- Email: [email protected]
Quantitative assessment of estrogenic activity in the water environment of Korea by the E-SCREEN assay
The Science of the Total Environment 263 Ž2000. 161᎐169
Quantitative assessment of estrogenic activity in the water environment of Korea by the E-SCREEN assay Seung-Min Oha , Se-Young Choung b, Yhun-Yhong Sheen c , Kyu-Hyuck Chung a,U a
College of Pharmacy, Sungkyunkwan Uni¨ ersity, 噛300, Chunchun-dong, Changan-gu, Suwon, Kyonggi-do 440-746, South Korea b College of Pharmacy, Kyoung Hee Uni¨ ersity, 噛1, Hoeki-dong, Dongdaemun-gu, Seoul 130-701, South Korea c College of Pharmacy, Ewha Womans Uni¨ ersity, 噛11-1, Daehyun-dong, Sudaemun-gu, Seoul 120-750, South Korea Received 23 February 2000; accepted 16 June 2000
Abstract In this study, the E-SCREEN assay was optimized and validated for the sensitive quantitative determination of the total estrogenicity in river samples. River water and sediment samples were collected and analyzed with the E-SCREEN. River water Ž10 l. was extracted using combined solid-phase extraction in static adsorption mode with Soxhlet extraction. Estrogenic pollutants adsorbed to the XAD-4 resin were recovered with 98.24" 5.90% efficiency by elution with ethyl acetate and dichloromethane Ž1:9.. The detection limit by 17-estradiol equivalent concentration ŽEEQ. of the E-SCREEN assay was 8.03 pg EEQrl. Among the water samples, the estrogenic activity was observed to be higher downstream of the Kumho river Ž7.43 ng EEQrl. and upstream of Kum river Ž2.05 ng EEQrl. than in other samples. More than 3 mg of equivalent sediment samples from the Kumho river, Kum river and Miho stream showed partial agonistic effects, and the Mankyung river showed a partial agonistic effect with only 1.5 mg of sediment. The highest value of RPE was 83.34 downstream of the Kumho river, and the lowest value of RPE was 6.52 downstream of the Miho stream. Full estrogen agonistic activities were observed downstream of the Kumho river and upstream of the Kum river. The partial agonistic activity was observed in upstream of the Kumho river, downstream of the Mankyung river, and upstream of the Miho stream, and no agonistic action was observed downstream of the Kum river or Miho stream, or upstream of the Mankyung river. The total estrogenic activity in the river water and sediment samples was between 0.50 pgrL and 7.4 ngrL, 3.39 pgrg and 10.70 pgrg. 䊚 2000 Elsevier Science B.V. All rights reserved. Keywords: River waters; Combined solid-phase extraction; E-SCREEN assay; MCF7-BUS cell; XAD-4; RPP; RPE; EEQ
U
Corresponding author. Tel.: q82-31-290-7714; fax: q82-31-292-8800. E-mail address: [email protected] ŽK. Chung.. 0048-9697r00r$ - see front matter 䊚 2000 Elsevier Science B.V. All rights reserved. PII: S 0 0 4 8 - 9 6 9 7 Ž 0 0 . 0 0 6 9 7 - 5
162
S.-M. Oh et al. r The Science of the Total En¨ ironment 263 (2000) 161᎐169
1. Introduction Organic contaminants, as either point source pollutants from incompletely treated sewage water or non-point source pollutants, such as pesticides, may be the source of endocrine disrupters in river water. Because of insufficient wastewater collection and treatment due to the shortage of facilities in Korea, river water may be contaminated with various chemicals including endocrine disrupters. Recently, since endocrine disrupters flowing into our environment become a big social issue, it is important to examine the estrogenic activity in our aquatic environment ŽColborn et al., 1993; Hileman, 1993, 1994; Soto et al., 1995; Waller et al., 1995.. However, most chemical analytical methods are not sufficient to identify all the environmental chemicals and to predict a combination of toxicity and bioavailability in biological organisms. Recent research has demonstrated that one of the most cost-effective ways to evaluate environmental pollution is to first use short-term, inexpensive bioassays to screen the samples for indications of toxic effects, and prioritizing the samples or sampled areas for chemical analyses andror more intensive studies. Thus, bioassays are becoming popular as screening methods for environmental compounds. Among these methods, cell culture bioassays are actively studied because they are rapid and sensitive in assessing the single andror combination effects of the chemicals on the ecosystem, as well as on humans. The E-SCREEN assay is a cell culture bioassay where the cell proliferation of human breast cancer MCF-7 cells in response to estrogen is measured, and it is a very sensitive assay by which one can measure single as well as multiple chemicals at the same time ŽSoto et al., 1995; Nagel et al., 1997; Blom et al., 1998.. Although cell proliferation assays may have their limitations ŽZacharewski, 1997., they can give quantitative estimates of test samples, so they can be quite applicable for environmental samples ŽKorner et al., 1999.. ¨ The Kumho, Kum, and Mankyung rivers, and the Miho stream are the representative rivers located in central Korea. Since the rivers flow through rural areas and several cities, they may
be polluted with various chemicals, including EDCs. Thus, in order to examine our aquatic environment, the estrogenic activities of these river water and sediment samples, previously isolated from water by an improved solid-phase extraction method with static mode, were assessed using the E-SCREEN assay.
2. Experimental methods 2.1. Examined ri¨ er waters and sample pre-treatment In May 1999, river water and sediments were collected from the upstream points and downstream points of the Kumho, Kum, and Mankyung rivers and the Miho river. Immediately after transport to laboratory, each river water sample Ž1 l. was acidified with concentrated sulfuric acid to pH 3. Active components were adsorbed for 24 h by stirring with 0.5 grl of Amberlite XAD-4 resin ŽFluka., based on the static adsorption mode of the combined solid-phase extraction method. Pollutants adsorbed to the resin were eluted with mixture of dichloromethane ŽDCM. and ethyl acetate Ž9:1. in a Soxhlet apparatus, and then dried under a reduced pressure. Residues were dissolved in 100% DMSO Ž625 l. and stored at y20⬚C until the assays were conducted. To prepare pore water samples, each sediment sample Ž10 g. was centrifuged at 5000 = g for 10 min, and the supernatant was taken and freeze-dried. The dried supernatant was redissolved in a methanolrwater mixture Ž1:1. and was purged with N2 gas, and finally dissolved in 1 ml of 100% DMSO. The sediments, after the removal of pore water, were dried for 24 h at room temperature. For the organic solvent extract, a sediment sample of 10 g of dried sediments was shaken in DCM for 4 h and filtered. The extract was dried under a reduced pressure and finally dissolved in 100% DMSO Ž500 l.. The water extract sample was extracted using deionized water instead of DCM. The samples that we used for the test were equivalent to 1.6 ml of river water, 10 mg of sediments, and pore water from 20 mg of sediments.
S.-M. Oh et al. r The Science of the Total En¨ ironment 263 (2000) 161᎐169
2.2. Cell culture MCF-7 BUS, human breast cancer, estrogen sensitive cells were kindly provided by Dr Soto at Tufts University. The cells were grown in 5% fetal bovine serum ŽHyclone, Logan, UT., Dulbeccos modified eagles medium ŽDMEM, Gibco, MD., under 5% CO 2 at 37⬚C. 2.3. CDFBS preparation In order to minimize the estrogenic activity from serum, FBS was stripped with 5% charcoal᎐0.5% dextran ŽOlea et al., 1996.. Charcoal Žacid washed, Sigma. was activated with cold sterile water immediately before use. A 5% charcoal᎐0.5% dextran T70 suspension was prepared. To this suspension, an equal volume of serum was added and maintained in suspension by rolling at 6 cyclesrmin at 37⬚C for 1 h. After centrifugation at 2000 = g for 50 min, the supernatant was taken and run through 0.45 and 0.2-m syringe filters ŽAxygen. and then stored at y20⬚C. 2.4. E-SCREEN assay The E-SCREEN assay using MCF-7-BUS cells was carried out according to the method of Perez et al. Ž1998.. The cells were harvested with 0.05% trypsin᎐0.53 mM EDTA⭈ 4Na ŽGibco. and resuspended in 5% FBS DMEM, and then plated onto a 24-well plate. The cell density was 10 000 cells per well and the culture was maintained at 37⬚C under 5% CO 2 for 24 h before the DMEM changed to phenol red containing free 10% CDFBS DMEM. The cells were exposed to various chemicals and test samples for 6 days, and then a sulforhodamine B ŽSRB. assay was conducted to measure cell proliferation. The cells were washed twice with phosphate buffered saline ŽPBS. before fixation with 10% TCA Ž500 l per well. at 4⬚C for 30 min. Each well was washed with sterile water 5᎐6 times, and to each dried well, 300 l of 1% acetic acid and 0.4% SRB was added and incubated for 15 min. The wells were washed 5᎐6 times with 1% acetic acid, and 1 ml of 10 mM trisma base ŽpH 10.5. was added into each dried well. The optical density was measured
163
at 490 nm using a microplate reader ŽMultiscan MCCr340 Pversion 2.3, Labsystems, Finland.. Each data point represented three independent experiments where three independent wells were used for the one experiment. 2.5. Quantitati¨ e measurement of estrogen action The relative proliferation effect ŽRPE., relative proliferation potency ŽRPP., and 17-estradiol equivalent concentration ŽEEQ. were calculated to estimate the estrogenic action of chemicals relative to 17-estradiol, which was the positive control. Based on the minimal concentration of 17-estradiol that resulted in the maximal cell proliferation effect, comparisons were made with the minimal concentration of test chemicals for the maximal cell proliferation effect to yield the RPP. The proliferative effect of test chemicals relative to positive control 17-estradiol was represented as RPE. Full agonistic activity was considered when the RPE was greater than 70, partial agonistic activity with RPE ranged from 25 to 70, and no effect when RPE was less than 25 ŽSoto et al., 1995.. As shown in Fig. 1, when bisphenol A and dieldrin were tested according to this method, the RPE of bisphenol A was 98.7 which showed that it was a full agonist and the RPE for dieldrin was 41.2, which indicated that it was a partial agonist. The RPP values of bisphenol
Fig. 1. Schematic representation of the dose᎐response curve to E 2 , bisphenol A and dieldrin. RPP: relative proliferative potency; RPE: relative proliferative effect s 100 = ŽPE ch emical y 1.rŽPE 2 y 1..
164
S.-M. Oh et al. r The Science of the Total En¨ ironment 263 (2000) 161᎐169
A and dieldrin were 0.01 and 0.001, respectively. This meant that bisphenol A and dieldrin needed 100 and 1000 times concentrations, respectively, to give the same effect as 17-estradiol. The EEQ was calculated as the concentration of 17estradiol that results in the same cell proliferation effect as the tested chemicals from the dose᎐response curve of the cell proliferation study with 17-estradiol as a positive control.
3. Results 3.1. Estrogenic acti¨ ity in ri¨ er water samples The water sample extracts were tested for MCF7-BUS cell proliferation, and the results are shown in Fig. 2, where estrogen effects were observed with all the samples. As shown in Fig. 3 and Fig. 4, the dose responsive effects of the Kumho and Kum river water samples were observed, respectively, indicating that this response caused by estrogenic chemicals in the water from both rivers. RPE values were calculated as shown in Table 1, where one can see estrogen action as well as the relative amounts of estrogenic chemicals in river water. The highest value
Fig. 2. Proliferative effect of MCF7-BUS cells grown in 24-well plates after 6 days incubation with river water samples Žtreated volume: 1.6 ml. in culture medium with 10% charcoal dextran treated fetal bovine serum. Values represent mean " S.D. Statistically different form the control Ž0.1% DMSO. group: U : P- 0.05; UU : P- 0.01.
of RPE was 83.34, downstream of the Kumho river, and the lowest value of RPE was 6.52, downstream of the Miho stream. Full estrogen agonistic activities were observed downstream of the Kumho river and upstream of the Kum river. The partial agonistic activity was observed up-
Table 1 Quantification of estrogenic effect of river water samples according to the E-SCREEN assay a Sampleb
PEc
RPE Ž%.d
Kumho
3.32" 0.16 6.15" 0.17 5.52" 0.29 1.70" 0.11 1.54" 0.15 2.92" 0.12 2.63" 0.08 1.40" 0.12 7.18" 0.15
37.57" 2.67 83.34" 4.83 73.06" 2.90 11.38" 0.77 8.79" 0.85 31.00" 1.98 26.43" 0.37 6.52" 0.77 100.00" 9.07
Upstream Downstream Kum Upstream Downstream Mankyung Upstream Downstream Miho Upstream Downstream 17-E2 Ž10y1 0 M. a
Values represent mean " S.D. Treated sample extracts are equivalent to 1.6 ml of river water. c The proliferative effect ŽPE. is the ratio of the highest cell number achieved with E 2 or the test compound, respectively, and the cell number of the negative control. d The relative proliferative effect ŽRPE. is calculated as 100 = ŽPE test compound y 1.rŽPE 17 -estradiol y 1.. b
Fig. 3. Concentration᎐response curves for estrogenic effect of River Kumho water by E-SCREEN assay. All the samples were dissolved in DMSO and 1 l of this solution was added to the cell culture medium Ž1 ml. for treatments. These were done in 24-well plates and lasted for 6 days. The data represent the means of three independent experiments. The coefficient of variation was - 10%.
S.-M. Oh et al. r The Science of the Total En¨ ironment 263 (2000) 161᎐169
165
3.2. Estrogenic acti¨ ity of sediment from ri¨ er sample
Fig. 4. Concentration᎐response curves for estrogenic effect of River Kum water by E-SCREEN assay. All the samples were dissolved in DMSO and 1 l of this solution was added to the cell culture medium Ž1 ml. for treatments. These were done in 24-well plates and lasted for 6 days. The data represent the means of three independent experiments. The coefficient of variation was - 10%.
stream of the Kumho river, downstream of the Mankyung river, and upstream of the Miho stream, and no agonistic action was observed downstream of the Kum river and the Miho streams, and upstream of the Mankyung river. It was interesting to note that estrogen agonistic activity was observed upstream of the Kum and Miho rivers, but no estrogenic activity was observed downstream of these rivers.
It was only possible to collect sediments from the downstreams of these rivers. The estrogenic effects of the water extracts, pore water and organic extracts of sediment are shown in Fig. 5. No estrogenic activity was observed in the pore water and water extract of the sediments, but highly induced estrogenic activity was observed in DCM extracts. From this result, estrogenic contaminants seem to be hydrophobic. The relative estrogenic effects of these chemicals are shown in Table 2. The DCM extracts showed partial estrogenic activity, whereas water extracts and pore waters associated with sediments showed no estrogenic effects. Fig. 6 shows the dose᎐response effects of DCM extracts from the river sediments. More than 3 mg of equivalent sediment samples from Kumho river, Kum river and Miho stream showed partial agonistic effects and, in the case of the Mankyung river, showed partial agonistic effects with only 1.5 mg of sediment. Thus, the Mankyung river sediments revealed approximately two times the estrogenic activity compared to the other river sediments. 3.3. Quantitati¨ e e¨ aluation based on EEQ In order to quantify the estrogenic activity of contaminants, the EEQ of each water sample was
Table 2 Quantification of estrogenic effect of river sediment and pore water samples according to the E-SCREEN assay a Sampleb
Kumho Kum Mankyung Miho 17 y E2 Ž10y10 M. a
Sediment DCM extracts c
Sediment water extracts
Pore water Ž20 mg-sediment.
PEd
RPE Ž%.e
PE
RPE Ž%.
PE
RPE Ž%.
3.24" 0.16 3.69" 0.29 3.82" 0.15 3.44" 0.08 ᎐
36.22" 1.63 43.45" 1.67 45.57" 1.68 39.51" 1.75 ᎐
1.52" 0.03 1.12" 0.02 1.38" 0.03 1.58" 0.06 7.18" 0.15
8.48" 0.17 2.00" 0.04 6.12" 0.14 9.31" 0.37 100.00" 9.07
1.14" 0.10 1.41" 0.13 1.16" 0.11 1.26" 0.12 ᎐
2.23" 0.10 6.59" 0.13 2.55" 0.11 4.19" 0.12 ᎐
Values represent mean " S.D. Treated sample extracts are equivalent to 10 mg of sediments, and pore water from 20 mg of sediments. c DCM: dichloromethane. d The proliferative effect ŽPE. is the ratio of the highest cell number achieved with E 2 or the test compound, respectively, and the cell number of the negative control. e The relative proliferative effect ŽRPE. is calculated as 100 = ŽPE test compound y 1.rŽPE 17yestradiol y 1.. b
166
S.-M. Oh et al. r The Science of the Total En¨ ironment 263 (2000) 161᎐169
Fig. 5. Proliferative effect of MCF7-BUS cells grown in 24-well plates after 6 days incubation with river sediment Žtreated volume: 10 mg-sediment. and pare water samples Žtreated volume: 20 mg sediment. in culture medium with 10% charcoal dextran fetal bovine serum. Values represent mean " S.D. Statistically different from the control Ž0.1% DMSO. group: U s P- 0.05; UU s P- 0.01; DCs dichloromethane extract; W s water extract.
calculated in comparison to 17-estradiol ŽTable 3.. The detection limit by EEQ calculation of the E-SCREEN assay was 8.03 pg EEQrl Ždata not shown.. Among the water samples, the estrogenic activity was observed to be higher downstream of the Kumho river Ž7.43 ng EEQrl. and upstream of the Kum river Ž2.05 ng EEQrl. than other samples. So far, there are few reports on the measurement of estrogenic activity in environmental samples. In this study, these high concentrations were close to the EEQ level Ž7.8᎐25 ng EEQrl. of municipal sewage plant effluents in Germany ŽKorner et al., 1999.; whereas upstream ¨ of the Kumho and Miho rivers and downstream of the Mankyung river were slightly contaminated ŽEEQ was in approx. picogram range..
4. Discussion MCF7-BUS human breast cancer cells containing large amounts of estrogen receptors are easy to culture and possess stable biological responses to estrogen compared to other human cell lines. Recently, Soto et al. Ž1995. demonstrated that, in spite of substantial modifications and simplifications, the proliferative response of MCF7-BUS
cells in relation to controls with and without 17-estradiol ŽE 2 . was a simple, reliable and sensitive screening-system for the detection and quantification of direct receptor-mediated estrogenic activity of single chemicals. In this study, we used 24-well plates for the E-SCREEN assay, and the detection limit of the E-SCREEN assay for the natural estrogen 17-estradiol was at 1 fM Ž0.27 fgrml. Ždata not shown.. The IC50 value for 17-estradiol was between 1 and 5 nM; i.e. two to three orders of magnitude lower than the physiological serum levels of E 2 in women. The quantitative results for natural estrogen, 17-estradiol tested by E-SCREEN with MCF7-BUS was in good agreement with those reported by other groups ŽSonnenschein et al., 1995; Soto et al., 1995; Villalobos et al., 1995; Korner et al., 1999.. ¨ Therefore, the E-SCREEN assay is suitable to establish 17-estradiol equivalency factors for environmental estrogens to rank their estrogenic potency relative to the natural ligand 17estradiol ŽKorner et al., 1999.. The most impor¨ tant prerequisite for applying the concept of estradiol equivalency factors to environmental samples is the additive behavior of the estrogenic activity of single substances in mixture. Korner et ¨ al. Ž1999., in their experiments with two different mixtures each containing three xenoestrogenic responses of the single compounds, thus demonstrated that the estrogen equivalency factor concept could be applied to environmental samples. Soto et al. Ž1994. also found additive behavior of various proliferation induced by non-steroidal xenoestrogens in estrogen receptor-positive MCF7 breast cancer cells, based on a common estrogen receptor-mediated mechanism. The different classes of substances such as steroids, alkylphenols, bisphenols, phthalates, and chlorinated hydrocarbons known to act like estrogen, require extraction methods which enable the simultaneous quantitative extraction of all these compounds from various environmental sources. For the extraction of estrogenic compounds, we found out that static adsorption was much better than the dynamic adsorption that is common method. We had much higher recovery Ž98.24" 5.90%. with XAD-4 than with XAD-2 Ždata are not shown.. This finding also is in good agree-
S.-M. Oh et al. r The Science of the Total En¨ ironment 263 (2000) 161᎐169
167
Fig. 6. Concentration᎐response curves for estrogenic effect of sediment samples of Kumho, Kum River and Mankyung, Miho stream by E-SCREEN assay. All the samples were dissolved in DMSO and 1 l of this solution was added to the cell culture medium Ž1 ml. for treatments. These were done in 24-well plates and lasted 6 days. The data represent the means of three independent experiments. The coefficient of variation was - 10%.
ment with the report from Korner et al. Ž1999.. ¨ Cytotoxicity of the extract was not observed in all the samples that we analyzed. The river samples induced a proliferative effect relative to 17-estradiol between 30 and 83%, indicating the presence of a full estrogen receptor agonist. Expressed as 17-estradiol equivalent concentrations, the levels of total estrogenic activity were 7.4 ng EEQrl and 2.0 ng EEQrl for Kumho downstream and Kum upstream, respectively. This was the first demonstration that the total estrogenicity was determined in the river
waters of South Korea. The values of this study indicate that rivers of South Korea are possibly contaminated with estrogenic chemicals.
5. Conclusion In this study, we have demonstrated that the E-SCREEN assay with human MCF7-BUS breast cancer cells is a sensitive and stable tool to analyze quantitatively complex mixtures such as river water for their total content of estrogenic active
S.-M. Oh et al. r The Science of the Total En¨ ironment 263 (2000) 161᎐169
168
Table 3 Quantification of estrogenic effect of river water, sediment and pore water samples according to the E-SCREEN assay a Sample
Water
Sediment b
Žpg EEQrl.
Kumho Kum Mankyung Miho a b
Upstream Downstream Upstream Downstream Upstream Downstream Upstream Downstream
24.05 7431.68 2046.86 0.91 0.66 10.74 6.04 0.50
DC extracts Žpg EEQrg.
Water extracts Žpg EEQrg.
᎐ 3.27 ᎐ 8.04 ᎐ 10.60 ᎐ 4.96
᎐ 0.10 ᎐ 0.05 ᎐ 0.08 ᎐ 0.11
Pore water Žpg EEQrg. ᎐ 0.02 ᎐ 0.04 ᎐ 0.02 ᎐ 0.03
Values represent mean " S.D. The 17-estradiol equivalent concentration ŽEEQ. is calculated with E 2 and the test compound.
compounds. The extracts of environmental samples did not show cytotoxic effects in the range of dilutions where a dose᎐response relationship of estrogenic activity was observed. The detection limit by the EEQ calculation of the E-SCREEN assay was 8.03 pg EEQrl Ždata not shown.. Among the water samples, the estrogenic activity was observed to be higher downstream of the Kumho river Ž7.43 ng EEQrl. and upstream of the Kum river Ž2.05 ng EEQrl. than in other samples. More than 3 mg of equivalent sediment samples from the Kumho river, Kum river and Miho stream showed partial agonistic effects and, in the case of the Mankyung river, showed a partial agonistic effect with only 1.5 mg of sediments. The highest value of RPE was 83.34 downstream of the Kumho river, and the lowest value of RPE was 6.52 downstream of the Miho stream. Full estrogen agonistic activities were observed downstream of the Kumho river and upstream of the Kum river. Partial agonistic activity was observed upstream of the Kumho river, downstream of the Mankyung river, and upstream of the Miho stream, and no agonistic action was observed downstream of the Kum river or Miho stream, or upstream of the Mankyung river. We have found, in all eight water samples and four DC extracts of sediments in South Korea, that the estrogenic activity was between 0.50 pg EEQrL and 7.4 ng EEQrL, 3.27 pg EEQrg and 10.60 pg EEQrg.
Acknowledgements This study was partly supported by the 1998 G7 grants from the ministry of environments in Korea. References Blom A, Ekman E, Johannisson A, Norrgren L, Pesonen M. Effects of xenoestrogenic environmental pollutants on the proliferation of a human breast cancer cell lineŽMCF-7.. Arch Environ Contam Toxicol 1998;34:306᎐310. Colborn T, vom Saal F, Soto AM. Developmental effects of endocrine-disrupting chemicals in wildlife and humans. Environ Health Perspect 1993;101:378᎐384. Hileman B. Concerns broaden over chlorine and chlorinated hydrocarbons. Chem Eng News 1993;71:11᎐20. Hileman B. Environmental estrogens linked to reproductive abnormalities, cancer. Chem Eng News 1994;72:19᎐23. Korner W, Hanf V, Schuller W, Kempter C, Metzger J, ¨ Hagenmaier H. Development of a sensitive E-SCREEN assay for quantitative analysis of estrogenic activity in municipal sewage plant effluents. Sci Total Environ 1999; 225:33᎐48. Nagel SC, Saal FS, Thayer KA, Dhar MG, Boechler M, Welshons WV. Relative binding affinity-serum modified access ŽRBA-SMA. assay predicts the relative in vivo bioactivity of the xenoestrogens bisphenol A and octylphenol. Environ Health Perspect 1997;105:70᎐76. Olea N, Pulgar R, Perez P et al. Estrogenicity of resin-based composites and sealants used in dentistry. Environ Health Perspect 1996;104Ž3.:298᎐305. Sonnenschein C, Soto A, Fernandez MF, Olea N, Olea-Serrano MF, Ruiz-Lopez MD. Development of a marker of estrogenic exposure in human serum. Clin Chem 1995; 41:1888᎐1895.
S.-M. Oh et al. r The Science of the Total En¨ ironment 263 (2000) 161᎐169 Perez P, Pulgar R, Olea-Serrano F et al. The estrogenicity of bisphenol A-related diphenylalkanes with various substituents at the central carbon and the hydroxy groups. Environ Health Perspect 1998;106Ž3.:298᎐305. Soto AM, Chung KL, Sonnenschein C. The pesticides endosulfan, toxaphene, and dieldrin have estrogenic effects on human estrogen-sensitive cells. Environ Health Perspect 1994;102:380᎐383. Soto AM, Sonnenschein C, Chung KL, Fernandez MF, Olea N, Serrano FO. The E-SCREEN assay as a tool to identify estrogens: An update on estrogenic environmental pollutants. Environ Health Perspect 1995;103ŽSuppl 7.:113᎐122.
169
Zacharewski T. In vitro bioassays for assessing estrogenic substances. Environ Sci Tech 1997;31:612᎐623. Villalobos M, Olea N, Brotons JA, Olea-Serraon MF, Ruiz de Almodovar JM, Pedraza V. The E-SCREEN assay: a comparison of different MCF7 cell stocks. Environ Health Perspect 1995;103:844᎐850. Waller AM, Minor DL, Mckinney JD. Using three-dimensional quantitative structure-activity relationships to examine estrogen receptor binding affinities of polychlorinated hydroxybiphenyls. Environ Health Perspect 1995; 103:702᎐707.
Quantitative assessment of estrogenic activity in the water environment of Korea by the E-SCREEN assay Seung-Min Oha , Se-Young Choung b, Yhun-Yhong Sheen c , Kyu-Hyuck Chung a,U a
College of Pharmacy, Sungkyunkwan Uni¨ ersity, 噛300, Chunchun-dong, Changan-gu, Suwon, Kyonggi-do 440-746, South Korea b College of Pharmacy, Kyoung Hee Uni¨ ersity, 噛1, Hoeki-dong, Dongdaemun-gu, Seoul 130-701, South Korea c College of Pharmacy, Ewha Womans Uni¨ ersity, 噛11-1, Daehyun-dong, Sudaemun-gu, Seoul 120-750, South Korea Received 23 February 2000; accepted 16 June 2000
Abstract In this study, the E-SCREEN assay was optimized and validated for the sensitive quantitative determination of the total estrogenicity in river samples. River water and sediment samples were collected and analyzed with the E-SCREEN. River water Ž10 l. was extracted using combined solid-phase extraction in static adsorption mode with Soxhlet extraction. Estrogenic pollutants adsorbed to the XAD-4 resin were recovered with 98.24" 5.90% efficiency by elution with ethyl acetate and dichloromethane Ž1:9.. The detection limit by 17-estradiol equivalent concentration ŽEEQ. of the E-SCREEN assay was 8.03 pg EEQrl. Among the water samples, the estrogenic activity was observed to be higher downstream of the Kumho river Ž7.43 ng EEQrl. and upstream of Kum river Ž2.05 ng EEQrl. than in other samples. More than 3 mg of equivalent sediment samples from the Kumho river, Kum river and Miho stream showed partial agonistic effects, and the Mankyung river showed a partial agonistic effect with only 1.5 mg of sediment. The highest value of RPE was 83.34 downstream of the Kumho river, and the lowest value of RPE was 6.52 downstream of the Miho stream. Full estrogen agonistic activities were observed downstream of the Kumho river and upstream of the Kum river. The partial agonistic activity was observed in upstream of the Kumho river, downstream of the Mankyung river, and upstream of the Miho stream, and no agonistic action was observed downstream of the Kum river or Miho stream, or upstream of the Mankyung river. The total estrogenic activity in the river water and sediment samples was between 0.50 pgrL and 7.4 ngrL, 3.39 pgrg and 10.70 pgrg. 䊚 2000 Elsevier Science B.V. All rights reserved. Keywords: River waters; Combined solid-phase extraction; E-SCREEN assay; MCF7-BUS cell; XAD-4; RPP; RPE; EEQ
U
Corresponding author. Tel.: q82-31-290-7714; fax: q82-31-292-8800. E-mail address: [email protected] ŽK. Chung.. 0048-9697r00r$ - see front matter 䊚 2000 Elsevier Science B.V. All rights reserved. PII: S 0 0 4 8 - 9 6 9 7 Ž 0 0 . 0 0 6 9 7 - 5
162
S.-M. Oh et al. r The Science of the Total En¨ ironment 263 (2000) 161᎐169
1. Introduction Organic contaminants, as either point source pollutants from incompletely treated sewage water or non-point source pollutants, such as pesticides, may be the source of endocrine disrupters in river water. Because of insufficient wastewater collection and treatment due to the shortage of facilities in Korea, river water may be contaminated with various chemicals including endocrine disrupters. Recently, since endocrine disrupters flowing into our environment become a big social issue, it is important to examine the estrogenic activity in our aquatic environment ŽColborn et al., 1993; Hileman, 1993, 1994; Soto et al., 1995; Waller et al., 1995.. However, most chemical analytical methods are not sufficient to identify all the environmental chemicals and to predict a combination of toxicity and bioavailability in biological organisms. Recent research has demonstrated that one of the most cost-effective ways to evaluate environmental pollution is to first use short-term, inexpensive bioassays to screen the samples for indications of toxic effects, and prioritizing the samples or sampled areas for chemical analyses andror more intensive studies. Thus, bioassays are becoming popular as screening methods for environmental compounds. Among these methods, cell culture bioassays are actively studied because they are rapid and sensitive in assessing the single andror combination effects of the chemicals on the ecosystem, as well as on humans. The E-SCREEN assay is a cell culture bioassay where the cell proliferation of human breast cancer MCF-7 cells in response to estrogen is measured, and it is a very sensitive assay by which one can measure single as well as multiple chemicals at the same time ŽSoto et al., 1995; Nagel et al., 1997; Blom et al., 1998.. Although cell proliferation assays may have their limitations ŽZacharewski, 1997., they can give quantitative estimates of test samples, so they can be quite applicable for environmental samples ŽKorner et al., 1999.. ¨ The Kumho, Kum, and Mankyung rivers, and the Miho stream are the representative rivers located in central Korea. Since the rivers flow through rural areas and several cities, they may
be polluted with various chemicals, including EDCs. Thus, in order to examine our aquatic environment, the estrogenic activities of these river water and sediment samples, previously isolated from water by an improved solid-phase extraction method with static mode, were assessed using the E-SCREEN assay.
2. Experimental methods 2.1. Examined ri¨ er waters and sample pre-treatment In May 1999, river water and sediments were collected from the upstream points and downstream points of the Kumho, Kum, and Mankyung rivers and the Miho river. Immediately after transport to laboratory, each river water sample Ž1 l. was acidified with concentrated sulfuric acid to pH 3. Active components were adsorbed for 24 h by stirring with 0.5 grl of Amberlite XAD-4 resin ŽFluka., based on the static adsorption mode of the combined solid-phase extraction method. Pollutants adsorbed to the resin were eluted with mixture of dichloromethane ŽDCM. and ethyl acetate Ž9:1. in a Soxhlet apparatus, and then dried under a reduced pressure. Residues were dissolved in 100% DMSO Ž625 l. and stored at y20⬚C until the assays were conducted. To prepare pore water samples, each sediment sample Ž10 g. was centrifuged at 5000 = g for 10 min, and the supernatant was taken and freeze-dried. The dried supernatant was redissolved in a methanolrwater mixture Ž1:1. and was purged with N2 gas, and finally dissolved in 1 ml of 100% DMSO. The sediments, after the removal of pore water, were dried for 24 h at room temperature. For the organic solvent extract, a sediment sample of 10 g of dried sediments was shaken in DCM for 4 h and filtered. The extract was dried under a reduced pressure and finally dissolved in 100% DMSO Ž500 l.. The water extract sample was extracted using deionized water instead of DCM. The samples that we used for the test were equivalent to 1.6 ml of river water, 10 mg of sediments, and pore water from 20 mg of sediments.
S.-M. Oh et al. r The Science of the Total En¨ ironment 263 (2000) 161᎐169
2.2. Cell culture MCF-7 BUS, human breast cancer, estrogen sensitive cells were kindly provided by Dr Soto at Tufts University. The cells were grown in 5% fetal bovine serum ŽHyclone, Logan, UT., Dulbeccos modified eagles medium ŽDMEM, Gibco, MD., under 5% CO 2 at 37⬚C. 2.3. CDFBS preparation In order to minimize the estrogenic activity from serum, FBS was stripped with 5% charcoal᎐0.5% dextran ŽOlea et al., 1996.. Charcoal Žacid washed, Sigma. was activated with cold sterile water immediately before use. A 5% charcoal᎐0.5% dextran T70 suspension was prepared. To this suspension, an equal volume of serum was added and maintained in suspension by rolling at 6 cyclesrmin at 37⬚C for 1 h. After centrifugation at 2000 = g for 50 min, the supernatant was taken and run through 0.45 and 0.2-m syringe filters ŽAxygen. and then stored at y20⬚C. 2.4. E-SCREEN assay The E-SCREEN assay using MCF-7-BUS cells was carried out according to the method of Perez et al. Ž1998.. The cells were harvested with 0.05% trypsin᎐0.53 mM EDTA⭈ 4Na ŽGibco. and resuspended in 5% FBS DMEM, and then plated onto a 24-well plate. The cell density was 10 000 cells per well and the culture was maintained at 37⬚C under 5% CO 2 for 24 h before the DMEM changed to phenol red containing free 10% CDFBS DMEM. The cells were exposed to various chemicals and test samples for 6 days, and then a sulforhodamine B ŽSRB. assay was conducted to measure cell proliferation. The cells were washed twice with phosphate buffered saline ŽPBS. before fixation with 10% TCA Ž500 l per well. at 4⬚C for 30 min. Each well was washed with sterile water 5᎐6 times, and to each dried well, 300 l of 1% acetic acid and 0.4% SRB was added and incubated for 15 min. The wells were washed 5᎐6 times with 1% acetic acid, and 1 ml of 10 mM trisma base ŽpH 10.5. was added into each dried well. The optical density was measured
163
at 490 nm using a microplate reader ŽMultiscan MCCr340 Pversion 2.3, Labsystems, Finland.. Each data point represented three independent experiments where three independent wells were used for the one experiment. 2.5. Quantitati¨ e measurement of estrogen action The relative proliferation effect ŽRPE., relative proliferation potency ŽRPP., and 17-estradiol equivalent concentration ŽEEQ. were calculated to estimate the estrogenic action of chemicals relative to 17-estradiol, which was the positive control. Based on the minimal concentration of 17-estradiol that resulted in the maximal cell proliferation effect, comparisons were made with the minimal concentration of test chemicals for the maximal cell proliferation effect to yield the RPP. The proliferative effect of test chemicals relative to positive control 17-estradiol was represented as RPE. Full agonistic activity was considered when the RPE was greater than 70, partial agonistic activity with RPE ranged from 25 to 70, and no effect when RPE was less than 25 ŽSoto et al., 1995.. As shown in Fig. 1, when bisphenol A and dieldrin were tested according to this method, the RPE of bisphenol A was 98.7 which showed that it was a full agonist and the RPE for dieldrin was 41.2, which indicated that it was a partial agonist. The RPP values of bisphenol
Fig. 1. Schematic representation of the dose᎐response curve to E 2 , bisphenol A and dieldrin. RPP: relative proliferative potency; RPE: relative proliferative effect s 100 = ŽPE ch emical y 1.rŽPE 2 y 1..
164
S.-M. Oh et al. r The Science of the Total En¨ ironment 263 (2000) 161᎐169
A and dieldrin were 0.01 and 0.001, respectively. This meant that bisphenol A and dieldrin needed 100 and 1000 times concentrations, respectively, to give the same effect as 17-estradiol. The EEQ was calculated as the concentration of 17estradiol that results in the same cell proliferation effect as the tested chemicals from the dose᎐response curve of the cell proliferation study with 17-estradiol as a positive control.
3. Results 3.1. Estrogenic acti¨ ity in ri¨ er water samples The water sample extracts were tested for MCF7-BUS cell proliferation, and the results are shown in Fig. 2, where estrogen effects were observed with all the samples. As shown in Fig. 3 and Fig. 4, the dose responsive effects of the Kumho and Kum river water samples were observed, respectively, indicating that this response caused by estrogenic chemicals in the water from both rivers. RPE values were calculated as shown in Table 1, where one can see estrogen action as well as the relative amounts of estrogenic chemicals in river water. The highest value
Fig. 2. Proliferative effect of MCF7-BUS cells grown in 24-well plates after 6 days incubation with river water samples Žtreated volume: 1.6 ml. in culture medium with 10% charcoal dextran treated fetal bovine serum. Values represent mean " S.D. Statistically different form the control Ž0.1% DMSO. group: U : P- 0.05; UU : P- 0.01.
of RPE was 83.34, downstream of the Kumho river, and the lowest value of RPE was 6.52, downstream of the Miho stream. Full estrogen agonistic activities were observed downstream of the Kumho river and upstream of the Kum river. The partial agonistic activity was observed up-
Table 1 Quantification of estrogenic effect of river water samples according to the E-SCREEN assay a Sampleb
PEc
RPE Ž%.d
Kumho
3.32" 0.16 6.15" 0.17 5.52" 0.29 1.70" 0.11 1.54" 0.15 2.92" 0.12 2.63" 0.08 1.40" 0.12 7.18" 0.15
37.57" 2.67 83.34" 4.83 73.06" 2.90 11.38" 0.77 8.79" 0.85 31.00" 1.98 26.43" 0.37 6.52" 0.77 100.00" 9.07
Upstream Downstream Kum Upstream Downstream Mankyung Upstream Downstream Miho Upstream Downstream 17-E2 Ž10y1 0 M. a
Values represent mean " S.D. Treated sample extracts are equivalent to 1.6 ml of river water. c The proliferative effect ŽPE. is the ratio of the highest cell number achieved with E 2 or the test compound, respectively, and the cell number of the negative control. d The relative proliferative effect ŽRPE. is calculated as 100 = ŽPE test compound y 1.rŽPE 17 -estradiol y 1.. b
Fig. 3. Concentration᎐response curves for estrogenic effect of River Kumho water by E-SCREEN assay. All the samples were dissolved in DMSO and 1 l of this solution was added to the cell culture medium Ž1 ml. for treatments. These were done in 24-well plates and lasted for 6 days. The data represent the means of three independent experiments. The coefficient of variation was - 10%.
S.-M. Oh et al. r The Science of the Total En¨ ironment 263 (2000) 161᎐169
165
3.2. Estrogenic acti¨ ity of sediment from ri¨ er sample
Fig. 4. Concentration᎐response curves for estrogenic effect of River Kum water by E-SCREEN assay. All the samples were dissolved in DMSO and 1 l of this solution was added to the cell culture medium Ž1 ml. for treatments. These were done in 24-well plates and lasted for 6 days. The data represent the means of three independent experiments. The coefficient of variation was - 10%.
stream of the Kumho river, downstream of the Mankyung river, and upstream of the Miho stream, and no agonistic action was observed downstream of the Kum river and the Miho streams, and upstream of the Mankyung river. It was interesting to note that estrogen agonistic activity was observed upstream of the Kum and Miho rivers, but no estrogenic activity was observed downstream of these rivers.
It was only possible to collect sediments from the downstreams of these rivers. The estrogenic effects of the water extracts, pore water and organic extracts of sediment are shown in Fig. 5. No estrogenic activity was observed in the pore water and water extract of the sediments, but highly induced estrogenic activity was observed in DCM extracts. From this result, estrogenic contaminants seem to be hydrophobic. The relative estrogenic effects of these chemicals are shown in Table 2. The DCM extracts showed partial estrogenic activity, whereas water extracts and pore waters associated with sediments showed no estrogenic effects. Fig. 6 shows the dose᎐response effects of DCM extracts from the river sediments. More than 3 mg of equivalent sediment samples from Kumho river, Kum river and Miho stream showed partial agonistic effects and, in the case of the Mankyung river, showed partial agonistic effects with only 1.5 mg of sediment. Thus, the Mankyung river sediments revealed approximately two times the estrogenic activity compared to the other river sediments. 3.3. Quantitati¨ e e¨ aluation based on EEQ In order to quantify the estrogenic activity of contaminants, the EEQ of each water sample was
Table 2 Quantification of estrogenic effect of river sediment and pore water samples according to the E-SCREEN assay a Sampleb
Kumho Kum Mankyung Miho 17 y E2 Ž10y10 M. a
Sediment DCM extracts c
Sediment water extracts
Pore water Ž20 mg-sediment.
PEd
RPE Ž%.e
PE
RPE Ž%.
PE
RPE Ž%.
3.24" 0.16 3.69" 0.29 3.82" 0.15 3.44" 0.08 ᎐
36.22" 1.63 43.45" 1.67 45.57" 1.68 39.51" 1.75 ᎐
1.52" 0.03 1.12" 0.02 1.38" 0.03 1.58" 0.06 7.18" 0.15
8.48" 0.17 2.00" 0.04 6.12" 0.14 9.31" 0.37 100.00" 9.07
1.14" 0.10 1.41" 0.13 1.16" 0.11 1.26" 0.12 ᎐
2.23" 0.10 6.59" 0.13 2.55" 0.11 4.19" 0.12 ᎐
Values represent mean " S.D. Treated sample extracts are equivalent to 10 mg of sediments, and pore water from 20 mg of sediments. c DCM: dichloromethane. d The proliferative effect ŽPE. is the ratio of the highest cell number achieved with E 2 or the test compound, respectively, and the cell number of the negative control. e The relative proliferative effect ŽRPE. is calculated as 100 = ŽPE test compound y 1.rŽPE 17yestradiol y 1.. b
166
S.-M. Oh et al. r The Science of the Total En¨ ironment 263 (2000) 161᎐169
Fig. 5. Proliferative effect of MCF7-BUS cells grown in 24-well plates after 6 days incubation with river sediment Žtreated volume: 10 mg-sediment. and pare water samples Žtreated volume: 20 mg sediment. in culture medium with 10% charcoal dextran fetal bovine serum. Values represent mean " S.D. Statistically different from the control Ž0.1% DMSO. group: U s P- 0.05; UU s P- 0.01; DCs dichloromethane extract; W s water extract.
calculated in comparison to 17-estradiol ŽTable 3.. The detection limit by EEQ calculation of the E-SCREEN assay was 8.03 pg EEQrl Ždata not shown.. Among the water samples, the estrogenic activity was observed to be higher downstream of the Kumho river Ž7.43 ng EEQrl. and upstream of the Kum river Ž2.05 ng EEQrl. than other samples. So far, there are few reports on the measurement of estrogenic activity in environmental samples. In this study, these high concentrations were close to the EEQ level Ž7.8᎐25 ng EEQrl. of municipal sewage plant effluents in Germany ŽKorner et al., 1999.; whereas upstream ¨ of the Kumho and Miho rivers and downstream of the Mankyung river were slightly contaminated ŽEEQ was in approx. picogram range..
4. Discussion MCF7-BUS human breast cancer cells containing large amounts of estrogen receptors are easy to culture and possess stable biological responses to estrogen compared to other human cell lines. Recently, Soto et al. Ž1995. demonstrated that, in spite of substantial modifications and simplifications, the proliferative response of MCF7-BUS
cells in relation to controls with and without 17-estradiol ŽE 2 . was a simple, reliable and sensitive screening-system for the detection and quantification of direct receptor-mediated estrogenic activity of single chemicals. In this study, we used 24-well plates for the E-SCREEN assay, and the detection limit of the E-SCREEN assay for the natural estrogen 17-estradiol was at 1 fM Ž0.27 fgrml. Ždata not shown.. The IC50 value for 17-estradiol was between 1 and 5 nM; i.e. two to three orders of magnitude lower than the physiological serum levels of E 2 in women. The quantitative results for natural estrogen, 17-estradiol tested by E-SCREEN with MCF7-BUS was in good agreement with those reported by other groups ŽSonnenschein et al., 1995; Soto et al., 1995; Villalobos et al., 1995; Korner et al., 1999.. ¨ Therefore, the E-SCREEN assay is suitable to establish 17-estradiol equivalency factors for environmental estrogens to rank their estrogenic potency relative to the natural ligand 17estradiol ŽKorner et al., 1999.. The most impor¨ tant prerequisite for applying the concept of estradiol equivalency factors to environmental samples is the additive behavior of the estrogenic activity of single substances in mixture. Korner et ¨ al. Ž1999., in their experiments with two different mixtures each containing three xenoestrogenic responses of the single compounds, thus demonstrated that the estrogen equivalency factor concept could be applied to environmental samples. Soto et al. Ž1994. also found additive behavior of various proliferation induced by non-steroidal xenoestrogens in estrogen receptor-positive MCF7 breast cancer cells, based on a common estrogen receptor-mediated mechanism. The different classes of substances such as steroids, alkylphenols, bisphenols, phthalates, and chlorinated hydrocarbons known to act like estrogen, require extraction methods which enable the simultaneous quantitative extraction of all these compounds from various environmental sources. For the extraction of estrogenic compounds, we found out that static adsorption was much better than the dynamic adsorption that is common method. We had much higher recovery Ž98.24" 5.90%. with XAD-4 than with XAD-2 Ždata are not shown.. This finding also is in good agree-
S.-M. Oh et al. r The Science of the Total En¨ ironment 263 (2000) 161᎐169
167
Fig. 6. Concentration᎐response curves for estrogenic effect of sediment samples of Kumho, Kum River and Mankyung, Miho stream by E-SCREEN assay. All the samples were dissolved in DMSO and 1 l of this solution was added to the cell culture medium Ž1 ml. for treatments. These were done in 24-well plates and lasted 6 days. The data represent the means of three independent experiments. The coefficient of variation was - 10%.
ment with the report from Korner et al. Ž1999.. ¨ Cytotoxicity of the extract was not observed in all the samples that we analyzed. The river samples induced a proliferative effect relative to 17-estradiol between 30 and 83%, indicating the presence of a full estrogen receptor agonist. Expressed as 17-estradiol equivalent concentrations, the levels of total estrogenic activity were 7.4 ng EEQrl and 2.0 ng EEQrl for Kumho downstream and Kum upstream, respectively. This was the first demonstration that the total estrogenicity was determined in the river
waters of South Korea. The values of this study indicate that rivers of South Korea are possibly contaminated with estrogenic chemicals.
5. Conclusion In this study, we have demonstrated that the E-SCREEN assay with human MCF7-BUS breast cancer cells is a sensitive and stable tool to analyze quantitatively complex mixtures such as river water for their total content of estrogenic active
S.-M. Oh et al. r The Science of the Total En¨ ironment 263 (2000) 161᎐169
168
Table 3 Quantification of estrogenic effect of river water, sediment and pore water samples according to the E-SCREEN assay a Sample
Water
Sediment b
Žpg EEQrl.
Kumho Kum Mankyung Miho a b
Upstream Downstream Upstream Downstream Upstream Downstream Upstream Downstream
24.05 7431.68 2046.86 0.91 0.66 10.74 6.04 0.50
DC extracts Žpg EEQrg.
Water extracts Žpg EEQrg.
᎐ 3.27 ᎐ 8.04 ᎐ 10.60 ᎐ 4.96
᎐ 0.10 ᎐ 0.05 ᎐ 0.08 ᎐ 0.11
Pore water Žpg EEQrg. ᎐ 0.02 ᎐ 0.04 ᎐ 0.02 ᎐ 0.03
Values represent mean " S.D. The 17-estradiol equivalent concentration ŽEEQ. is calculated with E 2 and the test compound.
compounds. The extracts of environmental samples did not show cytotoxic effects in the range of dilutions where a dose᎐response relationship of estrogenic activity was observed. The detection limit by the EEQ calculation of the E-SCREEN assay was 8.03 pg EEQrl Ždata not shown.. Among the water samples, the estrogenic activity was observed to be higher downstream of the Kumho river Ž7.43 ng EEQrl. and upstream of the Kum river Ž2.05 ng EEQrl. than in other samples. More than 3 mg of equivalent sediment samples from the Kumho river, Kum river and Miho stream showed partial agonistic effects and, in the case of the Mankyung river, showed a partial agonistic effect with only 1.5 mg of sediments. The highest value of RPE was 83.34 downstream of the Kumho river, and the lowest value of RPE was 6.52 downstream of the Miho stream. Full estrogen agonistic activities were observed downstream of the Kumho river and upstream of the Kum river. Partial agonistic activity was observed upstream of the Kumho river, downstream of the Mankyung river, and upstream of the Miho stream, and no agonistic action was observed downstream of the Kum river or Miho stream, or upstream of the Mankyung river. We have found, in all eight water samples and four DC extracts of sediments in South Korea, that the estrogenic activity was between 0.50 pg EEQrL and 7.4 ng EEQrL, 3.27 pg EEQrg and 10.60 pg EEQrg.
Acknowledgements This study was partly supported by the 1998 G7 grants from the ministry of environments in Korea. References Blom A, Ekman E, Johannisson A, Norrgren L, Pesonen M. Effects of xenoestrogenic environmental pollutants on the proliferation of a human breast cancer cell lineŽMCF-7.. Arch Environ Contam Toxicol 1998;34:306᎐310. Colborn T, vom Saal F, Soto AM. Developmental effects of endocrine-disrupting chemicals in wildlife and humans. Environ Health Perspect 1993;101:378᎐384. Hileman B. Concerns broaden over chlorine and chlorinated hydrocarbons. Chem Eng News 1993;71:11᎐20. Hileman B. Environmental estrogens linked to reproductive abnormalities, cancer. Chem Eng News 1994;72:19᎐23. Korner W, Hanf V, Schuller W, Kempter C, Metzger J, ¨ Hagenmaier H. Development of a sensitive E-SCREEN assay for quantitative analysis of estrogenic activity in municipal sewage plant effluents. Sci Total Environ 1999; 225:33᎐48. Nagel SC, Saal FS, Thayer KA, Dhar MG, Boechler M, Welshons WV. Relative binding affinity-serum modified access ŽRBA-SMA. assay predicts the relative in vivo bioactivity of the xenoestrogens bisphenol A and octylphenol. Environ Health Perspect 1997;105:70᎐76. Olea N, Pulgar R, Perez P et al. Estrogenicity of resin-based composites and sealants used in dentistry. Environ Health Perspect 1996;104Ž3.:298᎐305. Sonnenschein C, Soto A, Fernandez MF, Olea N, Olea-Serrano MF, Ruiz-Lopez MD. Development of a marker of estrogenic exposure in human serum. Clin Chem 1995; 41:1888᎐1895.
S.-M. Oh et al. r The Science of the Total En¨ ironment 263 (2000) 161᎐169 Perez P, Pulgar R, Olea-Serrano F et al. The estrogenicity of bisphenol A-related diphenylalkanes with various substituents at the central carbon and the hydroxy groups. Environ Health Perspect 1998;106Ž3.:298᎐305. Soto AM, Chung KL, Sonnenschein C. The pesticides endosulfan, toxaphene, and dieldrin have estrogenic effects on human estrogen-sensitive cells. Environ Health Perspect 1994;102:380᎐383. Soto AM, Sonnenschein C, Chung KL, Fernandez MF, Olea N, Serrano FO. The E-SCREEN assay as a tool to identify estrogens: An update on estrogenic environmental pollutants. Environ Health Perspect 1995;103ŽSuppl 7.:113᎐122.
169
Zacharewski T. In vitro bioassays for assessing estrogenic substances. Environ Sci Tech 1997;31:612᎐623. Villalobos M, Olea N, Brotons JA, Olea-Serraon MF, Ruiz de Almodovar JM, Pedraza V. The E-SCREEN assay: a comparison of different MCF7 cell stocks. Environ Health Perspect 1995;103:844᎐850. Waller AM, Minor DL, Mckinney JD. Using three-dimensional quantitative structure-activity relationships to examine estrogen receptor binding affinities of polychlorinated hydroxybiphenyls. Environ Health Perspect 1995; 103:702᎐707.