for 7 minutes. The amplification 481bp products (20 ␮l) were digested by 10 U of restriction endonuclease Ear1 at 37°C over night. Digest patterns were determined by resolution on 2% agarose gel. Presence of restriction site resulting in only 2 fragments (481bp and 385 bp) indicated mutant allele (B/B), and if A/B polymorphism incurred then it resulted in three fragments of 481bp, 385bp, and 96bp. All genotyping was performed by laboratory personnel blinded to case-control status, and blinded quality control samples were inserted to validate genotyping identification procedures; concordance for blinded samples was 100%. 2.3. Statistical analysis We used a ␹2 test to assess whether the GSTA1 and GSTP1 genotypes were in Hardy-Weinberg equilibrium and to determine P values for differences in genotype frequencies between cases and controls. The associations between the genotypes and bladder cancer risk were examined by using logistic regression to calculate odds ratios (ORs) and 95% confidence intervals (CIs). Based on the distribution of our data and the prior knowledge of bladder cancer, we adjusted OR for age and gender as potential confounders in multivariate logistic regression analysis.

3. Results Table 1 shows selected characteristics of patients with bladder cancer and their controls. As expected, the smoking prevalence among cases was higher (74%) than in controls (54%) and cases were more often (36%) occupationally exposed than controls (14%). We found that individuals

1187

Table 2 GSTA1, GSTM1, GSTT1, and GSTP1 genotypes in relation to risk of bladder cancer GST genotype GSTA1 AA AB BB AB⫹BB GSTM1 activea nullb GSTT1 activea nullb GSTP1 Ile/Ile Ile/Val Val/Val Ile/Val ⫹ Val/Val

Cases n (%)

Controls n (%)

OR (95% CI)

P

67 (33) 112 (56) 22 (11) 134 (67)

49 (40) 57 (47) 16 (13) 73 (60)

1.00 1.46 (0.89–2.41) 1.01 (0.45–2.25) 1.34 (0.82–2.20)

0.138 0.988 0.215

90 (72) 111 (55)

61 (50) 61 (50)

1.00 1.23 (0.77–1.99)

0.361

145 (72) 56 (28)

88 (72) 34 (28)

1.00 1.00 (0.59–1.70)

0.998

84 (42) 95 (47) 22 (11) 117 (58)

49 (40) 52 (43) 21 (17) 73 (60)

1.00 1.07 (0.63–1.79) 0.61 (0.29–1.29) 0.93 (0.58–1.52)

0.798 0.162 0.773

CI ⫽ confidence interval; OR ⫽ odds ratio. Adjusted for age, gender, and occupational exposure. a Active (present) if at least one active allele present. b Inactive (null) if no active alleles present.

who smoked had significantly increased risk of urinary bladder carcinoma (OR ⫽ 2.43, P ⫽ 0.001). We also observed that occupationally exposed individuals had significantly higher risk for urinary bladder carcinoma (OR ⫽ 3.48, P ⫽ 0.001). The distribution of GST polymorphisms among cases and controls is presented in Table 2. The prevalence of less active GST genotypes GSTA1-AB ⫹ BB, GSTM1-null and GSTP1Ile\Val⫹Val\Val is in agreement with the reported frequency in the literature. However, the prevalence of GSTT1-null genotype was slightly higher (28%) than expected in Caucasians. The GSTA1 and GSTP1 genotype frequencies were in Hardy-Weinberg equilibrium for cases and controls (P ⫽ 0.928 and P ⫽ 0.269, respectively). No significant difference was observed in the distributions of GSTM1, GSTT1, GSTA1, and GSTP1 gene variants between patients and controls. None of the examined polymorphisms was significantly associated with bladder cancer risk independently (Table 2). Our results do not support any joint or linear cumulative effect of less active GSTT1, GSTM1, GSTA1 and GSTP1 gene variants (data not shown). Combined effect of GSTA1, GSTM1, GSTP1 or GSTT1 polymorphisms and cigarette smoking on risk of bladder cancer is shown in Table 3. The results of gene-smoking interaction analyses indicated a significant effect between smoking and all common GST polymorphisms tested (P for trend ⫽ 0.001). Thus, in line with the biological role of GSTM1, smokers with GSTM -null genotype exhibited 3.6 times higher risk in comparison to non-smoking GSTM1 active individuals (Table 3). Interestingly, similar combined

1188

M. Matic et al. / Urologic Oncology: Seminars and Original Investigations 31 (2013) 1184 –1192

Table 3 Combined effect of smoking and GST genotype on risk of bladder cancer GST/smoking status GSTA1 AA/non-smokers AB ⫹ BB/non-smokers AA/smokers (ever) AB ⫹ BB/smokers (ever) P – trenda GSTM1 activeb/non-smokers nullc/non-smokers active/smokers (ever) null/smokers (ever) P – trend GSTT1 activeb/non-smokers nullc/non-smokers active/smokers (ever) null/smokers (ever) P – trend GSTP1 Ile/Ile/non-smokers Ile/Val ⫹ Val/Val/non-smokers Ile/Ile/smokers (ever) Ile/Val ⫹ Val/Val/smokers (ever) P – trend

Cases n (%)

Controls n (%)

OR (95%CI)

17 (9) 35 (17) 50 (25) 99 (49)

25 (21) 31 (25) 24 (20) 42 (34)

1.00 1.65 (0.73–3.72) 2.86 (1.20–6.50) 3.48 (1.60–7.54)

16 (8) 36 (18) 74 (37) 75 (37)

21 (18) 34 (28) 39 (32) 27 (22)

1.00 1.23 (0.50–2.80) 2.52 (1.20–5.40) 3.62 (1.60–8.10)

40 (20) 12 (6) 104 (52) 45 (22)

40 (33) 16 (13) 48 (39) 18 (15)

1.00 0.63 (0.20–1.50) 2.19 (1.20–3.90) 2.77 (1.30–5.80)

18 (9) 34 (17) 66 (33) 83 (41)

21 (17) 35 (29) 28 (23) 38 (31)

1.00 1.21 (0.50–2.80) 2.23 (1.00–5.10) 2.84 (1.30–6.30)

P

0.226 0.012 0.001 0.001

0.620 0.017 0.002 0.001

0.32 0.008 0.007 0.001

0.662 0.012 0.010 0.001

CI ⫽ confidence interval; OR ⫽ odds ratio. Adjusted for age, gender and smoking. a P value for trend. b Active (present) if at least one active allele present. c Inactive (null) if no active alleles present.

effect was observed for smoking and GSTA1 polymorphism. Smokers with less active GSTA1-AB ⫹ BB genotype exhibited 3.5 times higher risk of bladder cancer than non-smoking individuals with GSTA1-AA genotype (Table 3). In a manner similar to that observed for GSTM1 null and GSTA1-AB ⫹ BB, GSTT1 null and GSTP1Ile/Val⫹Val/ Val smoking interaction resulted in higher risk than smoking alone (Table 3). In order to test whether GST-smoking interaction is modified by the quantity of cigarettes smoked, cases and controls were further stratified into light and heavy smokers (Table 4). The risk observed did not significantly differ with respect to quantity of cigarettes smoked. However, heavy smokers with GSTM1-null genotype had 2 times higher risk of bladder cancer than GSTM1-null light smokers (OR ⫽ 4.8 vs. OR ⫽ 2.0) when GSTM1 active non-smokers served as reference group. The effect of GST genotype was analyzed also only in smokers and we found that heavy smokers with GSTM1 null genotype had 2.2 times higher risk of bladder cancer compared with GSTM1 active heavy smokers (OR ⫽ 2.2, P ⫽ 0.047). Since smokers with GSTM1-null genotype and less active GSTA1-AB ⫹ BB genotype exhibited the highest bladder carcinoma risk, we analyzed effect of GSTA1 and GSTM1 gene– gene interaction on urinary bladder carci-

noma risk in smokers (Table 5). Smokers carrying both GSTM1-null and GSTA1-AB ⫹ BB genotypes had the highest risk of bladder cancer (OR ⫽ 2.00, P ⫽ 0.123).

4. Discussion The results of this study showed that null or low-activity genotypes of the GSTA1, GSTM1, GSTT1 and GSTP1 did not contribute independently toward the risk of bladder cancer in Serbian group of patients. However, in association with smoking, significant trend in increase of urinary bladder cancer risk was observed for carriers of low activity or null GST genotypes. This effect was independent of the quantity of cigarettes smoked. GSTA1/M1 gene interaction enhanced the risk of bladder cancer in smokers. Glutathione transferases are of particular interest with respect to susceptibility to cancer because many genotoxic electrophiles are accepted as substrates. Polymorphism of GSTM1, GSTT1, and GSTP1 has been associated with occurrence of more than 20 types of cancer, including bladder cancer [46], which has been reported to be consistently associated with GSTM1 null genotype [47]. Although the risk of bladder cancer found in our study was higher in subjects with GSTM1-null genotype, the association was

M. Matic et al. / Urologic Oncology: Seminars and Original Investigations 31 (2013) 1184 –1192

1189

Table 4 Bladder cancer risk for GST genetic polymorphisms stratified by smoking GST/smoking status GSTA1 AA/non-smokers AA/light smokers AB ⫹ BB/light smokers AA/heavy smokers AB⫹BB/heavy smokers GSTM1 activea/non-smokers active/light smokers nullb/light smokers active/heavy smokers null/heavy smokers GSTT1 activea/non-smokers active/light smokers nullb/light smokers active/heavy smokers null/heavy smokers GSTP1 Ile/Ile/non-smokers Ile/Ile/light smokers Ile/Val ⫹ Ile/Val/light smokers Ile/Ile/heavy smokers Ile/Val ⫹ Ile/Val/heavy smokers

Cases (n%)

Controls (n%)

OR (95% CI)

P

17 (10) 10 (6) 28 (17) 40 (24) 71 (43)

25 (28) 8 (9) 13 (14) 16 (17) 29 (32)

1.00 2.13 (0.72–6.55) 3.26 (1.32–8.35) 3.31 (1.36–8.34) 3.42 (1.56–7.77)

0.211 0.013 0.010 0.030

16 (10) 20 (12) 18 (11) 54 (33) 57 (34)

22 (25) 10 (12) 11 (12) 29 (33) 16 (18)

1.00 2.85 (1.00–7.97) 2.02 (0.85–5.86) 2.33 (1.02–5.23) 4.76 (1.94–11.75)

0.043 0.161 0.047 0.001

40 (21) 28 (15) 10 (6) 76 (40) 35 (18)

40 (38) 18 (17) 3 (3) 30 (28) 15 (14)

1.00 1.72 (0.83–3.66) 3.65 (0.93–13.64) 2.35 (1.26–5.83) 2.64 (1.24–5.87)

0.213 0.062 0.009 0.019

18 (11) 17 (10) 21 (13) 49 (29) 62 (37)

21 (24) 10 (11) 11 (13) 18 (21) 27 (31)

1.00 2.41 (0.85–7.23) 2.32 (0.93–6.21) 2.71 (1.11–6.82) 3.06 (1.32–7.31)

0.124 0.092 0.029 0.012

CI ⫽ confidence interval; OR ⫽ odds ratio. Adjusted for age, gender, and occupational exposure. a Active (present) if at least one active allele present. b Inactive (null) if no active alleles present.

weak (OR ⫽ 1.23, P ⫽ 0.361) and did not reach magnitude obtained for GSTM1 in prior investigations, OR ⫽ 1.53 [26], OR ⫽ 1.58 [48]. Similar null findings, however, have been reported in the recent study of McGrath et al., who also failed to find association between GSTM1 polymorphism and bladder cancer [27]. Such findings might be due to the relatively small sample size resulting in low power to detect minor to modest genotype disease associations; therefore such associations cannot be ruled out. Another reason might be the variability of the results among different study populations and dependence of the risk degree on other population characteristics. Regarding population-specific charac-

teristics, it should be pointed out that large portion of bladder carcinoma cases in this study were from regions with Balkan endemic nephropathy. Our findings that individual GSTT1 or GSTP1 polymorphisms do not independently contribute to the risk of bladder cancer are in agreement with prior epidemiologic studies [37,28], namely, the potential impact of GSTT1 and GSTP1 polymorphism on bladder cancer susceptibility is less certain. Studies that have addressed the role of GSTT1 polymorphism and bladder cancer have been inconsistent. Of the published studies, some suggested increased risk [28,31,34,35] associated with the GSTT1-null genotype,

Table 5 GSTA1 and GSTM1 gene interaction in smokers and urinary bladder carcinoma risk GSTM1/GSTA1/smokers GSTM1 GSTM1 GSTM1 GSTM1

activea/GSTA1AA active/GSTA1AB⫹BB nullb/GSTA1AA null/GSTA1AB⫹BB

Cases n (%)

Controls n (%)

OR (95%CI)

Pc

21 (14) 53 (34) 29 (20) 46 (32)

16 (25) 23 (34) 8 (14) 19 (27)

1.00 1.60 (0.71–3.70) 1.90 (0.73–9.08) 2.00 (0.83–4.81)

0.257 0.200 0.123

CI ⫽ confidence interval; OR ⫽ odds ratio. Adjusted for age, gender, and occupational exposure. a Active (present) if at least one active allele present. b Inactive (null) if no active alleles present. c P value for interaction using likelihood ratio test.

1190

M. Matic et al. / Urologic Oncology: Seminars and Original Investigations 31 (2013) 1184 –1192

whereas others found no association and only one suggested a decreased risk [12]. Similarly, variant GSTP1 Val allele has been found to be associated with both increased or decreased risk of bladder cancer [41,49]. Since separate univariate analyses could misclassify individuals with null and low-activity genotypes of one gene into the group of non-null high activity genotypes [48], we performed both gene– gene interaction and linear cumulative analysis of all four GST genotypes tested. Still, we did not find any role of joint GST polymorphisms in determining individual’s susceptibility towards bladder cancer apart from environmental factors. Of the GST family, GSTA1 and other members of GST ␣ class have commonly been described as being the most versatile, since they are responsible for enzymatic biotransformation of compounds such as PAHs, N-acetoxy PhIP (a mutagenic derivative of the predominant heterocyclic amine carcinogen found in cooked meats), alkylating chemotherapeutic agents, steroids, byproducts of oxidative stress, and non-catalytic binding of haem, bilirubin, azodyes, and steroid hormones. Due to the fact that hepatic detoxification of N-acetoxy PhIP occurs via GSTA1 pathway, polymorphism of GSTA1 has been studied with respect to colorectal cancer and prostate cancer since evidence exists that PhIP is involved in their etiology. Although initial investigations [50 –52] gave conflicting results, recent meta-analyses revealed that neither colorectal nor the prostate cancer is associated with GSTA1 genetic polymorphism. Similarly, GSTA1 polymorphism was not associated with the incidence of hepatocellular cancer [53]. Our study has shown higher risk of bladder cancer in individuals carrying low-activity GSTA1 B allele, although the magnitude of association was relatively low and did not reach statistical significance (OR ⫽ 1.34, P ⫽ 0.215). Until now, the association between urothelial cancer and GSTA1 polymorphism has been addressed in only one Japanese study [42], which found increased association of GSTA1 polymorphism and urothelial cancer in never smokers. However, in our study strong modifying effect of GSTA1 polymorphism regarding the risk of bladder carcinoma was obtained in association with smoking. Smokers with low-activity GSTA1 genotype were at 3.5 times higher risk of bladder cancer in comparison with non-smoking subjects with high activity GSTA1. Similar association between low activity GSTA1 genotype and smoking has also been found for breast cancer risk [54]. It seems that modifying effect of GSTA1 on smoking-induced oxidative damage exists also for non-malignant diseases, since low-activity genotype has been shown to increase susceptibility to diabetes type 2 in smokers [55]. In addition to GSTA1, similar modifying effect was also obtained in our study for GSTM1-null genotype and smoking. Regarding GSTT1 and GSTP1 effects, the risk of bladder cancer obtained in GSTT1-null or GSTP1 Val/Val or Ile/Val smokers was not significantly higher than in smokers with GSTT1 active or GSTP1 Ile/Ile genotypes. Our findings that GSTM1-null and low-activity GSTA1 genotype increase susceptibility to bladder cancer in smok-

ers are biologically plausible, since cigarette smoking is recognized to be the most important risk factor for bladder cancer as potent source of PAHs and free radicals. However, it is important to note that uroepithelial cells do not express GSTA1, while their GSTM1 protein level is also relatively low in comparison to the dominant GSTP1 expression [56]. On the other hand, liver cells abundantly express GSTA1 and GSTM1 and thus participate in GSTA1 and GSTM1 mediated conjugation of PAH metabolites with glutathione, thereby enhancing their excretion in urine [57,58]. Taken together, these data suggest that liver, by its GSTs conjugating and peroxidase activity, plays a key role in protection against bladder carcinogens present in cigarette smoke. It seems reasonable to assume that carriers of GSTA1 B allele which have been shown to exhibit lower GSTA1 protein expression in liver (up to four times in case of GSTA1-BB genotype) than persons with GSTA1-AA genotype would be more susceptible to bladder cancer if they were also carriers of GSTM1-null genotype. Indeed, our results confirm such an assumption, since GSTA1GSTM1 gene interaction analysis revealed the highest risk of bladder cancer in smokers with low-activity GSTA1 and GSTM1 null genotype (OR ⫽ 2.00, P ⫽ 0.123). Further evidence in favor of GST modifying effect in smoking related carcinogenesis is a recent data showing that the extent of DNA damage in bi-nucleated cells of smokers is influenced by GST genetic polymorphism [59]. Namely, GSTM1-null individuals seemed to have higher susceptibility to tobacco smoke-induced DNA damage, assessed by micronuclei, compared with GSTM1-positive ones. In light of these findings, it would be tempting to address in the future whether risk GST genotypes, e.g., GSTM1-null and low activity GSTA1 in bladder cancer patients, are associated with elevated DNA damage end-points such as micronuclei or PAH-DNA adducts. It is well known that in case-control studies, selection bias might influence the results. Our control group was hospital-based and relatively small. Therefore, the use of population controls may have been more appropriate. In the future, attention should be paid on finding a larger sample, which would select participants throughout the country. Another important problem of a case-control study is that of the accuracy of the responses about remote events (smoking history), with the possibility of recall bias. Further limitation of the study might be the fact that we adjusted OR just for age, gender, and occupational exposure, without taking in consideration other potential confounding factors.

5. Conclusions Null or low-activity genotypes of the GSTA1, GSTM1, GSTT1, and GSTP1 did not contribute independently toward the risk of bladder cancer in our patients. However, in association with smoking, both low activity GSTA1 and

M. Matic et al. / Urologic Oncology: Seminars and Original Investigations 31 (2013) 1184 –1192

GSTM1-null genotype significantly increase individual susceptibility to bladder cancer.

References [1] Sexton WJ, Wiegand LR, Correa JJ, et al. Bladder cancer: A review of non-muscle-invasive disease. Cancer Control 2010;17:256 – 68. [2] Cancer incidence and mortality in Central Serbia, 2006. Belgrade: Institute of Public Health of Serbia. [3] Parkin DM, Bray F, Ferlay J, et al. Global cancer statistics, 2002. CA Cancer J Clin 2005;55:74 –108. [4] Zeegers MP, Tan FE, Dorant E, et al. The impact of characteristics of cigarette smoking on urinary tract cancer risk: A meta-analysis of epidemiologic studies. Cancer 2000;89:630 –9. [5] Kellen E, Zeegers M, Paulussen A, et al. Does occupational exposure to PAHs, diesel, and aromatic amines interact with smoking and metabolic genetic polymorphisms to increase the risk on bladder cancer? The Belgian case control study on bladder cancer risk. Cancer Lett 2007;245:51– 60. [6] IARC: Beryllium, cadmium, mercury, and exposures in the glass manufacturing industry. In IARC Monographs on the Evaluation of Carcinogenic Risk to Humans #58. Geneva: World Health Organization;1995:8 –13. [7] McGrath M, Michaud D, De Vivo I. Polymorphisms in GSTT1, GSTM1, NAT1, and NAT2 genes and bladder cancer risk in men and women BMC Cancer 2006;6:239. [8] Filiadis I, Hrouda D. Genetic factors in chemically-induced transitional cell bladder cancer. BJU Int 2000;86:794 – 801. [9] Jung I, Messing E. Molecular mechanisms and pathways in bladder cancer development and progression. Cancer Control 2000;7:325–34. [10] Simic T, Mimic-Oka J, Savic-Radojevic A, et al. Glutathione S-transferase T1-1 activity up-regulated in transitional cell carcinoma of urinary bladder. Urology 2005;65:1035– 40. [11] Engel LS, Taioli E, Pfeiffer R, et al. Pooled analysis and metaanalysis of glutathione S-transferase M1 and bladder cancer: A HuGE review. Am J Epidemiol 2002;156:95–109. [12] Kim WJ, Kim H, Kim CH, et al. GSTT1-null genotype is a protective factor against bladder cancer. Urology 2002;60:913– 8. [13] Vineis P. Epidemiology of cancer from exposure to arylamines. Environ Health Perspect 1994;102:7–10. [14] Hayes JD, Flanagan JU, Jowsey IR. Glutathione transferases. Annu Rev Pharmacol Toxicol 2005;45:51– 88. [15] Hayes JD, Strange RC. Glutathione S-transferase polymorphisms and their biological consequences. Pharmacology 2000;61:154 – 66. [16] Di Pietro G, Magno LA, Rios-Santos F. Glutathione S-transferases: An overview in cancer research. Expert Opin Drug Metab Toxicol 2010;6:153–70. [17] Eaton DL, Bammler TK. Concise review of the glutathione S-transferases and their significance to toxicology. Toxicol Sci 1999;49: 156 – 64. [18] Landi S. Mammalian class ␪ GST and differential susceptibility to carcinogens: A review. Mutat Res 2000;463:247– 83. [19] Ali-Osman F, Akande O, Antoun G, et al. Molecular cloning, characterization, and expression in Escherichia coli of full-length cDNAs of three human glutathione S-transferase Pi gene variants: Evidence for differential catalytic activity of the encoded proteins. J Biol Chem 1997;272:10004 –12. [20] Watson MA, Stewart RK, Smith GB, et al. Human glutathione Stransferase P1 polymorphisms: Relationship to lung tissue enzyme activity and population frequency distribution. Carcinogenesis 1998; 19:275– 80. [21] Hu X, O’Donnell R, Srivastava SK, et al. Active site architecture of polymorphic forms of human glutathione S-transferaseP1-1 accounts for their enantioselectivity and disparate activity in the glutathione

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

[37]

[38]

[39]

1191

conjugation of 7,8-dihydroxy-9, 10-oxy-7,8,9,10-tetrahydrobenzo (a) pyrene. Biochem Biophys Res Commun 1997;235:424 – 8. Sundberg K, Johansson AS, Stenberg G, et al. Differences in the catalytic efficiencies of allelic variants of glutathione transferase P1-1 towards carcinogenic diol epoxides of polycyclic aromatic hydrocarbons. Carcinogenesis 1998;19:433– 6. Coles BF, Morel F, Rauch C, et al. Effect of polymorphism in the human glutathione S-transferase A1 (hGSTA1) promoter on hepatic GSTA1 and GSTA2 expression. Pharmacogenetics 2001;11:663–9. Coles FB, Kadlubar FF. Human ␣ class glutathione S-transferases: Genetic polymorphism, expression, and susceptibility to disease. In: Helmut S, Lester P, editors. Glutathione Transferases and ␥-Glutamyl Transpeptidases, Methods Enzymology. London: Elsevier Academic Press, 2005. pp. 9 – 42. Garcia-Closas M, Malats N, Silverman D, et al. NAT2 slow acetylation, GSTM1 null genotype, and risk of bladder cancer: Results from the Spanish Bladder Cancer Study and meta-analyses Lancet 2005;366:649 –59. Johns LE, Houlston RS. Glutathione S-transferase ␮1 (GSTM1) status and bladder cancer risk: A meta-analysis. Mutagenesis 2000;15: 399 – 404. McGrath M, Michaud D, De Vivo I. Polymorphisms in GSTT1, GSTM1, NAT1, and NAT2 genes and bladder cancer risk in men and women BMC Cancer 2006;6:239. Hung RJ, Boffetta P, Brennan P, et al. GST, NAT, SULT1A1, CYP1B1 genetic polymorphisms, interactions with environmental exposures, and bladder cancer risk in a high-risk population. Int J Cancer 2004;110:598 – 604. Karagas MR, Park S, Warren A, et al. Gender, smoking, glutathioneS-transferase variants and bladder cancer incidence: A populationbased study. Cancer Lett 2005;219:63–9. Lee SJ, Cho SH, Park SK, et al. Combined effect of glutathione S-transferase M1 and T1 genotypes on bladder cancer risk. Cancer Lett 2002;177:173–9. Salagovic J, Kalina I, Habalova V, et al. The role of human glutathione S-transferases M1 and T1 in individual susceptibility to bladder cancer. Physiol Res 1999;48:465–71. Ma QW, Lin GF, Chen JG, et al. Polymorphism of glutathione S-transferase T1, M1, and P1 genes in a Shanghai population: Patients with occupational or non-occupational bladder cancer. Biomed Environ Sci 2002;15:253– 60. Giannakopoulos X, Charalabopoulos K, Baltogiannis D, et al. The role of N-acetyltransferase-2 and glutathione S-transferase on the risk and aggressiveness of bladder cancer. Anticancer Res 2002;22: 3801– 4. Brockmoller J, Cascorbi I, Kerb R, Roots I. Combined analysis of inherited polymorphisms in arylamine N-acetyltransferase 2, glutathione S-transferases M1 and T1, microsomal epoxide hydrolase, and cytochrome P450 enzymes as modulators of bladder cancer risk. Cancer Res 1996;56:3915–25. Kempkes M, Golka K, Reich S, et al. Glutathione S-transferase GSTM1 and GSTT1 null genotypes as potential risk factors for urothelial cancer of the bladder. Arch Toxicol 1996;71:123– 6. Simic T, Savic-Radojevic A, Pljesa-Ercegovac M, et al. Glutathione S-transferases in kidney and urinary bladder tumors. J. Nat Rev Urol 2009;6:281–9. Zeng FF, Liu SY, Wei W, et al. Genetic polymorphisms of glutathione S-transferase T1 and bladder cancer risk: A meta-analysis. Clin Exp Med 2010;10:59 – 68. Kellen E, Hemelt M, Broberg K, et al. Pooled analysis and metaanalysis of the glutathione S-transferase P1 Ile 105Val polymorphism and bladder cancer: A HuGE-GSEC review. Am J Epidemiol 2007; 165:1221–30. Bell DA, Taylor JA, Paulson DF, et al. Genetic risk and carcinogen exposure: A common inherited defect of the carcinogen-metabolism gene glutathione S-transferase M1 (GSTM1) that increases susceptibility to bladder cancer. J Natl Cancer Inst 1993;85:1159 – 64.

1192

M. Matic et al. / Urologic Oncology: Seminars and Original Investigations 31 (2013) 1184 –1192

[40] Lafuente A, Zakahary MM, el-Aziz MA, et al. Influence of smoking in the glutathione-S-transferase M1 deficiency-associated risk for squamous cell carcinoma of the bladder in schistosomiasis patients in Egypt. Br J Cancer 1996;74:836 – 8. [41] Cao W, Cai L, Rao JY, et al. Tobacco smoking, GSTP1 polymorphism, and bladder carcinoma Cancer 2005;104:2400 – 8. [42] Komiya Y, Tsukino H, Nakao H, et al. Human glutathion S-transferase A1 polymorphism and susceptibility to urothelial cancer in the Japanese population. Cancer Lett 2005;221:55–9. [43] Garcia-Closas M, Kelsey KT, Hankinson SE, et al. Glutathione Stransferase ␮ and ␪ polymorphisms and breast cancer susceptibility. J Natl Cancer Inst 1999;91:1960 – 4. [44] Pemble SE, Taylor JB. An evolutionary perspective on glutathione transferases inferred from class-␪ glutathione transferase cDNA sequences. Biochem J 1992;287:957– 63. [45] Harries LW, Stubbins MJ, Forman D, et al. Identification of genetic polymorphisms at the glutathione S-transferase Pi locus and association with susceptibility to bladder, testicular, and prostate cancer. Carcinogenesis 1997;18:641– 4. [46] Habdous M, Siest G, Herbeth B, et al. Glutathione S-transferases genetic polymorphisms and human diseases: Overview of epidemiologic studies. Ann Biol Clin (Paris) 2004;62:15–24. [47] Golka K, Schmidt T, Seidel T, et al. The influence of polymorphisms of glutathione S-transferases M1 and M3 on the development of human urothelial cancer. J Toxicol Environ Health A 2008;71:881– 6. [48] Yuan JM, Chan KK, Coetzee GA, et al. Genetic determinants in the metabolism of bladder carcinogens in relation to risk of bladder cancer. Carcinogenesis 2008;29:1386 –93. [49] Mittal RD, Srivastava DS, Mandhani A, et al. Genetic polymorphism of drug metabolizing enzymes (CYP2E1, GSTP1) and susceptibility to bladder cancer in North India. Asian Pac J Cancer Prev 2005;6: 6 –9.

[50] Sweeney C, Coles BF, Nowell S, et al. Novel markers of susceptibility to carcinogens in diet: Associations with colorectal cancer. Toxicology 2002;181/182:83–7. [51] Van der Logt EM, Bergevoet SM, Roelofs HM, et al. Genetic polymorphisms in UDP-glucuronosyltransferases and glutathione S-transferases and colorectal cancer risk. Carcinogenesis 2004;25:2407–15. [52] Ning B, Wang C, Morel F, et al. Human glutathione S-transferase A2 polymorphisms: Variant expression, distribution in prostate cancer cases/controls, and a novel form. Pharmacogenetics 2004;14:35– 44. [53] McGlynn KA, Hunter K, LeVoyer T, et al. Susceptibility to aflatoxin B1-related primary hepatocellular carcinoma in mice and humans. Cancer Res 2003;63:4594 – 601. [54] McCarty KM, Santella RM, Steck SE, et al. PAH-DNA adducts, cigarette smoking, GST polymorphisms, and breast cancer risk. Environ Health Perspect 2009;117:552– 8. [55] Oniki K, Umemoto Y, Nagata R, et al. Glutathione S-transferase A1 polymorphism as a risk factor for smoking-related type 2 diabetes among Japanese. Toxicol Lett 2008;178:143–5. [56] Savic-Radojevic A, Mimic-Oka J, Pljesa-Ercegovac M, et al. Glutathione S-transferase-P1 expression correlates with increased antioxidant capacity in transitional cell carcinoma of the urinary bladder. Eur Urol 2007;52:470 –7. [57] Rossi AM, Guarnieri C, Rovesti S, et al. Genetic polymorphisms influence variability in benzene metabolism in humans. Pharmacogenetics 1999;9:445–51. [58] Ahn J, Gammon MD, Santella RM, et al. Effects of glutathione S-transferase A1 (GSTA1) genotype and potential modifiers on breast cancer risk. Carcinogenesis 2006;27:1876 – 82. [59] Palma S, Cornetta T, Padua L, et al. Influence of glutathione S-transferase polymorphisms on genotoxic effects induced by tobacco smoke. Mutat Res 2007;633:1–12.