Context: GSTP1 is a subgroup of glutathione-S-transferase family, which provides cellular protection against free radical and carcinogenic compounds due to its detoxifying function. Altered GSTP1 activity due to down regulation of enzyme activity and DNA methylation has been reported in many tumors, although data for ovarian cancer are few. In this study, we aimed at determining the expression of GSTP1 in relation to the methylation of the GSTP1 promoter in epithelial ovarian cancer (EOC). Materials and Methods: GSTP1 mRNA expression and GSTP1 enzyme concentration were assessed by quantitative reverse transcriptase polymerase chain reaction (PCR) and enzyme-linked immunosorbent assay, respectively, in 88 EOCs, 14 low malignant potential (LMP) tumors, and 20 benign tumors. The promoter methylation of GSTP1 gene was evaluated by methylation-specific PCR. Results: Reduced GSTP1 mRNA expression was observed in 49% EOCs, 21.4% LMP, and 45% benign tumors. Significantly lower levels of plasma GSTP1 were observed in all tumor samples compared to normal. GSTP1 promoter methylation was detected in 10 (11.4%) EOCs and 1 (7.3%) LMP tumors. No methylation was observed in benign tumors and normal ovaries. Conclusions: Our results show that there is a significant down regulation of GSTP1 expression while hypermethylation of the GSTP1 gene promoter is not very frequent in EOC. Further studies are needed to study underlying mechanisms leading to decreased expression.

INTRODUCTION

Ovarian cancer is the most lethal tumor of the female genital tract and the second most frequent gynecological cancer. [1] Most cases are diagnosed in the late stages, if diagnosed early may be curable. Although 65-80% of the patients respond to first-line chemotherapy, most patients will relapse with drug-resistant disease. [2] Current prognostic indicators neither accurately predict clinical outcome nor provide biological insight into the disease. Thus, a better understanding of the molecular mechanisms responsible for ovarian cancer development and progression is likely to aid the improvement of the diagnosis and treatment of the disease.

Glutathione-S-transferases (GSTs) (2.5.1.18) are a family of enzymes that detoxify intracellular xenobiotics, primarily by catalysis of the nucleophilic attack of reduced glutathione on electrophilic compounds. In humans, the cytosolic GSTs can be divided into seven major classes: α (A), μ (M), π (P), σ (S), θ (T), ω (O), and ζ (Z). GST Pi (GSTP1 ) is the predominant isoform present in normal human tissues, and its expression has significant biological and clinical implications including drug-resistance and carcinogenesis. The loss of GSTP1 expression is associated with promoter hypermethylation and has been identified in prostate, breast, liver, renal, urinary bladder, and endometrial cancers. [3]

Malignancies of the colon, stomach, urinary bladder, uterus, cervix, lung, and larynx often contain increased amounts of GSTP1 compared to adjacent normal tissue. [4] Three studies published in 1989 have all indicated a considerable elevation of serum GSTP1 levels in gastrointestinal malignancies. [5],[6],[7]] The results for ovarian cancer are not very conclusive.

There is a strong need for the development of biomarkers that can spot the disease at an early stage which in turn would improve the survival rates. [8] Hypermethylation is assumed to be an early event of carcinogenesis and a possible biomarker for cancer detection, prognosis and potential chemoprevention, and a therapeutic target. [9] Aberrant promoter hypermethylation associated with an epigenetically mediated gene silencing constitutes an alternative to coding region mutation for loss of gene function in cancer. [10] The methylation-specific polymerase chain reaction (PCR) (MSP) protocol designed by Herman and Baylin was able to assay methylation status of CpG islands within a gene promoter that correlates with loss of transcription. [10]

In this study, we aimed at evaluating the mRNA expression, plasma GSTP1 levels, and methylation status of GSTP1 promoter in ovarian carcinoma and its association with clinicopathological features.

MATERIALS AND METHODS

Sample collection

A total of 122 consecutive patients which included 88 epithelial ovarian carcinoma (EOC), 14 low malignant potential (LMP) tumors, and 20 benign tumors, who underwent potentially curative surgery between October 2010 and September 2012, were enrolled in this study. 15 normal ovarian tissues were collected at the time of oophorectomy from women without any family history of breast and ovarian cancer. The median age at diagnosis was 48 years (range: 23-72 years). No preoperative chemotherapy was given. All tumors were graded according to WHO criteria and staged according to the Federation of Gynecology and Obstetrics (FIGO) classification. 57 (65%) were grade 3 tumors and 64 (73%) cases had clinical stage III disease. Serous tumors were the most common histological subtype (52 [59%]).

Patients′ information included FIGO stage, histological grade, and subtype. Study approval was given by the Institutional Review Board and Medical Ethics Committee, and written informed consent was obtained from all participants.

Specimens removed from surgery were snap frozen at − 80°C until the extraction of DNA. Histology reports and slides as well as clinical data of these patients were reviewed by a Gynecological Oncologist and a Pathologist before the study. The patients′ plasma sample for the analysis of GSTP1 was collected in EDTA vacutainer. Immediately, after collection of the sample, the tubes were centrifuged for 15 min at 5000 rpm at room temperature. The upper two-third of the plasma was collected, and care was taken not to aspirate the platelets on top of the cell layer. Plasma samples were stored in aliquots at −20°C until the assay. The sampling, handling, storage, and assay procedures were identical for all patients and controls.

DNA and RNA extraction

Genomic DNA was extracted from 25 mg of ovarian tissue using QIAamp DNA mini kit (Qiagen, Valencia, CA, USA) following the manufacturer′s instructions. Total RNA was isolated from 30 mg of tissue preserved in RNA Later™ (Applied Biosystems, Foster city, CA, USA) using RNeasy mini kit (Qiagen, Valencia, CA, USA) following the manufacturer′s instructions. The quality and quantity of the extracted DNA and RNA were determined using Eppendorf Biospectrophotometer Kinetics™ and electrophoretically on a 2% agarose gel.

cDNA synthesis and quantitative reverse transcriptase-polymerase chain reaction

cDNA synthesis from total RNA was performed using the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA) following the company protocol. The gene expression levels of GSTP1 were quantified using TaqMan technology on a StepOnePlus™ real time PCR system (Applied Biosystem, Foster City CA, USA). Gene specific primers and probe of GSTP1 (assay ID Hs002512067_s1) were available as TaqMan gene expression assays (Applied Biosystems, Foster City, CA, USA). The 18 s ribosomal RNA (18S rRNA) (assay ID Hs99999901_s1) was used as an endogenous control in the quantification.

The real time PCR was performed in 20 μl of reaction volume containing 10 μl TaqMan gene expression master mix, 1 μl TaqMan gene expression assay, 2 μl cDNA and 7 μl of nuclease free water. The thermal cycling conditions were 50°C for 2 min, 95°C for 10 min followed by 50 cycles of 95°C for 15 s and 60°C for 1 min. A no template control was included in each reaction, and all reactions were performed in triplicate. Serial dilutions of cDNA from normal ovarian tissues were amplified in parallel as a control of amplification efficiency with each experiment and for the establishment of a standard curve for relative quantification. Expression of GSTP1 mRNA was normalized for 18S rRNA as an internal reference. Relative expression levels were calculated as GSTP1/18S rRNA in tumor and normal tissues, respectively, using delta delta Ct method (ΔΔC T ).

Plasma GSTP1 estimation

Plasma GSTP1 level was estimated in tumors and normal ovarian samples using an enzyme-linked immunosorbent assay (ELISA) kit procured from Cusabio (Cat No CSB-EL009989M0) following the manufacturer′s instructions. The sandwich ELISA used in the current study had a detection limit and an intra and inter assay coefficient of variation comparable to previously published immunoassays for GSTP1. All standards and samples were measured in duplicate.

Methylation-specific polymerase chain reaction and bisulfite sequencing

Sodium bisulfite conversion of unmethylated cytosine residues to uracil in genomic DNA samples was performed with 600 ng of DNA using the EZ DNA methylation kit (D5005, Zymoresearch Corp, CA, USA).

The promoter methylation status of GSTP1 was investigated by nested MSP. The primer sequences used for the first round PCR were 5′ GGGATTTTAGGGYGTTTTTTTG 3′ (forward) and 5′ ACCTCCRAACCTTATAAAAATAAT 3′ (reverse). The region chosen for GSTP1 spans the area of the greatest CpG density immediately 5 ′ to the transcription start site, in an area studied previously for methylation changes. The PCR reaction was performed with 2 μL bisulfite modified DNA template in 50 μL reaction mixture containing 1.5 mM/L MgCl 2 , 10 pM/L of each forward and reverse primer, 0.2 mM/L of each of the four dNTPs, 5 μL 10x PCR buffer and 1U Taq polymerase (New England Biolabs Inc., England). The primer sets used were designed to bind to both methylated and unmethylated DNA. The resulting 159 bp PCR product was diluted ten folds and subjected to the second round PCR with primer sets 5′TTCGGGGTGTAGCGGTCGTC3′ (forward) and 5′GCCCCAATACTAAATCACG ACG3′ (reverse) for methylated DNA and 5′GATGTTTGGGGTGTAGTGGTTGTT3′ (forward) and 5′CCACCCCAATACTAAATCACAACA3′ (reverse) for unmethylated DNA, respectively.

The PCR reaction was performed in duplicate in a Veriti™ thermal cycler (Applied Biosystems, Foster City, CA, USA) and was subjected to denaturation at 95°C for 5 min followed by 35 cycles of denaturation at 95°C for 30 s, annealing at 50°C for 30 s and extension at 72°C for 30 s and the final extension of 5 min at 72°C. The second step PCR was performed using an identical PCR program with the exception of annealing temperature, which was 53°C for methylated and unmethylated sequences. The resulting PCR products were 91 bp and 97 bp for methylated and unmethylated alleles, respectively. For each step of bisulfite modification and PCR, a positive and negative control for methylated and methylated DNA, CpGenome Universal Methylated DNA (Zymoresearch Corp, CA, USA) and peripheral blood lymphocyte DNA were used, respectively. Water with no DNA template was included as a control for possible contamination. 10 μL of each PCR reaction was loaded onto 2% agarose gel stained with 0.1% ethidium bromide and analyzed using a gel documentation system (G Box F3, Syngene) with a 100 bp DNA ladder (Fermentas, Germany) as a molecular weight standard.

The results obtained from MSP were validated using direct bisulfite sequencing. Four methylated, and two unmethylated PCR products were selected randomly and purified for direct sequencing using the Big Dye Terminator V1.1 sequencing kit cycler (Applied Biosystems, Foster City, CA, USA) and the ABI PRISM 310 Genetic Analyzer cycler (Applied Biosystems, Foster City, CA, USA). The PCR products were sequenced from both ends. The sequenced region included 12 CpG sites.

Statistics

Chi-square test or Fischer′s exact probability test was used to analyze methylation frequency and mRNA expression of GSTP1 gene. Student′s t-test was used to analyze plasma GSTP1 levels using SPSS software package (22.0 version, Chicago, IL, USA). Correlation with clinicopathological parameters was analyzed using the Chi-square test or Fischer′s exact probability test. All tests were carried out two sided. P <0.05 was considered statistically significant.

RESULTS

0mRNA expression of GSTP1 gene

Results of GSTP1 gene expression are given in [Table 1]. The median relative expression level in the control was used as the cut-off value. Reduced expression was seen in 49% EOCs, 21.4% LMP tumors, and 45% benign tumors whereas all normal ovarian tissues showed normal GSTP1 mRNA expression. GSTP1 expression was significantly lower in EOCs and benign tumors as compared to normal samples (P - 0.001 and P - 0.004, respectively). Reduced GSTP1 expression showed an association with the stage of the tumor and the presence of ascites (P - 0.001 and 0.09, respectively) (data not shown).{Table 1}

Plasma GSTP1

is a cytosolic enzyme and plasma levels of GSTP1 enzyme were assessed, and the results obtained are shown in [Table 2]. Ovarian tumors had lower plasma GSTP1 levels compared to normal subjects (P - 0.136, P - 0.152, and P - 0.0749 in EOCs, LMPs and benign tumors, respectively).{Table 2}

Promoter methylation of GSTP1 gene

[Table 3] shows the methylation frequency of GSTP1 gene. Hypermethylation of the promoter region of GSTP1 was detected in 10 of 88 (11.4%) EOCs and 1 of 14 (7.1%) LMP tumors. No hypermethylation was observed in benign and normal ovarian tissues. There was no significant correlation between GSTP1 methylation and clinicopathological characteristics of the patients. [Figure 1] shows the representative agarose gel pattern of MSP product of GSTP1 promoter region.{Figure 1}{Table 3}

The results obtained by MSP were validated by direct bisulfite sequencing, and the results were concordant. [Figure 2] shows the results obtained by direct bisulfite sequencing. In unmethylated MSP products, all cytosine nucleotides including those in the CpG islands are changed to thymine and in methylated MSP products, cytosine nucleotides in the CpG islands remain as cytosine.{Figure 2}

Correlation between GSTP1 promoter hypermethylation and GSTP1 expression

We analyzed the relationship between GSTP1 mRNA expression and methylation status in the study cohort. Of the 10 cases of EOCs which showed hypermethylation of the gene promoter, 6 cases (60%) showed reduced mRNA expression of the GSTP1 gene [Table 4]. However, since the numbers were little no statistical significance was observed (P - 0.516). Low levels of plasma GSTP1 also correlated with GSTP1 promoter methylation and the corresponding reduced mRNA expression in EOCs.{Table 4}

DISCUSSION

GSTP1 is a major member of the cytosolic GST super family expressed in several tissues and generally serves as a protector of cellular macromolecules from the damage caused by cytotoxic and carcinogenic agents that are thought to function in the xenobiotic metabolism. [11] Little is known about the mechanism of release of GSTP1 from tumor tissue. GSTP1 has several roles in tumorigenesis. First, GSTP1 might act as a tumor suppressor gene, which leads to tumor growth, when activated. Second, GSTP1 might act like a caretaker gene, which leads to somatic genome alterations that promote tumor growth when inactivated.

We studied the expression of GSTP1 by measuring the mRNA and plasma levels of GSTP1. The data obtained from relative gene expression through quantitative reverse transcriptase-PCR from our study, shows a significant reduction in mRNA expression levels in 49% EOCs and 45% benign tumors. We found significantly lower levels of plasma GSTP1 in EOC, LMP, and benign tumors. The down regulation of GSTP1 enzyme activity may be secondary to decrease GSTP1 mRNA level as almost 50% of EOCs, and benign tumors showed reduced mRNA expression. GSTP1 polymorphism has also been suggested to be involved in the alteration of GSTP1 enzyme activity. We postulate that reduced enzyme levels may play a role in the development and progression of cancer.

The postulation that GSTP1 reflects the sum of GSTP1 expression may be supported by the fact that GSTP1 concentrations in the culture media of various tumor cell lines correlate well with these of their representative cell lysates. [12] The release of GSTP1 from the cells may not be a simple leakage because GSTP1 in the medium is a monomer whereas the intracellular form is a homodimer.

Previously, GSTP1 expression in tissue and serum was suggested as a cancer marker in several earlier studies with results showing inconsistent expression patterns in various cancers particularly gastric cancers. Niitsu et al. and Fan et al. have both showed that the plasma GSTP1 levels were higher in patients with gastric cancer and were associated with advanced stage disease. [7],[13] Similar results have been reported for patients with nonsmall cell lung cancer and patients with oral cancer. [14],[15] Down regulation of GSTP1 with loss of GSTP1 mRNA expression was regarded as a phenotype associated with malignant transformation in prostate carcinoma, esophageal carcinoma, breast cancer, and hepatocellular cancer. [16],[17],[18],[19]

The results for ovarian cancer have been inconsistent. Some researchers found a relationship between over expression of GSTP1 in malignant ovarian tissue and poor prognosis or bad response to chemotherapy whereas others could not detect such an association. [20] Less attention has been paid to the analysis of GSTP1 in body fluids of patients with ovarian cancer. Kolwijck et al. have observed high levels of GSTP1 in ovarian cyst fluid in EOC patients compared to benign tumors which significantly correlated with clinicopathological variables. [21]

Gene silencing of tumor suppressor genes is an important mechanism in tumorigenesis. It is hypothesized that promoter methylation may play a role in gene silencing and has been studied in a variety of tumors. We examined the methylation status of the CpG island in the GSTP1 promoter region to find any association of methylation with reduced GSTP1 mRNA expression. Our results showed hypermethylation in 11.4% EOCs and 7.1% LMP tumors, while none of the benign tumors and normal ovarian tissues expressed any methylated alleles.

Of the 10 cases of EOC, which were hypermethylated at the promoter region, reduced mRNA expression was seen in 6 cases while the one methylated case of LMP tumor showed normal expression showing that hypermethylation of the gene promoter is associated with gene silencing. Due to few methylated cases, reduced GSTP1 mRNA expression was not found to be significant and hence methylation may not be a major regulator in GSTP1 gene silencing. Reduced mRNA expression due to promoter methylation of GSTP1 has been reported in prostate and other cancers. [22],[23] However, the finding of reduced mRNA expression with no demonstrable methylated alleles also suggest that along with promoter methylation there may be other predisposing factors such as histone deacetylation, nonCpG methylation, methylation of a broader chromosomal region, or loss of transcription factors necessary for the maintenance of GSTP1 expression. Compton et al. have suggested that down regulation may also happen in the transcription level due to various gene interactions. [24]

In our study, we also observed that the methylated tumor tissue always showed a heterogeneous methylation pattern with both methylated and unmethylated alleles alternating each other. There was no significant correlation between GSTP1 methylation and patient age, thus, methylation of GSTP1 in ovarian cancer is not attributable to the aging process.

There are not many published studies which have studied promoter methylation of GSTP1 in EOC and no reported data for the Indian population. A study by Wiley et al. in ovarian cancer did not report any methylation while Makarla et al. have reported the frequency of 9% which was mainly observed in serous ovarian carcinomas. [25],[26] Our results of 11.4% promoter methylation in EOCs are similar to the findings of Makarla et al. and serous tumors were the most common histological subtype (59%) in our study group also. [25]

In summary, our results indicate that there is a significant down regulation in the expression of GSTP1 in EOC and an imbalance between redox and GST enzyme which may further damage DNA and promote malignancy increasing the mutation rate. Further, studies are clearly warranted to explore the underlying mechanisms leading to decreased GSTP1 expression and to understand the complexity of cellular detoxification mechanisms.

ACKNOWLEDGEMENT

We are grateful to Dr. V Shanmugam (Research Assistant, NIMHANS) for helping with the statistical analysis.