The GSTP1 105Val Allele Increases Breast Cancer Risk and Aggressiveness but Enhances Response to Cyclophosphamide Chemotherapy in North China

The glutathione-S-transferase (GST) family contributes to the inactivation of various toxic compounds formed as secondary metabolites during oxidative stress. GSTP1 accounts for the majority of the GST family enzymatic activity, and the activity of GSTP1 enzyme can be altered by the presence of the Ile105Val polymorphism. In this study, we examined the polymorphic frequency of GSTP1 Ile105Val genotype in 920 breast cancer patients and 783 healthy controls in women of North China. Results showed that GSTP1 105Val allele (Ile/Val and Val/Val) was associated with a higher breast cancer risk (OR = 1.38, 95% CI: 1.14–1.69; P = 0.001) and more aggressive tumors with histological grade III (OR = 1.15, 95% CI: 1.05–1.26; P = 0.001), lymph node metastases (OR = 2.35, 95% CI: 1.72–3.21; P < 0.001), as well as ER negative (OR = 1.77, 95% CI: 1.31–2.39; P < 0.001) than those carrying the Ile/Ile allele. However, the patients with the GSTP1 105Val genotype had a better disease free survival after cyclophosphamide (CTX)-based chemotherapy than those with Ile/Ile (HR = 0.77, 95% CI: 0.45–0.91; P < 0.001). Furthermore, in vitro cellular experiments demonstrated that breast cancer cells with the GSTP1 105Val allele were significantly more sensitive to CTX-induced proliferation inhibition. Thus, we conclude that the GSTP1 105Val allele increases breast cancer risk and aggressiveness and enhance response to CTX-based chemotherapy in women of North China. Detection of the GSTP1 Ile105Val genotype may help screen for high-risk populations and direct individualized therapy.


Introduction
Breast cancer is the most common malignancy and the second leading cause of cancer-related death in women worldwide [1]. In China, an increasingly modern lifestyle has been accompanied by a sharp jump in the breast cancer rate in urban areas. Clinical evidence has shown that women at the same pathologic stages of cancer undergoing the same treatment may have different outcomes [2]. In addition to the known risk factors of age, family history, age at childbirth, menopause and hormone therapy, the individual genetic variability also impacts drug metabolism and subsequent efficacy [3]. Chemotherapy has been established as the standard of care for breast cancer patients [4,5], especially those with locally advanced breast cancer. Among the chemotherapy regimens, CTX-based chemotherapy is recommended by the National Comprehensive Cancer Network (NCCN) as the clinical practice guidelines for breast cancer. Although chemotherapy improves disease-free survival (DFS) and overall survival (OS) of breast cancer patients [6], it is challenging to identify the patients who will benefit from chemotherapy and reduce the use of chemotherapy in those who will not benefit.
Glutathione-S-transferases (GSTs), a superfamily of dimeric phase II metabolic enzymes, are divided into six classes and play important roles in the metabolism of products of oxidative stress including by-products of lipid and DNA oxidation [7][8][9]. GSTP1 encodes the π-class of enzymes which accounts for approximately 90% of the enzymatic activity of the GST family, and its expression is found in many normal and malignant tissues [10]. The GSTP1 enzymatic activity can be altered by genetic polymorphisms. The GSTP1 Ile105Val (rs1695 or rs947894) single-nucleotide polymorphism (SNP) is a transition from an A to a G at nucleotide position 313 (A313G), leading to an Ile105Val amino acid change located near the substrate binding site of the enzyme [11]. The altered protein expression may lead to subsequent development of a malignant phenotype, whereas may enhance chemotherapy efficacy [11][12][13]. To date, researches have reported the correlation of GSTP1 Ile105Val polymorphisms with breast cancer risk and chemosensitivity [14,15] Several studies have reported that women with the GSTP1 105Val genotype in Shanghai of Southeast China [15,16], America [17], and India [18,19] have greater breast cancer risks. However, some studies have reported conflicting results in women from Italy [20] and Australia [21]. Thus, the role of GSTP1 in breast cancer is also controversial. In particular, the association of GSTP1 Ile105Val genetic polymorphisms with tumor aggressiveness and response to cyclophosphamide (CTX) and CTX-based chemotherapy remains unidentified.
Among individuals with similar GSTP1 expression levels in somatic cells, enzyme catalytic activity would be expected to vary according to the presence of variant GSTP1 genotypes. We speculate that the breast cancer tumors with the GSTP1 105Val variant genotype may have different biological characteristics and responses to CTX-based treatment because of altered enzymatic activity, which may ultimately lead to survival differences of patients. In current study, we examined the distribution frequencies of the GSTP1 Ile105Val genotype in breast cancer patients and age-matched healthy women of North China and evaluated the association of the genotypes with breast cancer risk, tumor aggressiveness and the survival of patients treated with a CTX-based regimen. Furthermore, we validated the differences of individuals with the GSTP1 105Ile and Val alleles in response to CTX cellular cytotoxicity through in vitro experiments.

Genotypic distribution of GSTP1 Ile105Val in breast cancer patients and control subjects
The alleles and genotypic frequencies of GSTP1 Ile105Val in the control population and patients are shown in Table 1

Association between GSTP1 Ile105Val genotypes and clinicopathological characteristics
To further characterize the significance of the GSTP1 Ile105Val genotypes in breast cancer, the associations with various clinicopathological characteristics including patient age, clinical staging, histopathological grading, as well as ER, PR and HER2 status, were analyzed. The results showed that the Ile/Val and Val/Val genotypes of GSTP1 Ile105Val significantly correlated with patient age, histological grade, lymph node involvement, and ER status. The tumors with the Val allele more frequently were histological grade III (OR = 1.15, 95% CI: 1.05-1.26; P = 0.001), ER negative (OR = 1.77, 95% CI: 1.31-2.39; P < 0.001), as well as involved lymph node metastases (OR = 2.35, 95% CI: 1.72-3.21; P <0.001) than tumors with the Ile/Ile allele. No further significant associations were observed between the SNP genotypes and other clinic pathological features (Table 2).

Association between the GSTP1 Ile105Val genotype and DFS in breast cancer patients
To analyze the relationship between the GSTP1 Ile105Val genotype of breast cancer patients and their prognosis after CTX-based chemotherapy, we compared the 5-year DFS rate between patients with Val allele and those with Ile/Ile genotype. The results indicated that the 5-year DFS rate of patients with the Val allele was higher than those with the Ile/Ile genotype (HR = 0.76, 95% CI: 0.61-0.93; P = 0.008; Table 3). Further Kaplan-Meier DFS analysis showed that the GSTP1 genotype was associated with DFS after analysis with the Cox proportional hazards model ( Figure 1).

Multivariate analysis for GSTP1 Ile105Val genotype and DFS in breast cancer patients
To determine whether the GSTP1 Ile105Val genotype is an independent factor associated with DFS in breast cancer patients, we performed multivariate analyses. Patient characteristics including the GSTP1 Ile105Val genotype, age, histological grade, clinical stage, lymph node involvement, ER status, PR status, and HER2 status were first evaluated with a univariate analysis. Only the variables with a P < 0.05 in the univariate analysis were included in the multivariate analysis using a backward stepwise Cox proportional hazards regression model (n = 788). Result showed that the GSTP1 Ile105Val genotype was an independent factor associated with the DFS of breast cancer patients (RR = 0.77, 95% CI: 0.45-0.91; P < 0.001; Table 4).

The GSTP1 Ile105Val polymorphism affects the breast cancer cell response to CTX
To investigate the effects of the GSTP1 Ile105Val genotype on breast cancer drug resistance in vitro, we analyzed the mRNA expression levels, protein levels and the GSTP1 Ile105Val genotype in different breast cancer cell lines. The results indicated that GSTP1 mRNA levels were high in T47D, MDA-MB-435, MDA-MB-231 and MDA-MB-468, but were barely detectable in MCF-7 cells (Figure 2A). The same tendency was found in their protein levels ( Figure 2B). The  Figure  2C). Furthermore, the GSTP1 non-expressing MCF-7 cells were transiently transfected with the GSTP1 105Ile/Ile and GSTP1 105 Val/Val GFP fusion plasmids, and than they were treated by 4-HC. The MTT cell proliferation assays confirmed that the GSTP1 105Val/Val genotype enhanced the sensitivity to 4-HC-induced proliferation inhibition compared to the GSTP1 105Ile/Ile genotype in MCF-7 transfected cells (P = 0.027; Figure 2D). The results indicate that GSTP1 Ile105Val affects breast cancer cell response to CTX in vivo.

Discussion
The genotypic distributions of GSTP1 Ile105Val were evaluated based on a large cohort of women with breast cancer and healthy populations in North China. In the healthy control group, the genotypic distributions were 66.3% of Ile/Ile, 29.3% of Ile/Val and 4.4% of Val/Val. The frequency of Val (Ile/Val and Val/Val) alleles (33.7%) is similar to other reports of a Chinese population in Shanghai in Southeast China (33.2%) [15] and Taiwan (33%) [22], but lower than that reported in Indian (54.0%) [19], Slovakian (51.8%) [23], European-American (58%) and African-American populations (65%) [22], and a little higher than that for Englishmen (28%) [11] and Italian (30%) [24] These data indicate that the genotypic distributions of GSTP1 Ile105Val in Chinese populations differ from those in Western and certain other Asian populations. It is known that the allelic frequencies of metabolic genes are not equally distributed throughout human populations, and the frequencies might follow diverse ethnic and/or geographicspecific patterns [23]. Genetic polymorphisms in genes coding metabolic enzymes have been thought to be related to breast cancer susceptibility [25]. In this study, we observed that the GSTP1 105Val (Ile/Val and Val/Val) allele carriers had a higher risk of breast cancer than those with the homozygous Ile/Ile(OR = 1.38). These results are consistent with results from studies of Americans [17], Indians [18,19], and Chinese in Shanghai in Southeast China [16]. However, in several studies on African-Americans [20], white women in North Carolina [26] and Caucasians [21], no significant differences were found between the GSTP1 Ile105Val polymorphism and breast cancer risk. Studies of the Finnish [27] and Koreans [14] showed that the GSTP1 105Val allele was associated with a lower risk of breast cancer. A recent meta-analysis showed that GSTP1 105Val was associated with an increased breast cancer risk in Chinese populations but not in non-Chinese populations [28], which is consistent with our results. The GSTP1 enzyme plays an important role in the metabolism and inactivation of various toxic compounds [29]. The Ile105Val polymorphism is a transition from an A to a G in nucleotide position 313 (A313G), leading to an Ile105Val amino acid change located near the substrate binding site of the enzyme. The altered protein expression may lead to subsequent accumulation of carcinogens in the body, resulting in the development of a malignant phenotype. The environmental pollution in developing countries such as China and India caused by increasing population levels and changes in modern lifestyle may contribute to the increased breast cancer risk in those with the GSTP1 105Val allele.
In addition, we found that breast cancer patients with the GSTP1 105Val allele were more likely to bear a tumor with histological grade III, lymph node metastases, as well as ER negative than those carrying the Ile/Ile allele. Our evidence indicates that the GSTP1 with the 105Val variant lost or reduced enzyme activity compared with Ile/Ile genotype leading to the accumulation of toxic substances in the body. The toxic damage to genomic DNA in somatic cells not only induces carcinogenesis [30] but also causes tumors with more aggressive characteristics such as poor differentiation, hormone-independent growth and metastatic potential.
The GSTP1 105Val genotype is an unfavorable factor for healthy females, however, it is a favorable factor for the cytotoxic efficacy of chemotherapy for breast cancer patients. Thus, the patients with the 105Val genotype may have better prognosis than those homozygous for Ile/Ile. The report in Shanghai revealed that breast cancer patients with the GSTP1 105Val allele had a 60% reduction in mortality risk after chemotherapy (HR = 0.4, 95% CI: 0.2-0.8) [30]. Our results further demonstrated that the GSTP1 105Val genotype provided a good prognosis for breast cancer patients in a Chinese population after receiving CTX-based chemotherapy (HR = 0.77, 95% CI: 0.45-0.91). However, this genotype was not associated with prognosis in breast cancer patients receiving CTX-containing chemotherapy in North American [31]. The GSTP1 enzyme exhibits specific and high activity in the conjugation of CTX and its toxic metabolites [32]. The SNP of GSTP1 Ile105Val substitutions in the coding sequence results amino acid changes within the GSTP1 substratebinding site [10,33]. Evidence has demonstrated that the GSTP1 105Val variant is associated with a lower thermal stability and altered catalytic activity to a variety of substrates compared with GSTP1 105Ile [30] and presents a reduced ability to detoxify chemotherapeutic agents, which results in lower clearance and better efficacy. We further investigated the effect of the GSTP1 Ile105Val genotype on breast cancer drug resistance through in vitro cellular experiments. The results confirmed that the breast cancer cells with the GSTP1 105Val/Val genotype exhibited increased sensitivity to 4-HC, which is an active derivative of CTX in vivo, than the cells with the Ile/Ile genotype. The results provide further evidence that the GSTP1 Ile105Val genotype affects the therapeutic response and survival of breast cancer patients treated with CTX.
In summary, our results demonstrated that among women in North China, the GSTP1 105Val allele carries a higher breast cancer risk and a risk of more aggressive tumors. However, patients with this allele have a trend toward improved survival after treatment with CTX-based chemotherapy. Therefore, the GSTP1 Ile105Val genotype could serve as a molecular test to screen for a high risk of breast cancer, to evaluate breast cancer aggressiveness and to predict the efficacy of CTXbased chemotherapy in Chinese populations. Due to the limited number of cases with rare genotype Val/Val of GSTP1 lle105Val in this study, the preventive, diagnostic and therapeutic values of the genetyping for women and breast cancer patients should be further evaluated and confirmed by large multicenter studies.

Materials and Methods Patients
A total of 920 breast cancer patients (aged 24-65 years, mean age 54.3 years) and 783 healthy women (aged 21-69 years, mean age 53.2 years) were recruited for this study. A ttest indicated that the mean age of these two population groups was equal (P = 0.120). All patients and control Table 3. The association between the GSTP1 Ile105Val genotype and 5-year DFS (n = 879).  Table 5. ER, PR and HER2 statuses in breast cancer tissues were determined through immunohistochemical staining. All patients received CTXbased chemotherapy for at least 4 cycles. CTX was administered through an intravenous injection line. The doses were within the standard range of 500-600 mg/m 2 . The imaging examination (ultrasound, X-ray, MRI, ECT and CT) and/or pathological diagnosis were performed to monitor for DFS status in the follow-up. DFS was defined as the time interval between primary surgery and any relapse (localregional, contra-lateral and/or distant), or terminal time of follow-up without any relapse events. 879 of 920 cases were followed-up with over five years. This study was approved by the Institutional Review Board of TMUCIH, and written consent was obtained from all participants.

Cell culture
The

Specimens and genomic DNA extraction
A volume of 2 mL of peripheral blood was collected from each individual and treated with EDTA-K2 anticoagulant. Nucleated cells were then separated by hypotonic lysis of red blood cells as described previously [34]. Genomic DNA from nucleated cells and cultured cells was extracted according to standard methods using proteinase K followed by phenol/ chloroform/isopropanol treatment or using QIAamp DNA Blood Mini Kit (Qiagen, Valencia, CA, USA). DNA concentrations were determined with a UV spectrophotometer. DNA integrity and purity were assayed through 1.5% agarose gel electrophoresis. TE buffer (10 mM Tris-HCl and 1.0 mM EDTA, pH 8.0) was used for resuspending DNA, with the final concentration adjusted to 200-500 ng/µL. The DNA solutions were frozen and stored at −80°C.

Genotyping by the TaqMan allelic discrimination assay
The primers and TaqMan probes for the genotypic analysis of GSTP1 Ile105Val were designed and optimized using Oligo 6.0 software (Molecular Biology Insights, West Cascade, USA), and synthesized by Sangon Biological Engineering Technology & Services Co, Ltd. (Shanghai, China). The PCR-TaqMan allelic discrimination assays were performed using the Platinum Quantitative PCR SuperMix-UDG System (Invitrogen) according to the manufacturer's instructions with the ABI 7500 TaqMan system (Applied Biosystems, Carlsbad, CA, USA).

DNA sequencing
To validate the data generated by PCR-TaqMan assay, 10% of the samples were randomly sequenced. The sequencing reactions were performed according to the conventional dideoxy chain-termination method using an ABI PRISMTM 3130 Genetic Analyzer (Applied Biosystem).

Construction and transfection of GSTP1 105 Ile and Val expression plasmids
The GSTP1 Ile105Val genotypes of the breast cancer cells MCF-7, T-47D, MDA-MB-231 and MDA-MB-435 were detected using the TaqMan allelic discrimination assay and were confirmed through DNA sequencing. PCR was performed to amplify GSTP1 105Ile (A/A) and Val (G/G) full-length cDNAs from cell lines with the corresponding genotype. The primers were 5'-CCAAGCTTACCATGCCGCCCTACACC-3' (forward) and 5'-CCGGATCCTGTTTCCCGTTGCCAT-3' (reverse), with BamHI and HindIII restriction endonuclease recognition sites (underlined) on the 5' ends. The PCR reactions were performed in a volume of 50 µL at 95°C for 2 min and 35 cycles of 95°C for 30s, 58°C for 1 min, and 72°C for 1 min, followed by 72°C for 10 min. The PCR products were subcloned into the pCR2.1 plasmids (Invitrogen) and expanded in DH5 alpha E. coli. Full-length cDNAs with GSTP1 105Ile and Val genotypes were digested from pCR2.1 using restriction endonucleases and subcloned into the NH2-terminus of green fluorescent protein (GFP) of the mammalian expression plasmid pEGFP-N1 (Clontech, Palo Alto, CA, USA). The GSTP1s sequences in the recombinant plasmids were confirmed through DNA sequencing, and GSTP1s expression levels were detected using reverse transcription-quantitative polymerase chain reaction (RT-QPCR) and western blot assays. For transient transfection, 2×10 5 MCF7 cells per well in 6-well plates were cultured without antibiotics overnight and then transfected with recombinant plasmids using Lipofectamine 2000 reagent (Invitrogen) according to the manufacturer's instructions.

RT-QPCR
Total RNA from cultured cells was extracted with TRIZOL reagent, reverse transcription (RT) was performed using the SuperScript First-Strand cDNAs Synthesis kit, and real-time quantitative PCR (QPCR) was performed using Platinum SYBR Green qPCR SuperMix-UDG. All reagents were Invitrogen products, and the reactions were performed according to the manufacturer's instructions. The primers for GSTP1 cDNA amplification were 5'-AGGACCTCCGCTGCAAATACATCT-3' and 5'-TCTCCCACAATGAAGGTCTTG-3'. The primers for the housekeeping gene glyceraldehyde 3-phosphate dehydrogenase (GAPDH) were as previously described [35]. QPCR was performed with the parameters of 50°C for 2 min, pre-denaturation at 95°C for 3 min, and 45 cycles at 95°C for 30 sec and 62°C for 1 min. Target gene expression quantification in samples was accomplished by measuring the fractional cycle number at which the amount of expression reached a fixed threshold (C T ). Triplicate C T values were averaged, and GAPDH C T was subtracted from GSTP1 C T to obtain ΔC T . The relative amount of GSTP1 mRNA was calculated as 2 -ΔCT .

Western blot
Cultured cells were solubilized with protein lysis buffer. The proteins were separated by size using SDS-PAGE and transferred to polyvinyldifluoride membranes (Pierce, Rockford, IL, USA). The membranes were blocked with 5% milk in TBST (10 mM Tris, 150 mM NaCl, 0.05% Tween 20, pH 8.3) for 60 min at room temperature and incubated with a 1:1000 dilution of rabbit polyclonal anti-GSTP1 antibody (Sigma, St. Louis, MO, USA) in TBST-milk overnight at 4 °C. Non-bound primary antibody was removed by washing in TBST, and bound antibody was detected using HRP-conjugated goat anti-rabbit IgG. The immunoreactive protein bands were visualized by enhanced chemiluminescence (ECL) reagents (GE Healthcare).

MTT assay
To assess the effect of the GSTP1 Ile105Val genotype on anti-cancer drug resistance, MDA-MB-435 (Ile/Ile), MCF-7 (Ile/ Val) and T47D (Val/Val) cells lines were assessed using the MTT assay. T47D cells (Val/Val), MDA-MB-435 cells (Ile/Ile) and MCF-7 cells (Ile/Val) were treated by 4-HC and analyzed the cell proliferation using MTT assay. MCF-7 cells, which do not express GSTP1, were transfected with pEGFP-GSTP1 105Ile (Ile/Ile), pEGFP-GSTP1 105Val (Val/Val), and the vector control pEGFP-N1, and then also were treated by 4-HC and analyzed using MTT assay. All the cells were plated at 1×10 4 cells per well in 96-well plates and incubated with 15 µg/mL 4-HC for 24, 48 and 72 h. Then, 10 µL of MTT (5 g/L in PBS) was added to each well and incubated at 37 °C for 4 h, and the medium was replaced by 100 µL DMSO to dissolve the formazan. Absorbance was measured at 570 nm using a spectrophotometer. Cell viability was calculated as the value relative to control cultures. The cells were tested in three independent assays with each containing triplicates

Statistical analyses
The Chi-square (χ 2 ) test or Fisher's exact test was used to compare the SNP genotypic distributions between the breast cancer group and the healthy controls, among cancer patients with various clinicopathological parameters, and for analysis of the Hardy-Weinberg equilibrium. Polytomous logistic regression was used to estimate the odds ratios (OR) and 95 percent confidence intervals (95% CI) as measures of association between the genotypes and breast cancer risk subtypes or to compare case subtypes to all controls. Survival analyses were performed according to the Kaplan and Meier methods and assessed using the log-rank test. All prognostic variables in the multivariate survival analysis were performed using a backward stepwise Cox proportional hazards regression model, and hazard ratio (HR) and relative risk (RR) were calculated from the Cox model. Statistical analyses were performed with the Statistical Package for the Social Sciences (SPSS, version 19.0). P-values less than 0.05 were considered statistically significant. All calculated data of experiments in vitro are expressed as mean ± standard deviation (SD).