Genetic Polymorphism of Glutathione S-transferase and Cervical Cancer Susceptibility in Northeastern Thailand

Background: Exposure to certain carcinogens together with host genetic predisposition likely has an influence on cervical carcinogenesis. Objective: Our aim was to evaluate synergistic effects of glutathione S-transferase (GST) polymorphisms and risk behaviors (i.e., smoking and contraceptive use) on squamous cell cervical cancer (SCCA) development in northeastern Thailand. Methods: Subjects were 198 (SCCA) patients and 198 age-matched healthy controls. Multiplex PCR and PCR-RFLP were used to determine GSTT1 and GSTA1 gene polymorphism, respectively. Results: Interaction between the four polymorphic loci of GSTs (GSTM1, GSTT1, GSTP1 and GSTA1) and increased risk for cervical cancer was not observed. The three genotypes of GSTM1 consistently showed significant risks of smoking with a lower OR for the Null individuals (1.741) at around one -fourth of wild type homozygous individuals (8.000). The effects of GST polymorphisms on cervical cancer risk under the use of hormonal contraceptives apparently did not occur. Conclusions: The predisposition of smoking risk is related to the GST genotype. It is suggested that knowing one’s own genotype data will contribute to the prevention of SCCA by controlling risk habits.


Introduction
Cervical cancer remains the most common cancer and the leading cause of cancer-related deaths among women worldwide [1], including in Thailand [2]. Cervical cancer is usually considered to be caused by human papillomavirus (HPV) infection [3]. Nowadays, preventive HPV vaccines are available but these vaccines are effective in clearing only certain types of HPV infection. Other risk factors such as chemical substances and genetic backgrounds might be associated with risk of cervical cancer development. Tobacco smoke contains numerous carcinogens [4]; polycyclic aromatic hydrocarbons (PAHs), and volatile N-nitrosamines; all of which are considered to be primary carcinogens [5].
Smoking exposure is a leading cause of many types of cancer such as of lung and cervix [6][7]. Estrogencontaining oral contraceptives might affect carcinogenesis apjcb.waocp.com Wannapa Settheetham-Ishida, et al: Genetic Polymorphism of Glutathione S-transferase and Cervical Cancer Susceptibility between reduced glutathione (GSH) and substances (including environmental xenobiotics, carcinogens, chemotherapeutic agents or products of oxidative stress) [10]. These metabolizing enzymes are encoded by the glutathione-S-transferase genes (GSTs). Eight classes of cytosolic GST (corresponding gene) are recognized in mammalian species (viz., alpha (GSTA), mu (GSTM), theta (GSTT), pi (GSTP), sigma (GSTS), kappa (GSTK), omega (GSTO), and zeta (GSTZ) [12]. Allelic variants of the GSTs result in altered detoxification efficiency and are found to be polymorphic in many human populations [13]. It is thus agreeable to assume allelic variants of GSTs affect respective carrier susceptibility to various diseases including cancers [14]. Polymorphism of these genes results in alteration of enzymatic activity. For example, the TT genotype of the GSTA1 single nucleotide polymorphism (SNP) T69C had a lower enzymatic activity compared to the CC genotype [15]. A functional polymorphism of the GSTP1 at codon 105 contributes to the change of enzymatic stability and activity of the hydrophobic substrate binding site or electrophile-binding active site of the GST Pi peptide [16]. It has been reported that product of the Val allele has less conjugation capacity than that of the Ile allele [17]. GSTM1 and GSTT1 have a common and broad range of substrate specificities and as such they detoxify the reactive metabolites of benzo-a-pyrene and other polycyclic aromatic hydrocarbons [18]. A complete deletion of the GSTM1-which is frequently found in several populations-results in a null (homozygous) or partial loss (heterozygous) of enzymatic activity [19]. Several studies have reported that the homozygous deletion type of the GSTM1 is correlated with several types of cancer (i.e., colorectal, head and neck, lung, breast and prostate) [20]. A number of investigations have not, however, found any association between GSTM1 polymorphism and cancers [10][11]. The complete absence of enzymatic activity caused by a homozygous status (null genotype) of the GSTT1 deleted allele has been reported to increase the risk of prostate and colorectal cancers [21][22]. To contrast, the null genotype of the GSTT1 was not significantly associated with an elevated risk of gastric and colorectal cancer in the Korean population nor any cervical abnormality in a multi-ethnic population [23][24].
With the aim of preventing or reducing the incidence of squamous cell carcinomas of the cervix (SCCA) among Northeastern Thai women, we conducted several epidemiological studies and found that HPV infection, smoking habit, and sexual behavior were the major risks for the development of SCCA [25]. As for the potential co-risk factors, we focused on genetic predispositions related to the GSTs and performed genotyping of the GSTM1 and GSTP1 [26][27]. In the current study, we added the GSTT1 and GSTA1 genotype data and re-evaluated for any synergistic effects of the four GST polymorphisms and the risk behaviors (i.e., smoking and contraceptive use on the SCCA development).

DNA extraction and genotyping
The peripheral blood samples were taken and kept in EDTA tubes from both the patient and control groups. Genomic DNA extraction was performed using the GF-1 Blood DNA Extraction Kit (Vivantis, USA).
GSTT1 polymorphism was detected using the multiplex PCR method with the following two primer pairs: 5'-CAGTTGTGAGCCACCGTACCC-3', a n d 5 ' -C G ATA G T T G C T G G C C C C C T C -3 ' , 5'-CCAGCTCACCGGATCATGGCCAG-3' and 5'-CCTTCCTTACTGGTCCTCACATCTC-3'. The PCR mixture was performed using a total volume of 25 μl, containing 1 μl of DNA template, 12.5μl of KOD buffer, 5 μl of dNTP, 0.2 mM of each primer, and 0.5 unit of KOD FX (Toyobo, Japan). The cycle condition follows: an initial denaturation at 94°C for 2 min, followed by 30 cycles of denaturation at 98°C for 10 sec, and annealing and extension at 68°C for 90 sec. The PCR product (1460 bp and/or 466 bp) was analyzed by electrophoresis on 1.5% agarose gels.
The GSTA1 C69T polymorphism was determined using the PCR-RFLP method. The sequences for the primer pair were: 5'-GCA TCA GCT TGC CCT TCA-3' and 5'-AAA CGC TGT CAC CGT CCT G 3'. The PCR product of the 400-bp long was digested by EarI (New England, USA) and electrophoresed on 2.5 % agarose gel (Vivantis, USA), then visualized using ethidium bromide staining. The T allele generates a 308 bp and 92 bp fragment, while the C allele remains uncut.

Statistical analysis
The genotype distribution was tested using the Hardy-Weinberg equilibrium. The genotype and allele frequencies between the cases and controls were compared using the χ2 test with Yates' correction when required. The associations between selected variables and risks for SCCA were represented using the odds ratio (OR) with apjcb.waocp.com Wannapa Settheetham-Ishida, et al: Genetic Polymorphism of Glutathione S-transferase and Cervical Cancer Susceptibility associated with a significantly increased risk of cervical cancer among northeastern Thai women [6]. Accordingly, we evaluated the degree of risk of smoking (passive) vis-à-vis each genotype-based subpopulation to determine whether or not any genetic predisposition existed ( Table 3). The effects of GST polymorphisms on cervical cancer risk under the use of hormonal contraceptives apparently did not occur (Table 4); however, it should be noted that the individuals with the GSTT1 null genotype were found to show a trend of increase in cervical cancer risk among women taking oral contraceptives (adjusted OR: 4.25 (95%CI=1.00-18.01, P=0.050). We could not find any altered risk by the use of oral contraceptives in each genotype-based sub population (data not shown).

Discussion
The frequency of the GSTT1 null genotype (35.86%) was similar to previous reports among Asian population (30.0-41.4%) [29][30][31]. The finding in the current study of no relationship between the GSTT1 deletion type and risk of cervical cancer agreed with other reports among various Asian and European populations [28][29][30][31][32]. Few studies have reported a significant correlation between the GSTT1 null genotype and cervical cancer [33]; however, a 95% CI by using StataSE for the uni-and multi-variate logistic regression analyses. Differences were considered statistically significant when the P-value was < 0.05.

Results
Genotype distribution and allele frequency of each GST polymorphism (GSTT1, GSTA1, GSTM1, and GSTP1) are presented in Table 1. No significant deviation from the Hardy-Weinberg equilibrium in the genotype distribution among the controls was observed (P > 0.05 by χ2 test). There was no sign of altered risks for the development of cervical cancer except for the GSTP1 G allele; for which the presence resulted in a reduction of risk [33]. Interaction between the four polymorphic loci of GSTs (GSTT1, GSAT1, GSTM1, and GSTP1) and an increased risk for cervical cancer was not observed (data not shown).
Polymorphisms of GST, moreover, are associated with a higher risk of developing cervical cancer among smoking women [28]. In the present study, the effects of GST polymorphisms on cervical cancer risk vis-à-vis the sexual partner's tobacco smoking were inconclusive in a standard case-control comparison, P>0.05 (   inter-ethnic differences including behavioral risk factors might account for this discrepancy. Decreased expression of GSTA1 in individuals with the TT genotype results in the accumulation of oxidative DNA damage, suggesting an altered potential for carcinogenesis of the GSTA1 polymorphism [34]; this explanation is quite reasonable. However, many conflicting results have been reported even for the same types of cancer; significant and not significant associations for bladder cancer [35][36] and colorectal cancer [37]. We failed to show an association between GSTA1 polymorphism and SCCA; the low T allele frequency requires a larger sample size to draw a conclusion. Since cancer types and ethnic backgrounds might underlie genetic predispositions, there remains room for the contribution of GSTA1 polymorphism to SCCA development. The three genotypes of GSTM1 consistently showed significant risks of smoking with a lower OR for the Null individuals at around one -fourth of wild type homozygous individuals. Conflicting results of GSTM1 polymorphism and SCCA development [28] may be attributed to this fact. We simply expect that the lower enzyme activity results in the higher OR; however, in another deletion type polymorphism, GSTT1, OR for the individuals with Null genotype and with the wild type homozygous were similar to each other. This indicates the presence of possible intervention of compensatory effects by other GSTs. As for the GSTA1, the low T allele frequency may limit us to draw the conclusion with the present small sample size. In a study analyzing the same set of GST polymorphisms and colorectal cancer, the presence of GSTP1 Ile/Val (A/G) significantly decreased the risk of colorectal cancer [38], whereas the smoking risk was higher among heterozygous individuals than the Ile (A) homozygous individuals. It would be premature to conclude the presence or absence of genetic predisposition of GST polymorphisms in SCCA development.
A modified influence of the GSTT1 deletion on the association between combined (estrogen plus progestin) oral contraceptive use and sex hormone metabolism was demonstrated [39]. The use was also involved in cervical cell proliferation resulting in increased risks of cervical cancer [6]. The use of oral contraceptives possibly enhances HPV E6/E7 expression and thus results in HPV-related cervical cancer development [40].
In the current study, we investigated the possible role of the common GST polymorphisms in the development of cervical cancer in northeastern Thailand. The results showed that the genotypes GSTM1, GSTT1, GSTP1, and GSTA1 did not contribute independently towards cervical carcinogenesis; however, in the genotype-based sub population, the predisposition of smoking risk is related to the GST genotype. It is therefore suggested that knowing one's own genotype data will contribute to the prevention of cervical carcinoma by controlling risk habits.