Estrogenic microenvironment generated by organochlorine residues in adipose mammary tissue modulates biomarker expression in ERα-positive breast carcinomas

bcr1534 BCJ Research article Estrogenic microenvironment generated by organochlorine residues in adipose mammary tissue modulates biomarker expression in ERα-positive breast carcinomas Muñoz-de-Toro Mónica monicamt@fbcb.unl.edu.ar Durando Milena Beldoménico M Pablo Beldoménico R Horacio Kass Laura García R Silvia Luque H Enrique eluque@fbcb.unl.edu.ar

Laboratorio de Endocrinología y Tumores Hormonodependientes, School of Biochemistry and Biological Sciences, Universidad Nacional del Litoral, Santa Fe, Argentina

Faculty of Sciences, University of Liverpool, UK

Central Laboratory, Pesticide Division, School of Chemical Engineering, Universidad Nacional del Litoral, Santa Fe, Argentina

Breast Cancer Research 1465-5411 2006 8 4 R47 http://breast-cancer-research.com/content/8/4/R47 16859517 10.1186/bcr1534 13 3 2006 1 6 2006 8 6 2006 13 7 2006 19 07 2006 2006 Muñoz-de-Toro et al.; licensee BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Introduction

Breast cancer is the most frequent malignant disease in women. Exposure to estrogens throughout a woman's life is a risk factor for the development of breast cancer. Organochlorine compounds (OCCs), such as pesticides and polychlorinated biphenyls, are persistent lipophilic chemicals identified as endocrine disruptors, mainly with estrogenic effects. To test the hypothesis that the amount and quality of organochlorine residues in adipose tissue adjacent to breast carcinoma affect the biological behavior of the tumor, we studied biomarker expression in breast carcinoma and the OCC body burden in patients from an urban area adjacent to Paraná fluvial system, Argentina.

Methods

The studied patients were 55 women who had undergone excision biopsies of a breast lesion diagnosed as invasive breast carcinoma. Analysis of OCC residues in breast adipose tissue was conducted by electron-capture gas–liquid chromatography. Estrogen receptor alpha (ERα), progesterone receptor (PR) and proliferative activity (Ki-67) levels were measured in paraffin-embedded biopsies of breast tumors by immunohistochemistry.

Results

All patients had high levels of organochlorine pesticides in their breast adipose tissue. The most frequently detected compounds were p,p'-dichlorodiphenyldichloroethylene, hexachlorobenzene and β-hexachlorocyclohexane. When the whole sample was analyzed, no correlation between ERα or PR expression and OCC levels were found. In the subgroup of ERα-positive breast carcinoma patients, however, there was a positive correlation between PR expression (an estrogen-induced protein) in the neoplastic cells and OCC levels in adipose tissue surrounding the tumor. More significantly, all the ERα-positive breast carcinomas from postmenopausal women exhibited high proliferation when organochlorine levels in the surrounding adipose tissue reached levels higher than 2600 ppb. No associations were found between the organochlorine body burden and any other marker of tumor aggressiveness, such as node involvement or tumor size.

Conclusion

The present results support the hypothesis that organochlorine residues in adipose tissue adjacent to breast carcinoma generate an estrogenic microenvironment that may influence the biological behavior of the tumor through ERα activation and ERα-dependent proliferation. These findings may have therapeutic implications, since interference between organochlorine compounds and hormonal therapy could be expected to occur.

Introduction

Breast cancer is the most frequent malignant disease in women. Exposure to estrogens throughout a woman's life, including the period of intrauterine development, is a risk factor for the development of breast cancer 123. Epidemiological studies suggest a strong correlation between estrogen and/or xenoestrogen exposure and cancer, which is supported by numerous in vivo and in vitro experimental studies that attest to the proliferative and carcinogenic potency of estrogens 4.

Organochlorines present in the environment include agricultural pesticides and industrial compounds. Chemically stable and strongly lipophilic, organochlorine compounds (OCCs) have slow degradation rates and tend to bioaccumulate in lipid-rich tissues. Because some OCCs act as estrogen agonists they have been linked to an increased incidence of breast cancer 56, although not all data have been consistent 789.

People are exposed to OCCs from the environment and from residues in food. We recently reported that women from Santa Fe City area, a littoral region in Argentina, had high levels of organochlorine pesticides in their breast adipose tissue, and, more significantly, p,p'-dichlorodiphenyldichloroethylene (p,p'-DDE) was present in all sampled women and at high concentrations. The diet was a major factor that influenced the organochlorine pesticide body burden. Fish, particularly fatty varieties caught in local rivers, as well as other fat-rich meats were significant sources of exposure 10.

It is of interest, alongside designing studies to establish a link between xenoestrogens and breast cancer risk, to determine whether or not changes in the tumor microenvironment, due to organochlorine accumulation and release, affect the biological behavior of the tumor. In a nested case–control study, it has recently been demonstrated that patients with increased numbers of epithelial cells expressing estrogen receptor alpha (ERα) in normal lobular units are at increased risk of subsequent invasive breast cancer 11. These results suggest that overexpression of estrogen receptor in benign breast epithelium may increase estrogen sensitivity, thereby creating a permissive state that can lead to malignancy. As ERα mediates mammary cell differentiation and function, inappropriate exposure of cells to estrogens or xenoestrogens might alter the behavior of an established neoplasia. This might have therapeutic implications 12. Contradictory results have been reported in studies that addressed the possible relationship between organochlorine levels, breast cancer aggressiveness and prognosis 13141516. These differences may be due to methodological approaches since organochlorine levels were measured in different physiological compartments (adipose tissue and serum samples), which results in inappropriate comparisons 1718.

We hypothesized that the amount and quality of the organochlorine residues in adipose tissue adjacent to breast carcinoma might affect the biological behavior of the tumor. To test this hypothesis we conducted a cross-sectional study evaluating whether or not the amount and/or quality of organochlorine residues adjacent to breast carcinomas are associated with the expression of biological markers (paying special attention to hormone-dependent and hormone-independent breast cancer markers). We studied a population of women with breast lesions who reside in Santa Fe City and nearby locations, an urban area on the coast of Paraná River in a region that is characterized by intensive agricultural activities. This agricultural activity favors an elevated prevalence of environmental exposures, which our laboratory has recently confirmed 10.

Materials and methods

Study subjects

Participants included in this report were a subsample of subjects who were enrolled in a study to assess the OCC body burden among residents of Santa Fe City area in Argentina 10. This subsample (n = 55) comprised women that underwent an excision biopsy of a breast lesion diagnosed as invasive breast carcinoma between March 1998 and April 1999 in regional hospitals of Santa Fe City. All patients were urban residents at the time of surgery, were nonoccupationally exposed to persistent OCCs and were subjected to no therapeutic intervention before biopsy. An informed consent form was signed by all participants, 49 of whom completed a questionnaire.

The questionnaire provided information about sociodemographic factors, menopausal status, weight loss, height, parity, breast-feeding, dietary habits and hormone replacement therapy. Since OCC residue removal during lipolysis could be expected, periods of breast-feeding and weight loss before surgery were included in the questionnaire. The interview was conducted by a trained social worker by telephone and/or home visit. Histopathological reports and medical records were available for all patients. The primary tumor extent (T1–T4) and lymph node status were determined using the American Joint Committee on Cancer criteria 19. Approval for this study was obtained from the Universidad Nacional del Litoral Institutional Review Board (Santa Fe, Argentina).

Tissue sample collection

At the time of biopsy 0.5–1.0 g breast adipose tissue adjacent to the lesion was collected and weighed, was frozen at -70°C and was stored until organochlorine analysis. Paraffin blocks of breast tumor samples, previously fixed in 10% buffered formalin for 24 hours at room temperature, were available for immunohistochemical evaluation of tumor biomarkers.

Laboratory assays

Detection of organochlorine in adipose tissue

Breast adipose tissue was analyzed for the presence of the following compounds: hexachlorobenzene, hexachlorocyclohexane isomers (α-hexachlorocyclohexane, β-hexachlorocyclohexane and lindane), aldrin, oxychlordane, α-chlordane, γ-chlordane, heptachlor, heptachlor epoxide, dieldrin, endrin, mirex, metoxychlor, DDE (p,p'-DDE and o,p'-DDE), tetrachlorodiphenylethane (p,p'-tetrachlorodiphenylethane, o,p'-tetrachlorodiphenylethane), dichlorodiphenyltrichloroethane(p,p'-DDT, o,p'-DDT) and eight polychlorinated biphenyl congeners (International Union of Pure and Applied Chemistry numbers 077, 118, 128, 138, 170, 180, 187 and 195).

Concentrations of organochlorines were measured following the method previously described by Muñoz-de-Toro and colleagues 10. Briefly, an aliquot of 500 mg adipose tissue was dehydrated and then extracted in acetone–petroleum ether (25:75, v/v). The extract was filtered and the container rinsed with petroleum ether. All the petroleum ether fractions were collected and evaporated to dryness, and the percentage of fat in each tissue sample was calculated. Then, an aliquot of 200 mg fat from each sample was dissolved in 5 ml petroleum ether. Solid-phase extraction cleanup was performed using Supelclean LC-Alumina-N SPE tubes (Supelco, Bellefonte, PA, USA), and a second cleanup was performed using Supelclean Envi-Florisil SPE tubes (Supelco). Ten milliliters of ethylic ether–petroleum ether (15:85, v/v) was used to elute the chlorinated compounds from the microcartridge.

A gas–liquid chromatograph (Varian™ model 3400, Varian Inc., Palo Alto, CA, USA) equipped with a ⁶³Ni electron-capture detector was used for the analysis of residues. The gas chromatography (GC) columns used were: a Megabore Capillary GC Column DB 608 (30 m × 0.53 mm ID, film thickness 0.83 μm; J&W Scientific, Taejon, Korea), a GC Capillary Column CP-Sil 8 CB (50 m × 0.25 mm ID, film thickness 0.12 μm; VWR International B.V., Amsterdam, The Netherlands), and a GC packed column (2 m × 2 mm) coated with 1.5% SP 2250 + 1.95% SP 2401 on a Supelcoport 100–120 ID mesh (VWR International B.V., West Chester, PA, USA).

Organochlorine pesticide standards Pestanal (Honeywell Riedel-de Haën Fine Chemicals, Seelze, Germany) and polychlorinated biphenyl congener standards (Ultra Scientific, North Kingstown, RI, USA) were used. All the solvents used were of pesticide-grade quality (Merck, Darmstadt, Germany).

The OCC presence and level were confirmed using a GC–mass spectrometry system (VG Trio 2; VG Analytical, Manchester, UK) in randomly selected samples. The operational quality control and limits of detection have been reported previously 10.

Tumor biomarkers

ERα expression, progesterone receptor (PR) expression and Ki-67 expression were measured by immunohistochemistry as previously described 20. Samples received microwave pretreatment for antigen retrieval. Primary antibodies used were ERα (clone 6F-11) and PR (clone 1A6) from Novocastra Laboratories (Newcastle upon Tyne, UK) and Ki-67 (clone MIB-1; Amac, Westbrook, ME, USA), diluted 1:60, 1:20 and 1:100, respectively. Incubation with primary antibody was performed at 4°C for between 14 and 16 hours. Biotinylated anti-mouse IgG (Sigma, St Louis, MO, USA) was used as a second antibody; biotin was detected with the streptavidin–peroxidase (Sigma) complex. Diaminobenzidine tetrahydrochloride (Sigma) was used as a chromogen.

Biomarker expression was measured in the epithelial compartment. A proliferation index was obtained by considering the percentage of all Ki-67-positive cells independently of staining intensity (2,000 cells were counted/tissue section and at least three sections per sample were included). The Ki-67 assay (MIB-1) values were categorized according to Trihia and colleagues 21 as follows: low proliferation, ≤ 9.5%; intermediate proliferation, >9.5 and ≤ 15.5%; and high proliferation, >15.5%. ERα and PR were quantified by considering the percentage of positive cells, and the results were categorized into three groups 22: negative, 0%; weakly positive, >0 and <10%; and strongly positive, ≥ 10%.

Data analysis

The variables evaluated were organochlorine concentrations, age at biopsy (years), menopause status at biopsy (pre/post), age at menopause (years), years between menopause and surgery, years lived in a rural area, occurrence of weight loss (>3 kg) during the year before biopsy (yes/no), body mass index (kg/m²), number of children, months of lactation (averaged when several lactations and categorized into three groups: less than 3 months; from 3 to 6 months; and more than 6 months), animal fat intake (categorized into two groups: low, patients who ate lean red meat and processed meat (sausage, salami, etc.) infrequently; and high, those who ate red meat and processed meat regularly), freshwater fish intake (less than once a week, or at least once a week), and dairy product intake (low, infrequent; intermediate, once a week; or high, every day). Each food type was evaluated individually and then all food types were evaluated together as a score summation.

In addition, the variables tumor size (categorized as T1, 0–2 cm; 2 < T2 < 5 cm; T3, ≥ 5 cm; and T4, any size that invades muscle or skin), lymph node metastasis (none or present), and ERα, PR and MIB-1 expression (percentages and categories, described earlier) were evaluated. Expressions of ERα and PR were also dichotomized as positive and negative to conduct stratified analysis.

Nondetectable levels of OCCs were estimated as the mean between 0 and the detection limit. The OCC values were log-transformed when parametric tests were used. The central tendency of organochlorine residues was reported as geometric means.

All possible pairs of variables were tested for associations by appropriate statistical tests. Parametric tests were used when the assumptions were completely met, and nonparametric tests were used when assumptions were not met (Mann–Whitney tests to compare two groups, Kruskal–Wallis tests to compare three or more groups of a nominal variable, Jonckheere–Terpstra tests to compare three or more groups of an ordinal variable, and Spearman's Rho correlation to compare two continuous variables). Pearson and Mantel–Haenszel chi-square tests and Fisher's exact tests were used to compare proportions.

Results are reported for the pairs that were of interest (i.e. organochlorine residues and tumor biomarker expression), whereas the other pairs were used to assess the probability that an association of interest was caused by confounding or interaction effects of a third variable. Although multivariable analysis was not possible due to the limited sample size, stratification was conducted every time it was of interest to assess the associations within values of a third variable.

Results

Patient and tumor characteristics

Table 1 presents summary descriptive information on the study subjects. All tumors were classified as invasive carcinomas, and the tumor size varied widely (T1 = 30.2%, T2 = 52.8%, T3 = 5.7%, and T4 = 11.3% of our breast carcinoma patients). Node-positive patients (one or more axillary lymph nodes with neoplastic cells) comprised 46.9% of participants.

Table 1

Characteristics of patients studied

Continuous variables

Age (years)

60.5 ± 11. 6 (n = 51)

Body mass index (kg/m²)

25.4 ± 4.8 (n = 46)

Number of full-term pregnancies

2.2 ± 1.9 (n = 48)

Rural residence (years)

10.4 ± 17.8 (n = 45)

Categorical variables

Breastfeeding (n = 45)

<3 months total

35.6

>3 and <6 months total

11.1

>6 months total

53.3

Menopausal status (n = 49)

Premenopausal

24.5

Postmenopausal

75.5

Weight loss (>3 kg) during the year before biopsy (n = 42)

76.2

Yes

23.8

Freshwater fish consumption (n = 47)

Less than once a week

68.1

At least once a week

31.9

Dairy product intake (n = 46)

Infrequent

6.5

At least once a week

17.4

Every day

76.1

Animal fat intake (n = 46)

Low

34.8

High

65.2

Continuous variables presented as the mean ± standard deviation; categorical variables presented as the percentage.

Data on tumor biomarker expression are summarized in Table 2. According to hormone receptor expression, 50.0% of carcinomas appeared to be hormone-dependent tumors (ERα-positive and PR-positive) and 30.8% were hormone-independent tumors (ERα-negative and PR-negative). Breast carcinomas from premenopausal patients showed a lower percentage of ERα-positive cells (12.2 ± 19%) than those from postmenopausal patients (42.3 ± 35.5%; Mann–Whitney test, P = 0.021). ERα expression in breast tumors from node-positive patients (23.3 ± 32.4%) appeared to be lower than in tumors from node-negative women (43.6 ± 34.2%), but this difference was not statistically significant (Mann–Whitney test, P = 0.065). PR-negative breast carcinomas showed a higher proliferative index (33.4 ± 15.7%) than PR-positive carcinomas (22.0 ± 11.5%; Mann–Whitney test, P = 0.015).

Table 2

Biomarkers of hormone dependence and proliferative activity in breast carcinomas

Immunophenotype^a

Percentage of cases (n)

Hormone dependence

ERα-positive/PR-positive

50.0 (26)

ERα-positive/PR-negative

11.5 (6)

ERα-negative/PR-positive

7.7 (4)

ERα-negative/PR-negative

30.8 (16)

MIB-1 category

Low proliferation

12.8 (6)

Intermediate proliferation

17.0 (8)

High proliferation

70.2 (33)

^aCriteria adopted to establish cut-off values are described in Materials and methods. ERα, estrogen receptor alpha; PR, progesterone receptor.

There was a positive association between animal fat intake and tumor proliferative index (Fisher's exact test, P = 0.025): 69.7% of tumors with high proliferation were from patients with high fat consumptions. On the other hand, dairy product intake was positively correlated with ERα expression in carcinomas: 0%, 23.8% and 42.1% for low consumption, intermediate consumption and high consumption, respectively (Jonckheere–Terpstra test, P = 0.017).

Organochlorine levels in breast adipose tissue

A considerable high frequency of occurrence of organochlorine residues and elevated concentrations of organochlorine residues were found in breast adipose tissue of all participants (n = 55) (Table 3). The residues found most frequently were p,p'-DDE in all patients analyzed, hexachlorobenzene in 87.3%, β-hexachlorocyclohexane in 74.5%, heptachlor epoxide in 36.4%, oxychlordane in 32.7%, and heptachlor in 21.8%. p,p'-DDT was detected in three patients (5.5%). The concentrations of organochlorine residues in breast adipose tissue were very high, with maximum values of 4794 ppb for p,p'-DDE and 1780 ppb for β-hexachlorocyclohexane.

Table 3

Concentrations of organochlorine residues in breast adipose tissue adjacent to a malignant lesion

Compound

Frequency (%)

Geometric mean (95% confidence interval)

Range (ppb)

p,p'-Dichlorodiphenyldichloroethylene

100

541.9 (398.3–737.3)

20–4,794

Hexachlorobenzene

87.3

33.9 (22.2–51.6)

BDL-252

β-Hexachlorocyclohexane

74.5

81.7 (43.1–154.8)

BDL-1,780

Heptachlor epoxide

36.4

5.5 (3.8–7.9)

BDL-222.0

Oxychlordane

32.7

6.5 (4.0–10.5)

BDL-622.0

Heptachlor

21.8

2.8 (2.3–3.4)

BDL-24

p,p'-Dichlorodiphenyltrichloroethane

5.5

2.5 (1.9–3.1)

BDL-159

o,p'-Dichlorodiphenyldichloroethylene

1.8

2.2 (1.8–2.6)

BDL-255

Dieldrin

1.8

2.1 (1.9–2.3)

BDL-34

Total polychlorinated biphenyls

1.8

2.1 (1.9–2.4)

BDL-81

Total organochlorines

100

885.3 (646.6–1,212.1)

20.0–5,861

BDL, below detection limit.

Organochlorine levels and characteristics of the adjacent tumors

Analysis of all subjects found no associations between the organochlorine body burden and any marker of tumor aggressiveness such as node involvement, tumor size or proliferative activity. Moreover, the 12 patients who died within 4 years following surgery had organochlorine levels that did not differ from those of the surviving studied subjects (Mann–Whitney test, P = 0.303).

Regarding levels of organochlorine and the expression of biomarkers in adjacent breast tumors, concentrations of organochlorine appeared higher when surrounding hormone-dependent (ERα-positive and PR-positive) tumors (geometric means for total organochlorines in hormone-dependent tumors = 1169.2 ng/g versus hormone-independent tumors = 711.1 ng/g); however, this trend was not statistically significant (Mann–Whitney test, P = 0.379). Interestingly, in the subgroup of ERα-positive tumors, a significant positive correlation was observed between PR expression and total organochlorine levels (Spearman's Rho correlation, P = 0.044) (Figure 1), mainly due to p,p'-DDE levels since the strongest correlation was observed between PR expression and p,p'-DDE (Spearman's Rho correlation, P = 0.010).

Figure 1

Organochlorine levels modulate progesterone receptor expression in estrogen receptor-α positive breast carcinomas

Organochlorine levels modulate progesterone receptor expression in estrogen receptor-α positive breast carcinomas. The percentage of PR expression in ERα-positive breast carcinoma was positively correlated with organochlorine levels in the adjacent adipose tissue (Spearman's Rho test, P = 0.044).

The estrogenic effect of OCCs was also evident when tumor proliferative activity was evaluated in ERα-positive breast tumors from postmenopausal women. As shown in the subgroup of ERα-positive breast tumors from postmenopausal patients (Figure 2), all the adjacent tumors had high cellular proliferation when levels of OCC residues were 2600 ppb or higher (above the dashed line) (Fisher exact test, P = 0.017 – statistical analyses of tumors with low and intermediate proliferation were grouped). Furthermore, it is interesting to note the bimodal distribution of the OCC concentrations in the group of highly proliferative ERα-positive breast carcinomas from postmenopausal patients (Figure 2). The subgroup clustered above the 2600 ppb threshold were patients that had been menopausal for a long time (mean 31 years), while those with low total OCCs (<2600 ppb) and high proliferation had been in menopause for a shorter time (mean 17 years; Mann–Whitney test, P = 0.032). Moreover, when ERα-positive tumors from premenopausal patients were analyzed, all tumors with high proliferation had low total OCC levels (data not shown).

Figure 2

High organochlorine levels promote high proliferation in estrogen receptor-α positive breast carcinomas of postmenopausal

High organochlorine levels promote high proliferation in estrogen receptor-α positive breast carcinomas of postmenopausal. All ERα-positive breast carcinomas from postmenopausal women had high cellular proliferation when levels of organochlorine residues were 2600 ppb or higher (above the dashed line) (Fisher's exact test, P = 0.017 – statistical analysis of tumors with low and intermediate proliferation grouped).

In other words, patients that had been menopausal for a long time exhibited high proliferation if OCC levels surrounding the tumor were high, whereas highly proliferative ERα-positive breast carcinomas with a microenvironment of low OCC levels were seen either in premenopausal women or in women that went through menopause more recently.

Discussion

This study presents evidence regarding an estrogenic effect of OCCs on human mammary tumors that developed in a rich OCC environment. The PR expression in ERα-positive breast carcinomas was positively associated with OCC concentration in the adjacent adipose tissue, while in the subgroup of postmenopausal women, when low serum levels of endogenous estrogen are expected, high levels of OCCs in the adipose tissue surrounding breast tumor were associated with high proliferative activity. The present results support the hypothesis that organochlorine residues in adipose tissue adjacent to breast carcinoma generate an estrogenic microenvironment that may influence the biological behavior of the tumor.

We recently demonstrated that women living in an urban area of an agricultural and industrial region of Argentina had high levels of organochlorine residues in their breast adipose tissue 10. More significantly, p,p'-DDE was present in all subjects at high concentrations. In the present report, the study subjects were a subsample of women from our previous work 10 who were diagnosed with invasive breast carcinoma. As expected, these women also exhibited high levels of organochlorine residues in the adipose tissue adjacent to the breast tumors.

Epidemiologic studies have identified numerous prognostic parameters in breast cancer patients. Among these parameters, tumor size, cell proliferation, lymph node involvement and steroid receptor status are accepted as the most significant 23. Even though our sample size is small, our results agree with randomized clinical studies performed to provide prognostic information in patients with breast carcinoma 23. We have found that premenopausal and node-positive women showed a lower expression of ERα in their tumors, and that PR-negative carcinomas have a higher proliferative index. These observations agree with large well-designed clinical studies 23 and attest to the representativeness of our sample.

The relation of diet, especially fat intake, to recognized breast cancer prognostic indicators was investigated in humans as well in animal models. In rats, a high-fat diet significantly stimulated tumor proliferation and cell kinetics in chemically induced mammary carcinomas 24, while in women an increase in saturated fat intake was positively associated with tumor growth and/or spread 25. Consistent with this, our tumors from patients that reported high fat consumption exhibited high proliferative indices. In the present study, dairy product intake was positively correlated with tumor ERα expression. This result suggests that these dietary components could modify the tumor biological behavior, improving the response to endocrine therapy and/or survival in breast carcinoma patients.

Previous studies on estrogenic organochlorine levels and breast cancer have focused mainly on risk assessment while few of them have taken into account hormone receptor status and/or biomarkers of tumor aggressiveness. Dewailly and colleagues 26 and Liljegren and colleagues 27 found that patients with ERα-positive tumors had higher DDE and a higher polychlorinated biphenyl body burden than controls, but Muscat and colleagues 16 failed to confirm this finding. In the present study, a trend of higher concentrations of total organochlorines surrounding hormone-dependent (ERα-positive and PR-positive) tumors was apparent; this trend was not statistically significant, possibly due to the small sample size. Most organochlorine pesticides found in our samples were defined as xenoestrogens 28; consistent with this, a positive association was established in the ERα-positive breast tumor subgroup between PR expression and OCC levels in adjacent adipose tissue. This result suggests an upregulation of PR expression via an ERα-dependent mechanism.

DDT isomers and metabolites bind to and transcriptionally activate human ERα 29. Moreover, organochlorine has been shown estrogenic in MCF-7 cells by the expression of PR 29. Using a transgenic mouse model, Villa and colleagues 30 showed that PR mRNA was significantly induced by DDT isomers in tissues positive for estrogen receptors, and that the estrogenic action of p,p'-DDE was more efficient in organs that expressed ERβ, such as the mammary gland. Organochlorines might induce the observed changes in PR expression either directly by binding ERs or indirectly by increasing sensitivity to endogenous estrogens, or affecting the amount and type of estrogen metabolites with estrogenic properties. The local estrogen synthesis in neoplastic breast tissue depends on two principal pathways: the 'aromatase pathway', which transforms androgens into estrogens; and the 'sulfatase pathway', which converts estrone sulfate into estrone 3132. It is tempting to suggest that OCC could increase the local biosynthesis of estrogens through the modulation of the aforementioned enzymatic pathways.

Regarding the organochlorine body burden and the prognosis of breast carcinoma patients, adverse effects have been suggested by Demers and colleagues 13, Hoyer and colleagues 14 and Muscat and colleagues 16. In the present study, the markers of aggressiveness such as tumor size or node involvement show no associations with the organochlorine body burden, while the proliferative activity was associated with OCC levels when low endogenous estrogen serum levels are expected (postmenopause). An elevated cellular proliferation in breast carcinoma patients that have been menopausal for a long time depended on the levels of OCCs, whereas high cellular proliferation in those premenopausal patients or those that had more recently gone through menopause appeared to be independent of OCC concentrations. Our result agrees with results recently obtained in vitro, where proliferation of MCF-7 was induced by xenoestrogens when tumor cells were cultured in a medium devoid of estrogen 33. Our study has the characteristic that breast tumors in our series were diagnosed later than in developed countries, where active breast cancer screening programs work properly. Based on this observation, it is suggested that tumors in our series developed and progressed in an organochlorine-rich environment; following their natural history without intervention, thus providing more accurate information regarding the OCC effect on breast tumor biological behavior.

The present results highlight the importance of differentiating between tumor types, in terms of the estrogen receptor status and PR status, when analyzing the effects of both estrogens and xenoestrogens. The emerging picture is that epithelium and stromal compartments interact via a complex network of diffusible factors, resulting in the modulation of cell growth and differentiation, which facilitates tumor development and/or progression in a permissive, xenoestrogen-rich microenvironment. It is not known how the induction of cell proliferation and the increased expression of PR in tumors developed in an organochlorine-rich microenvironment would affect disease-related survival, as hormone-receptor-positive tumors may be associated with a more favorable prognosis compared with hormone-receptor-negative tumors 34.

In summary, the present results support the hypothesis that the amount and quality of the organochlorine residues in adipose tissue adjacent to breast carcinoma is responsible for generating an estrogenic microenvironment that may influence the tumor biological behavior. In the ERα-positive breast carcinoma, organochlorine levels in adipose tissue were associated with an increase in PR expression and cell proliferation, both mechanisms induced through estrogen action and ERα activation. While several in vitro works have shown the xenoestrogen effect on breast tumor cell lines 282933, present results suggest the occurrence of in vivo estrogenic effects of organochlorine residues on human breast tumors. Also, the results suggest that tumors of women at advanced menopause depend upon a microenvironment of high xenoestrogen levels to induce high cell proliferation. New studies with larger samples would be necessary to confirm this preliminary result.

Conclusion

Abbreviations

DDE = dichlorodiphenyldichloroethylene; DDT = dichlorodiphenyltrichloroethane; ERα = estrogen receptor alpha; GC = gas chromatography; OCC = organochlorine compound; PR = progesterone receptor; T1–T4 = tumor size.

Competing interests

The authors declare that they have no competing interests.

Authors' contributions

MM-d-T participated in designing the study, acquiring data from medical records and histopathological reports, analyzing and interpreting data, and outlining and reviewing the manuscript. MD participated in designing the study, searching the literature, acquiring data (inmunohistochemistry), analyzing data and drafting the manuscript. PMB participated in designing the study, analyzing data (statistical analysis) and reviewing the manuscript. HRB participated in designing the study, and analyzing and interpreting data from electron-capture gas–liquid chromatography. LK participated in designing the study, searching the literature, acquiring data (inmunohistochemistry), analyzing data and reviewing the manuscript. SRG carried out the gas–liquid chromatography, and analyzed and interpreted data. EHL participated in designing the study, acquiring study samples, analyzing and interpreting data, and preparing the manuscript. All authors read and approved the final manuscript.

Acknowledgements

The authors are grateful to Dr R Giardina for histological assessment and tumor classification. They specially thank SM Sanchez for outstanding efforts in identifying patients, encouraging their participation, and conducting the interviews and the patients that agreed to participate. This work was supported by grants from the Universidad Nacional del Litoral (CAI+D program), Roemmers Foundation and the Argentine National Agency for the Promotion of Science and Technology (ANPCyT). MD is a fellow and LK and EHL are Career Investigators of the Argentine National Council for Science and Technology (CONICET).

Evidence of prenatal influences on breast cancer risk Ekbom A Trichopoulos D Adami HO Hsieh CC Lan SJ Lancet 1992 340 1015 1018 10.1016/0140-6736(92)93019-J 1357410 Effect of twinship on incidence of cancer of the testis, breast, and other sites (Sweden) Braun MM Ahlbom A Floderus B Brinton LA Hoover RN Cancer Causes Control 1995 6 519 524 10.1007/BF00054160 8580300 Prenatal and perinatal risk factors for breast cancer in young women Weiss HA Potischman NA Brinton LA Brogan D Coates RJ Gammon MD Malone KE Schoenberg JB Epidemiology 1997 8 181 187 9229211 Is estradiol a genotoxic mutagenic carcinogen? Liehr JG Endocr Rev 2000 21 40 54 10.1210/er.21.1.40 10696569 Environmental organochlorine exposure as a potential etiologic factor in breast cancer Wolff MS Toniolo PG Environ Health Perspect 1995 103 Suppl 7 141 145 1518872 8593861 The link between the insecticide heptachlor epoxide, estradiol, and breast cancer Cassidy RA Natarajan S Vaughan GM Breast Cancer Res Treat 2005 90 55 64 10.1007/s10549-004-2755-0 15770527 Breast cancer risk in relation to adipose concentrations of organochlorine pesticides and polychlorinated biphenyls in Long Island, New York Stellman SD Djordjevic MV Britton JA Muscat JE Citron ML Kemeny M Busch E Gong L Cancer Epidemiol Biomarkers Prev 2000 9 1241 1249 11097233 Breast adipose tissue concentrations of polychlorinated biphenyls and other organochlorines and breast cancer risk Aronson KJ Miller AB Woolcott CG Sterns EE McCready DR Lickley LA Fish EB Hiraki GY Holloway C Ross T Cancer Epidemiol Biomarkers Prev 2000 9 55 63 10667464 Organochlorine exposures and breast cancer risk in New York City women Wolff MS Berkowitz GS Brower S Senie R Bleiweiss IJ Tartter P Pace B Roy N Wallenstein S Weston A Environ Res 2000 84 151 161 10.1006/enrs.2000.4075 11068929 Organochlorine levels in adipose tissue of women from a littoral region of Argentina Muñoz-de-Toro M Beldomenico HR Garcia SR Stoker C De Jesus JJ Beldomenico PM Ramos JG Luque EH Environ Res 2006 102 107 112 10.1016/j.envres.2005.12.017 16480710 Breast cancer risk associated with estrogen receptor expression in epithelial hyperplasia lacking atypia and adjacent lobular units Gobbi H Dupont WD Parl FF Schuyler PA Plummer WD Olson SJ Page DL Int J Cancer 2005 113 857 859 10.1002/ijc.20632 15499614 Association of hormone replacement therapy to estrogen and progesterone receptor status in invasive breast carcinoma Chen WY Hankinson SE Schnitt SJ Rosner BA Holmes MD Colditz GA Cancer 2004 101 1490 1500 10.1002/cncr.20499 15378477 Risk and aggressiveness of breast cancer in relation to plasma organochlorine concentrations Demers A Ayotte P Brisson J Dodin S Robert J Dewailly E Cancer Epidemiol Biomarkers Prev 2000 9 161 166 10698476 Organochlorine exposures influence on breast cancer risk and survival according to estrogen receptor status: a Danish cohort-nested case-control study Hoyer AP Jorgensen T Rank F Grandjean P BMC Cancer 2001 1 8 37543 11518544 10.1186/1471-2407-1-8 Organochlorines and breast cancer risk by receptor status, tumor size, and grade (Canada) Woolcott CG Aronson KJ Hanna WM SenGupta SK McCready DR Sterns EE Miller AB Cancer Causes Control 2001 12 395 404 10.1023/A:1011289905751 11545454 Adipose concentrations of organochlorine compounds and breast cancer recurrence in Long Island, New York Muscat JE Britton JA Djordjevic MV Citron ML Kemeny M Busch-Devereaux E Pittman B Stellman SD Cancer Epidemiol Biomarkers Prev 2003 12 1474 1478 14693740 Comparison of organochlorine pesticide and polychlorinated biphenyl residues in human breast adipose tissue and serum Archibeque-Engle SL Tessari JD Winn DT Keefe TJ Nett TM Zheng T J Toxicol Environ Health 1997 52 285 293 10.1080/009841097159584 9354175 Exposure of women to organochlorine pesticides in Southern Spain Botella B Crespo J Rivas A Cerrillo I Olea-Serrano MF Olea N Environ Res 2004 96 34 40 10.1016/j.envres.2003.10.001 15261782 American Joint Committee on Cancer AJCC Cancer Staging Manual Philadelphia Lippincott-Raven 1997 172 177 Proliferative activity and steroid hormone receptor status in male breast carcinoma Muñoz de Toro MM Maffini MV Kass L Luque EH J Steroid Biochem Mol Biol 1998 67 333 339 10.1016/S0960-0760(98)00124-1 9883990 Ki-67 expression in breast carcinoma: its association with grading systems, clinical parameters, and other prognostic factors – a surrogate marker? Trihia H Murray S Price K Gelber RD Golouh R Goldhirsch A Coates AS Collins J Castiglione-Gertsch M Gusterson BA Cancer 2003 97 1321 1331 10.1002/cncr.11188 12599241 Clinicopathologic characteristics of 143 patients with synchronous bilateral invasive breast carcinomas treated in a single institution Intra M Rotmensz N Viale G Mariani L Bonanni B Mastropasqua MG Galimberti V Gennari R Veronesi P Colleoni M Cancer 2004 101 905 912 10.1002/cncr.20452 15329896 Prognostic factors in breast cancer: current and new predictors of metastasis Hayes DF Isaacs C Stearns V J Mammary Gland Biol Neoplasia 2001 6 375 392 10.1023/A:1014778713034 12013528 Effects of switching from a high-fat diet to a low-fat diet on tumor proliferation and cell kinetics of 7,12-dimethylbenz(a)anthracene-induced mammary carcinoma in rats Noguchi M Ohta N Kitagawa H Earashi M Thomas M Miyazaki I Mizukami Y Oncology 1992 49 246 252 1495754 Dietary fat in relation to prognostic indicators in breast cancer Verreault R Brisson J Deschenes L Naud F Meyer F Belanger L J Natl Cancer Inst 1988 80 819 825 3392742 High organochlorine body burden in women with estrogen receptor-positive breast cancer Dewailly E Dodin S Verreault R Ayotte P Sauve L Morin J Brisson J J Natl Cancer Inst 1994 86 232 234 8283497 Case-control study on breast cancer and adipose tissue concentrations of congener specific polychlorinated biphenyls, DDE and hexachlorobenzene Liljegren G Hardell L Lindstrom G Dahl P Magnuson A Eur J Cancer Prev 1998 7 135 140 9818775 The E-SCREEN assay as a tool to identify estrogens: an update on estrogenic environmental pollutants Soto AM Sonnenschein C Chung KL Fernandez MF Olea N Serrano FO Environ Health Perspect 1995 103 Suppl 7 113 122 1518887 8593856 Transcriptional activation of the human estrogen receptor by DDT isomers and metabolites in yeast and MCF-7 cells Chen CW Hurd C Vorojeikina DP Arnold SF Notides AC Biochem Pharmacol 1997 53 1161 1172 10.1016/S0006-2952(96)00879-9 9175721 Target-specific action of organochlorine compounds in reproductive and nonreproductive tissues of estrogen-reporter male mice Villa R Bonetti E Penza ML Iacobello C Bugari G Bailo M Parolini O Apostoli P Caimi L Ciana P Toxicol Appl Pharmacol 2004 201 137 148 10.1016/j.taap.2004.05.007 15541753 Local estrogen biosynthesis in males and females Simpson E Rubin G Clyne C Robertson K O'Donnell L Davis S Jones M 1999 6 131 137 10731101 The selective estrogen enzyme modulators in breast cancer: a review Pasqualini JR Biochim Biophys Acta 2004 1654 123 143 15172700 Xenoestrogen action in breast cancer: impact on ER-dependent transcription and mitogenesis Hess-Wilson JK Boldison J Weaver KE Knudsen KE Breast Cancer Res Treat 2006 96 279 292 10.1007/s10549-005-9082-y 16328721 Harris JR Lippman ME Morrow M Osborne CK eds Diseases of the breast Philadelphia, PA: Lippincott Williams & Wilkins 2000