Short abstract
The transformation of ovarian surface epithelial cells, stromal cells, sex cord, or germ cells initiates ovarian malignancy. Epithelial ovarian cancer (EOC) is clinically silent with vague, non-specific symptomatology and is generally diagnosed at an advanced stage, resulting in a high mortality rate. The known main etiological factors are: age at menarche and menopause (early menarche or late menopause), use of oral contraception (estrogen and/or progesterone), family history, genetic factors, inflammation, occupational and environmental exposure. The study is intended to assess the association between blood organochlorine pesticide (OCP) levels and polymorphic status of phase I and phase II metabolizing enzymes (CYP1A1, GSTM1, and GSTT1) in the pathogenesis of epithelial ovarian cancer. The study included 200 subjects in total, of which 100 were epithelial ovarian cancer cases and 100 were controls. Estimation of blood organochlorine pesticide levels was carried out using gas chromatography and significantly high levels of beta-hexachlorocyclohexane (β-HCH), endosulfan-I, endosulfan-II, dichlorodiphenyltrichloroethane (pʹpʹ-DDT), dichlorodiphenyldichloroethylene (pʹpʹ-DDE) were observed in cases as compared to controls (P-value = 0.029, 0.042, 0.044, 0.039 and 0.037 respectively). For studying the polymorphism of CYP1A1, GSTM1/T1, PCR-RFLP, AS-PCR and multiplexing were performed and the frequency of null deletion of GSTM1/T1 was significantly higher in epithelial ovarian cancer cases. Regression model testing was also performed to check the interactive effect of organochlorine pesticide levels and polymorphic variant of genes keeping CA-125 as the dependent variable and observed a statistically significant role of genotypic/environmental interaction in epithelial ovarian cancer cases in the North Indian population.
Impact statement
Among pervasive environmental toxins, OCPs are one of the largest and most hazardous classes of contaminants in use around the world. Because these compounds possess the estrogen mimicking properties, the accumulation of these compounds in the human body may be a risk for several hormone-dependent diseases. EOC is hormonally dependent cancer and the mortality rate due to this disease is increasingly prevalent, and it has become imperative to explore the role of OCPs in the disease pathogenesis. The present study highlights the significant association of high OCPs level in the disease pathogenesis. It is also observed that the polymorphism in xenobiotic metabolism enzymes alters the predisposition of OCPs. The synergistic effect of gene polymorphism (CYP1A1, GSTM1, and GSTT1) and non-occupational exposure to OCPs was also assessed considering CA125 level as dependent variable with the risk of EOC and concluded that there exist a potential role of genotypic/environmental interaction in the etiology of EOC.
Keywords: Epithelial ovarian cancer, organochlorine pesticide, risk assessment, xenobiotic metabolizing enzymes, CYP1A1, GSTM1, GSTT1
Introduction
The transformation of ovarian surface epithelial cells, stromal cells, sex cord, or germ cells initiates ovarian malignancy. EOC is increasingly prevalent in pre-and post-menopausal women and is more prevalent than germinal or embryonal ovarian malignancies commonly observed in young age women.1 EOC is clinically silent with vague, non-specific symptomatology and is generally diagnosed at an advanced stage, resulting in a high mortality rate. The known main etiological factors are: age at menarche and menopause (early menarche or late menopause), use of oral contraception (estrogen and/or progesterone), family history, genetic factors, inflammation, occupational and environmental exposure. The numbers of man-made environmental toxins are increasing and many of them pose credible health risk. Among pervasive organic environmental toxins, pesticides are one of the largest and most hazardous classes of contaminants in use worldwide.2 It is estimated that around 85–90% of all the pesticides used in agriculture never reach their target organism, but instead are dispersed throughout the environment, affecting off-target species, including humans.3
Evidence indicates that the genes whose polymorphic variants are responsible for impaired ability to metabolize environmental carcinogens could affect an individual’s cancer susceptibility.4 Individual susceptibility to cancer due to environment toxin exposure may be influenced by polymorphism of metabolic susceptible enzymes gene families and identification of these risk modifier genes is critical to define high-risk genotypes.5 These enzymes are responsible for the metabolism of diverse range of xenobiotics, carcinogens, drugs, steroidal hormone, and toxins. Phase I includes Cytochrome P-450 (CYP450) system and Phase II includes glutathione S-transferase (GST).5–8
CYP450 is a diverse superfamily of monooxygenase hemoenzymes responsible for oxidative metabolism by oxidation of its numerous exogenous and endogenous substrates. CYP1A1 is one of the Phase I xenobiotic metabolizing enzymes amenable for the metabolism of various carcinogens and steroidal hormones including estradiol.8,9 Polymorphism in this gene may alter its activity and could affect the metabolism of carcinogens and steroidal hormones. In contrast, GSTs are the important phase II xenobiotic metabolizing enzymes involved in the conjugation of reactive chemical intermediates activated by phase I CYP450 enzymes and perform a significant role in detoxification and subsequently in the excretion of these compounds. GSTM1 and GSTT1 are two important genes in the GST family and are known to be polymorphically distributed in humans. These enzymes are involved in conjugation of electrophilic intermediates of xenobiotics and a variety of other substrates with glutathione, rendering the products more water-soluble and therefore, protect cellular macromolecules (DNA and protein) from free radical damage. Homozygous deletions in both of these genes are responsible for phenotypic absence of enzyme activity.10,11Variation/deletion amongst these genes may subject individuals to increased cancer susceptibility throughout their lifetime.4,12 A numbers of studies have analyzed the role of GSTM1 and GSTT1 genes in increased risk of numerous cancers including ovarian cancer but the results are inconsistent. In some populations/ethnic groups, deletions of these GSTs confer an increased risk, while in others no profound risk has been reported.4,12–15 Keeping in view the above facts, the present study is intended to assess correlation between gene polymorphism (CYP1A1, GSTM1, GSTT1), OCP levels in Epithelial Ovarian Cancer cases and controls. To estimate the impact of gene polymorphism and OCP levels upon CA-125 blood levels, linear regression modeling will be utilized.
Materials and methods
The study was conducted at Environmental Biochemistry and Molecular Biology Laboratory, Department of Biochemistry in collaboration with Department of Obstetrics and Gynaecology, University College of Medical Sciences associated with GTB Hospital (University of Delhi), Delhi, India. The study was conducted with the approval of Institutional Ethical Committee – Human Research (IEC-HR). In the present study, 100 histological/cytological confirmed cases of EOC with equal number of controls were recruited. For each case, an otherwise healthy woman with non-specific peri-menopausal complaints were taken as control and matched for age, BMI, and other factors. Normal bilateral ovaries were confirmed on trans-vaginal ultrasonography. To reduce the biases, case and control subjects were recruited from the same geographical/residential area.
A venous blood sample (3 mL) was collected in an EDTA vial and kept at 4°C; 1 mL was used for OCP estimation and the remainder of the sample was used for genomic analysis.
Estimation of organochlorine pesticides residue levels
Pesticide residue was extracted by the following methodology.16 A blood sample (1 mL) was placed in a 100 mL conical flask. Hexane and acetone were added in 2:1 ratio and the flask was kept on the mechanical shaker for 3 to 4 h at room temperature. The clear top hexane layer was collected in a clean flask and the same procedure was repeated twice. The cleanup of the collected solvent was done by USEPA method using florisil (Sigma) column. Sample was allowed to pass through the florisil column and 20 mL of hexane was further added for complete elution. The process was repeated thrice for remaining residues. The entire extract was transferred in a round bottom flask and the hexane was evaporated to concentrate the sample upto 1 mL for further analysis. Quantification of OCP residue was done by Perkin Elmer gas Chromatograph equipped with 63Ni Electron capture detector (Perkin Elmer Pvt. Ltd Singapore).
Genetic polymorphism study
DNA was extracted from whole blood using commercially available Quick gDNA isolation kit (Zymo research, USA). Agarose gel electrophoresis was done to check the quality of genomic DNA. Electrophoretic movement of genomic DNA was analyzed in 0.8% agarose gel at 100 V for 1 h using 0.5X TBE buffer (Tris borate EDTA buffer). For genetic polymorphism of CYP1A1 gene, two polymorphic sites CYP1A1m1 (T6235C) and CYP1A1m2 (−1462 V) were analyzed using PCR-RFLP and AS-PCR as described by Kumar et al.9 Genotyping of GSTM1/GSTT1 deletion was done by the method described by Sharma et al.17
Statistical analysis
Independent t-test was performed to compare the quantitative date and Chi square was used for qualitative data. Binary logistic regression was applied for comparing the genotypic variation of gene. Multiple linear regression was done for assessing the combined effect of genetic and environmental factor on disease risk.
Results
The two groups of the study were age matched which is reflected by the mean age of both the groups which were 45.28 ± 12.23 for cases and 45.64 ± 12.35 for control group. The other demographic factors which were compared (area of residence, socio economic status, education, religion, dietary habit) but none were found to be significantly associated except the source of drinking water. Cases and control patients recruited lived in similar residential areas and have similar lifestyle. Other potential risk factors considered to be relevant for EOC (parity, menstrual history, family history, BMI, and use of oral contraceptive pills) were significantly different between cases and control patients (Table 1). As CA-125 is an established diagnostic as well as prognostic marker for EOC, its level was also taken into consideration for cases and it was observed that serum CA-125 level was significantly elevated among cases with the mean value of 965.71 U/mL as compared to the reference range 0–35 U/mL. Significantly high levels of β-HCH, endosulfan-I, endosulfan-II, pʹpʹ-DDT, pʹpʹ-DDE and heptachlor were found in cases as compared to the controls with the P-values = 0.029, 0.042, 0.044, 0.039, 0.037, and 0.001, respectively (Table 2). While comparing the genotypic variation of CYP1A1 m1 and m2, no differences were observed. In contrast, the frequency of null deletion of GSTM1/GSTT1 was statistically significantly in EOC cases compared to controls (Table 3).
Table1.
Comparison of various socio-demographic features, risk/protective factors in the two groups.
| Characteristics | Cases (n=100) | Controls (n=100) | P-value |
|---|---|---|---|
| Age (in years) | 45.28±12.23 | 45.64±12.35 | 0.851 |
| Residential area | |||
| Urban | 88 (88%) | 98 (98%) | 0.674 |
| Rural | 12 (12%) | 2 (2%) | |
| Socioeconomic status | |||
| Lower | 0 (0%) | 0 (0%) | 0.668 |
| Middle | 91 (91%) | 93 (93%) | |
| Higher | 9 (9%) | 7 (7%) | |
| Occupation | |||
| Unemployed | 100 (100%) | 95 (95%) | 0.742 |
| Employed | 0 (0%) | 5 (5%) | |
| Education | |||
| Illiterate | 67 (67%) | 71 (71%) | 0.538 |
| Literate | 33 (33%) | 29 (29%) | |
| Religion | |||
| Hindu | 90 (90%) | 90 (90%) | 0.975 |
| Muslim | 10 (10%) | 10 (10%) | |
| Dietary habit | |||
| Vegetarian | 68 (68%) | 61 (61%) | 0.526 |
| Non-vegetarian | 32 (32%) | 39 (39%) | |
| Water resource | |||
| Govt. supply | 60 (60%) | 85 (85%) | 0.040* |
| Ground water | 40 (40%) | 15 (15%) | |
| Living style | |||
| Colony | 96 (96%) | 95 (95%) | 0.786 |
| Industrial area | 4 (4%) | 5 (5%) | |
| Parity | |||
| Nullipara | 9 (9%) | 9 (9%) | 0.987 |
| Multipara | 91 (91%) | 91 (91%) | |
| Menstrual history | |||
| Early menarche | 5 (5%) | 3 (3%) | 0.813 |
| Normal onset menarche | 95(95%) | 97 (97%) | |
| Family history | |||
| Present | 4 (4%) | 3 (3%) | 0.893 |
| Absent | 96 (96%) | 97 (97%) | |
| BMI (kg/m2) | 24.78±3.69 | 24.01±4.53 | 0.805 |
| Oral contraceptive intake | |||
| Taken | 5 (5%) | 10 (10%) | 0.600 |
| Not taken | 95 (95%) | 90 (90%) |
*P-value ≤ 0.05 is considered as significant.
Table 2.
Estimation of organochlorine pesticide (OCP) levels in cases and controls.
| OCPs (ng/mL) | Cases (n=100)mean ± S.D | Controls (n=100)mean ± S.D | P-value |
|---|---|---|---|
| α-HCH | 6.89±3.66 | 6.46±2.54 | 0.289 |
| β-HCH | 5.47±2.34 | 3.65±1.72 | 0.029* |
| γ-HCH | 4.44±2.32 | 4.31±2.55 | 0.356 |
| Dieldrin | 1.96±0.76 | 1.85±1.33 | 0.329 |
| Endosulfan-I | 2.01±1.52 | 1.52±0.91 | 0.044* |
| Endosulfan-II | 2.78±1.21 | 1.32±0.77 | 0.042* |
| pʹpʹ-DDT | 1.79±1.22 | 0.98±0.32 | 0.039* |
| pʹpʹ-DDE | 1.89±1.22 | 0.82±0.36 | 0.037* |
| Heptachlor | 5.99±2.41 | 3.10±1.21 | 0.001* |
*P-value ≤ 0.05 is considered as significant.
Table 3.
Genotypic distribution of CYP1A1 m1 and m2 and GSTM1/GSTT1 genotypes among EOC cases and controls.
| Genotypes | Cases(n=100) | Controls(n=100) | Odd ratio(OR) | 95% CI | P-value |
|---|---|---|---|---|---|
| CYP1A1m1 | |||||
| wt/wt | 26 (26%) | 34 (34%) | Ref. | ||
| wt/mt | 56 (56%) | 52 (52%) | 1.408 | 0.746–2.657 | 0.291 |
| mt/mt | 18 (18%) | 14 (14%) | 1.681 | 0.708–3.994 | 0.239 |
| CYP1A1m2 | |||||
| wt/wt | 27 (27%) | 35 (35%) | Ref. | ||
| wt/mt | 55 (55%) | 50 (50%) | 1.426 | 0.758–2.681 | 0.271 |
| mt/mt | 18 (18%) | 15 (15%) | 1.556 | 0.665–3.637 | 0.308 |
| GSTM1/GSTT1 | |||||
| GSTM1+/GSTT1+ | 33 (33%) | 47 (47%) | Ref. | ||
| GSTM1+/GSTT1− | 26 (26%) | 23 (23%) | 1.610 | 0.787–3.295 | 0.192 |
| GSTM1−/GSTT1+ | 22 (22%) | 21 (21%) | 1.492 | 0.708–3.144 | 0.293 |
| GSTM1−/GSTT1− | 19 (19%) | 9 (9%) | 3.007 | 1.211–7.466 | 0.018* |
Ref.: Reference, *P < 0.05 significant.
To assess the role of gene-environment interaction, multiple linear regression model testing was performed considering CA-125 level as the dependent variable and interaction between gene polymorphism and OCP levels was independent variables. While performing regression model testing, various confounding factors including age, family history, parity, and drinking water source were adjusted. B is unstandardized coefficient and this represents the mean CA125 difference. For this analysis, P < 0.005 was considered as statistically significant as the Bonferroni correction was performed due to multiple comparison tests (nine times) and the corrected P-value was 0.05/9 = 0.005. The calculation was done by assuming adjusted co-variants as constant: CA125 = B1 ×CYP1A1m1+ B2 × β-HCH + B3 × CYP1A1 × β-HCH, considering β-HCH = 1, CYP1A1m1 = 1; taking B1 = 26.107, B2 = 2.022 and B3 = −5.223 (Table 4). The interaction of CYP1A1m1 genotype and level of β-HCH among case group correlated with mean increase of 22.90 U/mL in CA-125 level. However, no significant increase in CA-125 was observed with CYP1A1m2 (Table 5). The interaction of GSTM1 genotype and level of β-HCH and DDT among cases group correlated with mean increase of 20.98 U/mL and 11.04 U/mL in CA-125 levels, respectively (Table 6). The interaction of GSTT1 genotype and level of β-HCH and DDT among case group correlated with mean increase of 22.12 U/mL and 11.05 U/mL in CA-125 levels, respectively (Table 7).
Table 4.
Regression model testing an interactive effect of OCPs (ppb) and CYP1A1m1 genotype on CA125 in EOC cases.
|
Blood sample of EOC cases |
||
|---|---|---|
| CYP1A1 m1 interaction | B-value | P-value |
| CYP1A1m1 (variants vs. wild type) | 3.459 | 0.627 |
| α-HCH | 0.353 | 0.695 |
| Interaction term of CYP1A1m1-α-HCH levels | −0.301 | 0.759 |
| CYP1A1m1 (variants vs. wild type) | 26.107 | 0.001 |
| β-HCH | 2.022 | 0.095 |
| Interaction term of CYP1A1m1-β-HCH levels | −5.223 | 0.001* |
| CYP1A1m1 (variants vs. wild type) | −5.868 | 0.234 |
| ϒ-HCH | −1.208 | 0.179 |
| Interaction term of CYP1A1m1-ϒ-HCH levels | 1.828 | 0.096 |
| CYP1A1m1 (variants vs. wild type) | −3.435 | 0.544 |
| Dieldrin | −3.562 | 0.172 |
| Interaction term of CYP1A1m1-dieldrin levels | 3.046 | 0.301 |
| CYP1A1m1 (variants vs. wild type) | −6.362 | 0.135 |
| Endosulfan-I | −0.917 | 0.546 |
| Interaction term of CYP1A1m1-Endosulfan-I levels | 3.671 | 0.057 |
| CYP1A1m1 (variants vs. wild type) | −7.594 | 0.335 |
| Endosulfan-II | −1.273 | 0.740 |
| Interaction term of CYP1A1m1-Endosulfan-II levels | 3.344 | 0.393 |
| CYP1A1m1 (variants vs. wild type) | −4.980 | 0.322 |
| pp-DDT | −1.696 | 0.610 |
| Interaction term of CYP1A1m1-ppDDT levels | 3.849 | 0.264 |
| CYP1A1m1 (variants vs. wild type) | −3.932 | 0.433 |
| ppDDE | 3.727 | 0.116 |
| Interaction term of CYP1A1m1-ppDDE levels | 1.395 | 0.608 |
| CYP1A1m1 (variants vs. wild type) | −7.039 | 0.246 |
| Heptachlor | −0.507 | 0.615 |
| Interaction term of CYP1A1m1-Heptachlor levels | 1.402 | 0.204 |
*P value <0.005 is significant; B: regression coefficient.
Table 5.
Regression model testing an interactive effect of OCPs (ppb) and CYP1A1m2 genotype on CA125 in EOC cases.
|
Blood sample of EOC cases |
||
|---|---|---|
| CYP1A1 m2 interaction | B-value | P-value |
| CYP1A1m2 (variants vs. wild type) | −1.00 | 0.988 |
| α-HCH | −0.220 | 0.800 |
| Interaction term of CYP1A1m2-α-HCH levels | 0.371 | 0.702 |
| CYP1A1m2 (variants vs. wild type) | −15.229 | 0.008 |
| β-HCH | −0.356 | 0.705 |
| Interaction term of CYP1A1m2-β-HCH levels | 3.127 | 0.005 |
| CYP1A1m2 (variants vs. wild type) | 1.031 | 0.825 |
| ϒ-HCH | −0.205 | 0.802 |
| Interaction term of CYP1A1m2-ϒ-HCH levels | 0.340 | 0.738 |
| CYP1A1m2 (variants vs. wild type) | −3.650 | 0.541 |
| Dieldrin | −3.793 | 0.158 |
| Interaction term of CYP1A1m2-dieldrin levels | 3.353 | 0.263 |
| CYP1A1m2 (variants vs. wild type) | −7.469 | 0.058 |
| Endosulfan-I | −1.529 | 0.253 |
| Interaction term of CYP1A1m2-Endosulfan-I levels | 4.862 | 0.006 |
| CYP1A1m2 (variants vs. wild type) | 0.205 | 0.961 |
| Endosulfan-II | 1.402 | 0.234 |
| Interaction term of CYP1A1m2-Endosulfan-II levels | 0.466 | 0.746 |
| CYP1A1m2 (variants vs. wild type) | 0.169 | 0.969 |
| pp-DDT | 1.002 | 0.703 |
| Interaction term of CYP1A1m2-ppDDT levels | 0.886 | 0.747 |
| CYP1A1m2 (variants vs. wild type) | −1.204 | 0.805 |
| ppDDE | 4.318 | 0.058 |
| Interaction term of CYP1A1m2-ppDDE levels | 0.374 | 0.887 |
| CYP1A1m2 (variants vs. wild type) | −1.652 | 0.751 |
| Heptachlor | 0.149 | 0.846 |
| Interaction term of CYP1A1m2-Heptachlor levels | 0.633 | 0.464 |
B: regression coefficient.
Table 6.
Regression model testing an interactive effect of OCPs (ppb) and GSTM1 genotype on CA125 in EOC cases.
|
Blood sample of EOC cases |
||
|---|---|---|
| GSTM1 interaction | B-value | P-value |
| GSTM1 (null vs. present type) | −3.692 | 0.432 |
| α-HCH | 0.419 | 0.329 |
| Interaction term of GSTM1-α-HCH levels | −0.737 | 0.248 |
| GSTM1 (null vs. present type) | 23.449 | 0.001 |
| β-HCH | 3.091 | 0.001 |
| Interaction term of GSTM1-β-HCH levels | −5.552 | 0.001* |
| GSTM1 (null vs. present type) | 1.159 | 0.784 |
| ϒ-HCH | 1.035 | 0.101 |
| Interaction term of GSTM1-ϒ-HCH levels | −2.172 | 0.011 |
| GSTM1 (null vs. present type) | 2.451 | 0.575 |
| Dieldrin | 1.794 | 0.221 |
| Interaction term of GSTM1-dieldrin levels | −5.648 | 0.006 |
| GSTM1 (null vs. present type) | 0.034 | 0.993 |
| Endosulfan-I | 2.333 | 0.060 |
| Interaction term of GSTM1-Endosulfan-I levels | −4.039 | 0.015 |
| GSTM1 (null vs. present type) | −1.644 | 0.697 |
| Endosulfan-II | 1.627 | 0.040 |
| Interaction term of GSTM1-Endosulfan-II levels | −2.276 | 0.101 |
| GSTM1 (null vs. present type) | 12.398 | 0.001 |
| pp-DDT | 8.833 | 0.001 |
| Interaction term of GSTM1-ppDDT levels | −10.187 | 0.001* |
| GSTM1 (null vs. present type) | 4.623 | 0.418 |
| ppDDE | 5.991 | 0.005 |
| Interaction term of GSTM1-ppDDE levels | −5.367 | 0.033 |
| GSTM1 (null vs. present type) | 0.735 | 0.871 |
| Heptachlor | 0.822 | 0.085 |
| Interaction term of GSTM1-Heptachlor levels | −1.516 | 0.026 |
*P value <0.005 is significant; B: regression coefficient.
Table 7.
Regression model testing an interactive effect of OCPs (ppb) and GSTT1 genotype on CA125 in EOC cases.
|
Blood sample of EOC cases |
||
|---|---|---|
| GSTT1 interaction | B-value | P-value |
| GSTT1 (null vs. present type) | −4.257 | 0.373 |
| α-HCH | 0.391 | 0.359 |
| Interaction term of GSTT1-α-HCH levels | −0.599 | 0.354 |
| GSTT1 (null vs. present type) | 24.871 | 0.001 |
| β-HCH | 3.127 | 0.001 |
| Interaction term of GSTT1-β-HCH levels | −5.875 | 0.001* |
| GSTT1 (null vs. present type) | 1.889 | 0.661 |
| ϒ-HCH | 0.753 | 0.214 |
| Interaction term of GSTT1-ϒ-HCH levels | −2.331 | 0.008 |
| GSTT1 (null vs. present type) | 3.173 | 0.468 |
| Dieldrin | 1.644 | 0.258 |
| Interaction term of GSTT1-dieldrin levels | −5.868 | 0.005 |
| GSTT1 (null vs. present type) | 0.877 | 0.822 |
| Endosulfan-I | 2.527 | 0.040 |
| Interaction term of GSTT1-Endosulfan-I levels | −4.311 | 0.010 |
| GSTT1 (null vs. present type) | 0.197 | 0.963 |
| Endosulfan-II | 1.497 | 0.052 |
| Interaction term of GSTT1-Endosulfan-II levels | −2.978 | 0.046 |
| GSTT1 (null vs. present type) | 12.433 | 0.001 |
| pp-DDT | 8.869 | 0.001 |
| Interaction term of GSTT1-ppDDT levels | −10.248 | 0.001* |
| GSTT1 (null vs. present type) | 5.376 | 0.321 |
| ppDDE | 6.256 | 0.002 |
| Interaction term of GSTT1-ppDDE levels | −5.677 | 0.020 |
| GSTT1 (null vs. present type) | 1.107 | 0.810 |
| Heptachlor | 0.711 | 0.131 |
| Interaction term of GSTT1-Heptachlor levels | −1.581 | 0.025 |
*P value <0.005 is significant; B: regression coefficient.
Discussion
It has been observed that the prevalence of EOC is significantly higher in that group who were using ground water as a direct source of their drinking water (Table 1) which is one of the possible routes of pesticide exposure to the population. This finding is in consistent with the other reports in which ground water of Delhi and nearby region was tested and found the presence of pesticide contaminant in significant level.16,18
The results show that there is significant high levels of pesticide (β-HCH, endosulfan-I, endosulfan-II, pʹpʹ-DDT, pʹpʹ-DDE and heptachlor) among cases group as compared to controls (Table 2) which is in consistent with one of our earlier published study in which high levels of pesticides were reported among EOC subjects.16 The possible reason for this finding may explain by the earlier study in which it has been found that there is maximum traceability of OCPs in water samples of Yamuna river and its canals in Delhi and nearby states. Moreover, these compounds are strongly lipophilic in nature with slow degradation rate and due to this property these become persistent in environment once entered.16,19,20
CYP450 and GST are xenobiotic metabolizing enzymes which have a key role in the vital detoxification process. These genes have reported to be polymorphically distributed amongst distinct human ethnic population. Functional polymorphisms affect enzymatic activity of the corresponding enzyme and can potentially alter the metabolism of harmful substances which may cumulatively promote disease throughout a person lifetime. Moreover, polymorphisms show ethnicity-dependent frequency distribution; therefore, risk of developing disorders/diseases in a population might depend upon frequencies of some of these high-risk alleles. There is growing evidence that among the risk factors that have been hypothesized to be associated with EOC, genetic predisposition plays a significant role. In Table 3, the genotypic variations were compared statistically. For CYP1A1 m1 and m2, the frequency of heterozygous and homozygous mutant type was found to be higher among EOC cases compared to that of controls. These observations are in agreement with earlier studies21,22 where the authors have mentioned about the risk of EOC due to the polymorphic variation in CYP1A1 gene but no statistically significant results were demonstrable when polymorphic variants of CYP1A1 m1 and m2 (CYP1A1 MspI and Ile/Val) were checked. In contrast, another study23 has shown a positive significant association between the polymorphic variant of CYP1A1 (Ile/Val) with increased risk of EOC. One of the possible reasons for this finding may be the ethnicity-related differences among the different population which may determine the susceptibility of an individual. As per the earlier studies, genes which are associated with metabolic detoxification pathway (CYP1A1, GSTM1 and GSTT1) are found to be associated as risk factors for EOC.24 Polymorphic variation in CYP genes may be associated with increased toxification process, whereas GST genes found to be associated with impaired detoxification process.25
While comparing the GST genotypes, it was observed that the frequency of GSTM1/GSTT1 null deletion was significantly higher among EOC cases and found to be associated with increased disease risk (Table 3). It is an enzyme family which is considered to play a crucial role in cellular protection and oxidative stress due to foreign toxic compounds.26 Null deletion of GTM1/T1 is an indicator of high risk for disease as null deletion signifies the phenotypic absence of enzyme activity or homozygous deletion of the whole gene. Null deletion of these genes will lead to the impaired catalysis of the conjugate of glutathione and detoxification process of organic peroxidases.27,28 Moreover, the presence of null deletion may impair the detoxification process for OCPs and may promote oxidative stress and DNA damage contributing to the pathogenesis of EOC.29,30
The present study identifies the high-risk allele of CYP1A1, GSTM1, and GSTT1 and their possible association with epithelial ovarian cancer. Moreover, this study also highlights that the presence of variant type of allele may promote the deposition of OCPs throughout the body, thereby promoting OCP-associated disease risks. Here we have taken EOC as our study model as it is one of the hormonally (estrogen) dependent cancers for which OCPs may contribute estrogen mimicking properties.
In the present study, regression model testing was performed to assess the role of gene–environment interaction in the etiology of EOC. For this model testing, CA125 which is a diagnostic as well as prognostic marker for EOC was taken as dependent variable and OCP levels and genotypic variation in genes were taken as independent variables. In other words, the effect on the level of CA125 was checked when variant type of gene is present and high level of OCP was detected.
A significant increase in CA125 level was observed when interaction term was assessed between β-HCH and CYP1A1m1 (Table 4). In other words, when CYP1A1m1 gene is polymorphic, increased level of β-HCH resulted in an estimated increase of 20.90 U/mL in CA125 levels. No significant interaction was observed with CYP1A1 m2 and pesticide levels (Table 5). Significant interaction is also observed between GSTM1 and GSTT1 null genotypes and increasing level of β-HCH and DDT resulting in the increase in CA-125 level. In association with presence of GSTM1 and GSTT1 null genotype and increased level of β-HCH and DDT, mean elevated in CA125 by 20.98 U/mL, 11.04 U/mL, 22.12 U/mL, and 11.05 U/mL, respectively was observed (Tables 6 and 7). These findings are supported by a previously published study in which it was observed that the presence of null genotypes for GSTM1 and T1 and increased levels of pesticides are associated with increased risk of bladder cancer.17 Other published studies have observed the null genotype of GSTM1 being associated with increased DNA adduct formation when exposure of carcinogens such as polycyclic aromatic hydrocarbons (PAHs) is present.31,32 The role of gene–environment interaction in the etiopathogenesis of various types of cancer including colon, oro-pharyngeal, and laryngeal cancer in other studies which have noted that the presence of variant type of GSTM1, GSTT1 and ADH (aldehyde dehydrogenase) may increase the risk of developing cancer when the exposure to certain environmental stimuli including PAHs or alcohol is present.33–35 The present study further supports that there is a plausible role of gene–environment interaction in the etiology of EOC. Moreover, further mechanistic-based studies using in vitro and/or animal models should be carried out to validate the present findings. The regression model testing is one of the relevant statistical tools to assess the role of genotype-environmental OCP interaction to date had not been carried out for EOC. The present study is first in its kind where the role of gene–environment interaction is explored for the etiology of EOC using statistical tool which can be used for drug targeting and therapeutic intervention.
Authors’ contributions
TS has carried out the experimental work, data interpretation and manuscript writing, GT was involved in experimentation, BDB and KG were involved in concept formation and editing of the manuscript, DM was involved in sample collection.
DECLARATION OF CONFLICTING INTERESTS
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
FUNDING
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: One of the authors (TS) is thankful to Indian Council of Medical research (ICMR), Government of India for awarding the fellowship support as Senior Research Fellow as well as Research Associate vide ref. no. 3/1/2/7 (Env)/2014-NCD-I to pursue this research work as a part of the PhD thesis under University of Delhi, Delhi, India.
References
- 1.Colombo N, Van Gorp T, Parma G, Amant F, Gatta G, Sessa C, Vergote I. Ovarian cancer. Crit Rev Oncol Hematol 2006; 60:159–79 [DOI] [PubMed] [Google Scholar]
- 2.Gurler AZ, Erdal G, Kizilaslan H. Future of global agriculture: an ecological evaluation. J Appl Sci 2005; 5:1762–6 [Google Scholar]
- 3.Repetto R, Baliga S. Pesticides and the immune system: the public health risk. Washington, DC: World Resources Institute, 1996 [PubMed] [Google Scholar]
- 4.Rebbeck TR. Molecular epidemiology of the human glutathione S-transferase genotypes GSTM1 and GSTT1 in cancer susceptibility. Cancer Epidemiol Biomarkers Prev 1997; 6:733–43 [PubMed] [Google Scholar]
- 5.Nebert DW, Russell DW. Clinical importance of the cytochromes P450. Lancet 2002; 360:1155–62 [DOI] [PubMed] [Google Scholar]
- 6.Guengerich FP. Cytochrome P-450 3A4: regulation and role in drug metabolism. Annu Rev Pharmacol Toxicol 1999; 39:1–17 [DOI] [PubMed] [Google Scholar]
- 7.Guengerich FP. Cytochromes P450, drugs, and diseases. Mol Interv 2003; 3:194–204 [DOI] [PubMed] [Google Scholar]
- 8.Gonzalez FJ. Role of gene knockout and transgenic mice in the study of xenobiotic metabolism. Drug Metab Rev 2003; 35:319–35 [DOI] [PubMed] [Google Scholar]
- 9.Kumar V, Singh S, Yadav CS, Ahmed RS, Gupta S, Pasha ST, Tripathi AK, Banerjee BD. CYP1A1 and CYP3A4 polymorphic variations in delhi population of Northern India. Environ Toxicol Pharmacol 2010; 29:126–30 [DOI] [PubMed] [Google Scholar]
- 10.Pemble S, Schroeder KR, Spencer SR, Meyer DJ, Hallier E, Bolt HM, Ketterer B, Taylor JB. Human glutathione S-transferase theta (GSTT1): cDNA cloning and the characterization of a genetic polymorphism. Biochem J 1994; 300:271–6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Seidegård J, Pero RW, Miller DG, Beattie EJ. A glutathione transferase in human leukocytes as a marker for the susceptibility to lung cancer. Carcinogenesis 1986; 7:751–3 [DOI] [PubMed] [Google Scholar]
- 12.Rebbeck TR, Walker AH, Jaffe JM, White DL, Wein AJ, Malkowicz SB. Glutathione S-transferase-mu (GSTM1) and -theta (GSTT1) genotypes in the etiology of prostate cancer. Cancer Epidemiol Biomarkers Prev 1999; 8:283–7 [PubMed] [Google Scholar]
- 13.Srivastava DS, Mandhani A, Mittal B, Mittal RD. Genetic polymorphism of glutathione S-transferase genes (GSTM1, GSTT1 and GSTP1) and susceptibility to prostate cancer in Northern India. BJU Int 2005; 95:170–3 [DOI] [PubMed] [Google Scholar]
- 14.Mallick S, Romana M, Blanchet P, Multigner L. GSTM1 and GSTT1 polymorphisms and the risk of prostate cancer in a Caribbean population of African descent. Urology 2007; 69:1165–9 [DOI] [PubMed] [Google Scholar]
- 15.Lima MM, Jr, Oliveira MN, Granja F, Trindade AC, De Castro Santos LE, Ward LS. Lack of association of GSTT1, GSTM1, GSTO1, GSTP1 and CYP1A1 polymorphisms for susceptibility and outcome in Brazilian prostate cancer patients. Folia Biol 2008; 54:102–8 [PubMed] [Google Scholar]
- 16.Sharma T, Banerjee BD, Mazumdar D, Tyagi V, Thakur G, Guleria K, Ahmed RS, Tripathi AK. Association of organochlorine pesticides and risk of epithelial ovarian cancer: a case control study. J Reprod Health Med 2015; 1:76–82 [Google Scholar]
- 17.Sharma T, Jain S, Verma A, Sharma N, Gupta S, Arora VK, Banerjee BD. Gene environment interaction in urinary bladder cancer with special reference to organochlorine pesticide: a case control study. Cancer Biomark 2013; 13:243–51 [DOI] [PubMed] [Google Scholar]
- 18.Lari SZ, Khan NA, Gandhi KN, Meshram TS, Thacker NP. Comparison of pesticide residue in surface water and ground water of agriculture intensive areas. J Environ Health Sci Engineer 2014; 12:11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Alavanja MC, Bonner MR. Pesticides and human cancers. Cancer Invest 2008; 23:700–11 [DOI] [PubMed] [Google Scholar]
- 20.Kaushik CP, Sharma HR, Jain S, Dawra J, Kaushik A. Pesticide residues in river Yamuna and its canals in Haryana and Delhi, India. Environ Monit Assess 2008; 144:329–40 [DOI] [PubMed] [Google Scholar]
- 21.Terry KL, Titus-Ernstoff L, Garner EO, Vitonis AF, Cramer DW. Interaction between CYP1A1 polymorphic variants and dietary exposures influencing ovarian cancer risk. Cancer Epidemiol Biomarkers Prev 2003; 12:187–90 [PubMed] [Google Scholar]
- 22.Holt SK, Rossing MA, Malone KE, Schwartz SM, Weiss NS, Chen C. Ovarian cancer risk and polymorphisms involved in estrogen catabolism. Cancer Epidemiol Biomarkers Prev 2007; 16:481–9 [DOI] [PubMed] [Google Scholar]
- 23.Heubner M, Wimberger P, Riemann K, Kasimir BS, Otterbach F, Kimmig R, Siffert W. The CYP1A1 Ile462Val polymorphism and platinum resistance of epithelial ovarian neoplasms. Oncol Res 2009; 18:343–7 [DOI] [PubMed] [Google Scholar]
- 24.Mrozikiewicz PM, Grześkowiak E, Seremak-Mrozikiewicz A, Bogacz A, Barlik M, Andrzej S, Bartkowiak-Wieczorek J, Krzysztof D. Importance of CYP1A1 polymorphism and its transcriptional regulation in ovarian and endometrial cancer. Ginekol Pol 2011; 82:925–32 [PubMed] [Google Scholar]
- 25.Zusterzeel PL, Visser W, Peters WH, Merkus HW, Nelen WL, Steegers EA. Polymorphism in the glutathione S-transferase P1 gene and risk for preeclampsia. Obstet Gynecol 2000; 96:50–4 [DOI] [PubMed] [Google Scholar]
- 26.Hayes JD, Strange RC. Glutathione S-transferase polymorphisms and their biological consequences. Pharmacology 2000; 61:154–66 [DOI] [PubMed] [Google Scholar]
- 27.Strange RC, Spiteri MA, Ramachandran S, Fryer AA. Glutathione-S-transferase family of enzymes. Mutat Res 2001; 482:21–6 [DOI] [PubMed] [Google Scholar]
- 28.Bessa SS, Ali EM, Hamdy SM. The role of glutathione S- transferase M1 and T1 gene polymorphisms and oxidative stress-related parameters in egyptian patients with essential hypertension. Eur J Intern Med 2009; 20:625–30 [DOI] [PubMed] [Google Scholar]
- 29.Beeghly A, Katsaros D, Chen H, Fracchioli S, Zhang Y, Massobrio M, Risch H, Jones B, Yu H. Glutathione S-transferase polymorphisms and ovarian cancer treatment and survival. Gynecol Oncol 2006; 100:330–7 [DOI] [PubMed] [Google Scholar]
- 30.Coughlin SS, Hall IJ. Glutathione S-transferase polymorphisms and risk of ovarian cancer: a HuGE review. Genet Med 2002; 4:250–7 [DOI] [PubMed] [Google Scholar]
- 31.Srám RJ, Binková B. Molecular epidemiology studies on occupational and environmental exposure to mutagens and carcinogens, 1997–1999. Environ Health Perspect 2000; 108: 57. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Binkova B, Chvatalova I, Lnenickova Z, Milcova A, Tulupova E, Farmer PB, Sram RJ. PAH–DNA adducts in environmentally exposed population in relation to metabolic and DNA repair gene polymorphisms. Mutat Res Fundament Mol Mech Mutagenesis 2007; 620:49–61 [DOI] [PubMed] [Google Scholar]
- 33.Hunter DJ. Gene–environment interactions in human diseases. Nat Rev Genet 2005; 6:287–98 [DOI] [PubMed] [Google Scholar]
- 34.Sinha R, Chow WH, Kulldorff M, Denobile J, Butler J, Garcia-Closas M, Weil R, Hoover RN, Rothman N. Well-done, grilled red meat increases the risk of colorectal adenomas. Cancer Res 1999; 59:4320–4 [PubMed] [Google Scholar]
- 35.Eriksson C. The role of acetaldehyde in the actions of alcohol (update 2000). Alcohol Clin Exp Res 2001; 25:l5–32 [DOI] [PubMed] [Google Scholar]