Combined in Situ and in Vitro Assessment of the Estrogenic Activity of Sewage and Surface Water Samples

Sascha Pawlowski*,1, Thomas Ternes{dagger}, Martin Bonerz{dagger}, Tatjana Kluczka{ddagger}, Bart van der Burg§, Heinz Nau{ddagger}, Lothar Erdinger and Thomas Braunbeck*

* Department of Zoology, University of Heidelberg, Im Neuenheimer Feld 230, D-69120 Heidelberg, Germany; {dagger} ESWE-Institute for Water Research and Water Technology, Söhnleinstrasse 158, D-65201 Wiesbaden, Germany; {ddagger} Department of Food Toxicology, Center of Food Science, School of Veterinary Medicine Hannover, Bischofsholer Damm 15, D-30173 Hannover, Germany; § Bio Detection Systems b.v., Kanaal Laboratorium, Badhuisweg 3, 1031 CM Amsterdam, The Netherlands; and Department of Hygiene, University of Heidelberg, Im Neuenheimer Feld 354, D-69120 Heidelberg, Germany

Received April 16, 2003; accepted May 27, 2003


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIAL AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
In order to investigate the estrogenic activities of two municipal sewage treatment plant (STP; sites A and B) effluents and of Rhine water sampled at Worms (site C; Rhine-Neckar triangle, Germany), data from in situ experiments measuring hepatic vitellogenin expression from caged rainbow trout (Oncorhynchus mykiss) were compared with data from in vitro bioassays (yeast estrogen screen [YES], ER luciferase assay with HEK 293 cells [HEK], primary rainbow trout hepatocytes [PH]) and chemical analysis. Three sampling campaigns were carried out at each site between November 2000 and September 2001. Vitellogenin (VTG)-mRNA expression in male rainbow trout exposed for two weeks ranged from 3 ± 5 to 619 ± 188 and from 226 ± 38 to 3373 ± 1958 pg/µg total RNA at sites A and B, respectively. E2-equivalents obtained from the in vitro bioassays gave values up to 0.21 ± 0.04 nM (57.3 ± 10.2 ng/l, PH), 0.07 ± 0.03 nM (20.2 ± 6.9 ng/l; YES) and 0.008 ± 0.002 nM (2.1 ± 0.7 ng/l; HEK). In contrast, in one-year-old rainbow trout exposed at site C, no VTG-mRNA induction could be observed after two weeks of exposure. In vitro bioassays (YES, HEK, PH) indicated estrogenic activity at site C, which, however, was lower than at the investigated STP effluents. Chemical analysis of representative water samples from site A identified steroidal estrogens up to 5.6 ng/l 17ß-estradiol (E2), 19 ng/l estrone as well as 1.5 ng/l 17{alpha}-ethinylestradiol. Furthermore, the sum of fecal- and phytosteroids, resorcyclic lactones, and flavonoid concentrations were 280 (A) and 1.200 ng/l (B). In addition, site C (river Rhine) contained 3.9 ng/l E2 and 250 ng/l of fecal- and phytosteroids, respectively. Thus, STP effluents and Rhine water contain biologically relevant concentrations of estrogenic compounds, the activity of which can be detected by means of various bioassays.

Key Words: estrogenicity; STP effluent; surface water; Rhine; rainbow trout; primary hepatocytes; HEK 293 cells; yeast estrogen screen; solid phase extracted water samples; chemical analysis.


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIAL AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The potential impact of natural and man-made compounds that interfere with the hormonal systems of wildlife and humans is undergoing significant examination. These so-called endocrine disruptors have been linked to changes in sex ratio, embryonic damage, and reduced fecundity in various vertebrate species (Birnbaum, 1995Go; Fry and Toone, 1981Go; Guillette et al., 1994Go; Sumpter, 1995Go), as well as to an increase in breast and testicular cancers in humans (Carlsen et al., 1995Go; Giwercman et al., 1993Go; Toppari et al., 1996Go). Estrogens and xenoestrogens, as one group of hormonally active compounds, have been detected in sewage treatment plant effluents from various countries across the world as well as in surface waters at environmentally relevant concentrations (Belfroid et al., 1999Go; Desbrow et al., 1998Go; Ternes et al., 1999cGo). Correspondingly, estrogenic activity could be detected in sewage treatment plant (STP) effluents and river waters across Europe using in vitro and in vivo bioassays (Jobling and Sumpter, 1993Go; Körner et al., 2001Go; Sumpter, 1995Go). Previous studies linked the main estrogenic activities of municipal sewage treatment plant effluents to steroidal estrogens such as 17ß-estradiol (E2), 17{alpha}-ethinylestradiol (EE2), and estrone (E1; Desbrow et al., 1998Go) at concentrations of up to tens of ng/l (Belfroid et al., 1999Go; Körner et al., 2001Go; Ternes et al., 1999cGo). Despite the fact that several correlations between in vitro assays and in situ experiments have been drawn so far (Beresford et al., 2000Go; Tilton et al., 2002Go), there is still a lack of information for further risk assessment within the aquatic environment.

In Germany, data on estrogenic effects on aquatic organisms including fish due to STP effluents are scarce. A previous study using extracted water samples from two municipal STPs and river Rhine surface water sampled at the Rhine-Neckar-triangle (southern Germany) indicated estrogenic activities in both STP effluents and the surface water (Pawlowski et al., unpublished data). In the same study, steroidal estrogens (SE) and phytoestrogens (PE) could be identified in the STP effluents using chemical analysis at values up to 26 ng/l (SE) and 1.2 µg/l (PE), and in Rhine water of about 4 ng/l for steroidal estrogens and 0.25 µg/l for phytoestrogens. In order to evaluate the impact of domestic discharge on aquatic organisms, e.g., fish, the present study was designed to compare vitellogenin-mRNA induction in caged male or one-year-old immature rainbow trout (Oncorhynchus mykiss) with in vitro data obtained from (1) primary rainbow trout hepatocyte cultures (Islinger et al., 1999Go; Pawlowski et al., 2000Go), the (2) yeast estrogen screen (YES; Routledge and Sumpter, 1996Go), and (3) transfected human embryonic kidney 293 cells (HEK 293 cell—ER-luciferase assay; Kuiper et al., 1998Go; Meerts et al., 2001Go) exposed to extracts from water obtained from the sites where fish were exposed over nine months of sampling (from November 2000 to September 2001). Biological effects were related to data from corresponding chemical analyses.


    MATERIAL AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIAL AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Sampling sites.
All sampling sites (A, B: sewage treatment plant effluents; C: river Rhine) were located at the Rhine-Neckar triangle in Southwest Germany. Both STPs were equipped with an activated sludge process as well as denitrification, nitrification, and phosphate elimination facilities. Further details concerning capacity and location of the STPs investigated could be obtained from Table 1Go and Figure 1Go, respectively. Briefly, the effluents of the first municipal treatment plant (site A) enter the river Rhine via its tributary, the river Neckar, whereas the effluents from the second plant (site B) directly enter the river Rhine (Fig. 1Go). Both STPs mainly receive domestic influents, but capacities expressed as person equivalents differ within one order of magnitude (Table 1Go). The third sampling site (site C) was located at the river Rhine itself at Worms, receiving, among others, effluents from the STP 2 at site B.


View this table:
[in this window]
[in a new window]
 
TABLE 1 Technical Data and Water Flow Conditions of the Investigated Sampling Sites during the Whole Sampling Period
 


View larger version (29K):
[in this window]
[in a new window]
 
FIG. 1. Location of sampling sites within the Rhine-Neckar-triangle, southwest Germany. Sites A and B represent two municipal sewage treatment plants discharging into the rivers Neckar and Rhine, respectively, whereas site C represents the Rhine itself at the city of Worms.

 
Fish stocks.
Two-year-old male rainbow trout weighing 250–350 g were obtained from a local trout farm (Huber, Kirchardt, Germany) and kept in 1000 l indoor glasfiber tanks receiving aerated tap water (3 l/min, oxygen content 6.2 ± 0.5 mg/l, temperature 12 ± 2°C, 12/12 h dark/light regimen) for at least eight weeks prior to the start of the test. One-year-old rainbow trout between 10 and 25 cm in length were obtained from the same supplier and kept under similar conditions. They were sexually immature, but gonadal developments with gender-selective sex steroid patterns are already present at this age. Thus, one-year-old rainbow trout differ from juvenile rainbow trout.

Fish exposure at the STPs.
To prevent oxygen depletion in STP effluents, bypass systems consisting of 600 l glass fiber tanks were connected to the effluent via water pumps (4000 L/h; Eheim, Deizisau, FRG). The tank was covered with stainless steel grids to prevent loss of experimental fish. The water was aerated using a commercially available air pump (Rebie, Bielefeld, FRG). Measured oxygen concentrations in the tanks varied between 3.9 and 5.4 mg/l, and temperature ranged from 10 to 19°C, depending on the season (for details, see Table 1Go). In each experiment, five male rainbow trout were exposed to the effluents for a period of two weeks in November 2000 (site A), March (site B), May and September 2001 (sites A and B; Table 1Go). Control fish were kept under similar conditions, receiving tap water only.

Fish exposure at the river Rhine.
Due to space limitations at the Rheingütestation Worms, one-year-old rainbow trout (between 15 and 25 cm) were used to measure estrogenic effects in Rhine water. For each experiment, 15 one-year-old rainbow trout were transferred to 100 l glass aquaria, receiving water from lane 4 of the river Rhine (right side), which is under the influence of, among others, the sewage effluent from site B. Exposure of fish was carried out over periods of 2 weeks in March, May, and September 2001 (Table 1Go). Control fish were kept under similar conditions, receiving tap water only.

Water sampling and solid phase extraction.
While fish were exposed to sewage effluents and river Rhine water, respectively, water samples (0.5 l) were taken concomitantly on a daily basis from every site and extracted using C18 solid phase-extraction columns (1 g, Bakerbond, J. T. Baker, Deventer, The Netherlands) within 24 h upon arrival in the laboratory. All subsequent steps were carried out in duplicate: Water samples were filtered over 0.1 µm glass fibre filters (type C5, MembraPure, Bodenheim, FRG), acidified with concentrated H2SO4 to pH 2, divided into two 1-liter samples and extracted using reverse phase C18 solid phase extraction columns (RP-C18 SPE; 1 g; Bakerbond, J. T. Baker, Deventer, The Netherlands), which had been conditioned with 3 x 3 ml hexane, 3 x 3 ml acetone as well as 1 x 3 ml deionized water according to a protocol by Spengler et al. (2001)Go. After extraction, columns were centrifuged for 10 min at 2000 x g and dried under a nitrogen stream. Elution was carried out with 2 x 5 ml acetone. Eluted samples were blown to dryness under a nitrogen stream, reconstituted in 2.5 ml ethanol and stored at 4°C until use in the yeast estrogen screen (up to four weeks). All extracted water samples from one exposure regime (lasting 14 days) were mixed to obtain an average estrogenic activity over the entire experimental period. Deionized water and tap water were treated and extracted according to the same protocol and were used as process and field controls, respectively.

The yeast estrogen screen.
All medium constituents were obtained from Sigma (Deisenhofen, FRG) except for chorophenol red-ß-galactopyranoside (CPRG; Roche Diagnostics, Mannheim, FRG). Recombinant yeast cells (Saccharomyces cerevisiae) were kindly provided by John P. Sumpter from Brunel University, Uxbridge, U.K. Yeast cells were cultured, and the assay was carried out as described by Routledge and Sumpter (1996)Go with slight modifications. Briefly, 200 µl of the respective water sample extract (concentrated by a factor of 2000 if compared to the initial water sample) were added to the first column of a 96 well plate (TTP, Renner, Dannstadt, FRG) and then diluted with 100 µl ethanol in a 1:2 series across the plate, resulting in a 1000x concentrated extract as the highest concentration. 17ß-Estradiol (E2) was used as a positive control with 10 nM as the highest concentration in the first well. Ethanol (100 µl) was used as negative control in the YES. Values were compared with E2 induction rates and expressed as estradiol equivalents (E2-EQs).

ER-luciferase assay with HEK 293 cells.
Human embryonic kidney (HEK) 293 cells were kindly provided by Paul van der Saag, Hubrecht Laboratory, Netherlands Institute for Developmental Biology at Utrecht, The Netherlands. Extracted water samples were tested in HEK 293 cells stably transfected with recombinant human estrogen receptor (HEK 293 ER{alpha} or 293 ERßs-luciferase assays; Kuiper et al., 1998Go; Lemmen et al., 2002Go). Cells were cultured and exposed to extracted water samples as described by Meerts et al. (2001)Go and Kuiper et al. (1998)Go, however, with 5% CO2 instead of 7.5%. Briefly, cells were trypsinized and resuspended in steroid-free assay medium and seeded in 96-well plates at a density of 15,000 cells per well in 200 µl assay medium. After 48 h of incubation, when cell density had reached 50–60% confluence, the assay medium was replaced by incubation medium containing a 1000-fold dilution of the extracted water samples. Solvent concentrations never exceeded 0.1% in the culture medium, resulting in 2x concentrated water sample extracts in the test. 17ß-Estradiol dissolved in ethanol was used as a positive control at concentrations of 0.1, 1, 10, 100, and 1000 pM. After 24 h at 37°C, the plates were transferred onto ice and the medium was removed by suction. Cells were lysed, deep-frozen, and analyzed for luciferase activity. Each sample was tested in five independent replicates for its estrogenicity in the ER-luciferase assay. Values were compared with E2 induction rates and expressed as estradiol equivalents (E2-EQs).

Primary rainbow trout hepatocyte cultures.
Two-year-old male rainbow trout weighing 250–350 g were used as donors for primary hepatocyte cultures. Fish were anesthetized with benzocaine (Sigma), and hepatocytes were isolated as described by Islinger et al. (1999)Go with slight modifications by Pawlowski et al. (2000)Go. For culture, 2 ml aliquots of the hepatocyte suspension (1 x 106 cells/ml) were seeded in 24-well cell culture plates (TTP, Renner, Dannstadt, FRG) and incubated at 18°C.

Prior to measuring estrogenicity of extracted water samples, the solvent ethanol was replaced by dimethyl sulfoxide (DMSO). After 24 h of cell regeneration, 1 ml of the supernatant medium was replaced by the appropriate amount of fresh medium containing 0.15 % (v/v) DMSO for negative control, as well as 17ß-estradiol (E2) as positive control (0.1 and 1 nM) or the respective extract (concentration factors of 12.5, 25, and 50x) dissolved in DMSO. After 48 h of exposure at 18°C, 1 ml of medium was replaced.

For vitellogenin-mRNA (VTG-mRNA) determination, primary rainbow trout hepatocyte cultures were harvested after 96 h of exposure. Total RNA was prepared, and VTG-mRNA was quantified as described previously (Islinger et al., 1999Go; Pawlowski et al., 2000Go). Samples were tested in independent cell cultures from two different fish, VTG-mRNA detection was carried out in six replicates. The amount of VTG-mRNA was determined using VTG-mRNA standards ranging from 300 to 0.12 pg. Values were compared with E2 induction rates and expressed as 17ß-estradiol equivalents (E2-EQs).

Isolation and quantification of vitellogenin-mRNA in exposed rainbow trout.
After two weeks of exposure, fish were anesthetised with benzocaine and opened ventrally. Pieces of liver (0.3 mm3) were cut and stored immediately in RNAlater (Qiagen, Hilden, FRG) for up to 6 h at room temperature. Total RNA was prepared using QIA-Shredder and RNeasy Mini-Kits (Qiagen; Islinger et al., 1999Go; Pawlowski et al., 2000Go). Concentration and purity of RNA were determined by photometry at 260/280 nm, and RNA was stored at -70°C until use. Intactness of the RNA samples was checked by denaturating agarose/formaldehyde gel electrophoresis and ethidium bromide staining for 28 S and 18 S rRNA. Vitellogenin-mRNA was quantified in a dot blot/RNase protection assay (Islinger et al., 1999Go; Pawlowski et al., 2000Go; Zhan et al., 1997Go). Vitellogenin-mRNA detection was carried out in six replicates each from three independent fish individuals. Values were compared with E2 induction rates and expressed as estradiol equivalents (E2-EQs).

Chemical analysis of water samples.
For chemical analysis of estrogens and phytoestrogens, water samples (1 l from each site collected over 24 h) were taken on 10 September 2001. Water samples were kept at 4°C during transport and were concentrated with RP-C18 (see above) within the next 24 h. The analytical methods for estrogens, fecal- and phytosteroids, and flavanoids were described in detail by Ternes et al. (1999aGo,bGo,cGo).

Statistical analysis.
In the YES, values higher than the triple standard deviation of the negative controls were registered as statistically significant. Statistically significant differences between means of luciferase activity in the ER luciferase assay, and vitellogenin-mRNA induction in vivo were first analysed for normality/homogeneity of variance before subjected to one-way ANOVA followed by Dunn’s post-hoc test (Sigma Stat, SPSS-Jandel, Erkrath, F.R.G.). Given the limited number of replicates (n = 2), vitellogenin-mRNA induction in primary rainbow trout hepatocytes was not evaluated statistically.


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIAL AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Solid Phase-Extracted STP Effluent Samples in the YES
Estrogenic activity of extracted water samples from all three sites (A–C) could be detected with the yeast estrogen screen during all sampling periods from November 2000 (Nov 00) until September 2001 (Sep 01, Table 2Go). Extracted water samples from site A displayed highest estrogenic activity with values ranging from about 0.07 ± 0.03 (20 ± 7 ng/l) to 0.05 ± 0.01 nM (13 ± 0.1 ng/l) E2-EQs in Nov 00 and Sep 01, respectively. Site B displayed slightly lower estrogenic activity with values ranging from 0.04 ± 0.00 nM (12 ± 1 ng/l) to 0.05 ± 0.01 nM (13.4 ± 2.5 ng/l) E2-EQs in Sep and May 01, respectively.


View this table:
[in this window]
[in a new window]
 
TABLE 2 Estrogenic Activity of Solid Phase Extracted Water Samples from the Two Sewage Treatment Plants (A, B) and the River Rhine (C) in the Yeast Estrogen Screen Expressed as 17ß-Estradiol (E2) Equivalents in nM (ng/l)
 
At site C, estrogenic activity was about 10 times lower that at sites A and B, respectively. Thus, values in May 01 were about 0.004 ± 0.000 nM (1.1 ± 0.1 ng/l) E2-EQs and about 0.005 ± 0.001 nM (1.3 ± 0.2 ng/l) E2-EQs in Mar 01.

Extracted Water Samples in HEK 293 Cells
With respect to the induction of estrogen receptors (ER) {alpha} and ß activity in HEK 293 cells, no similarities in the reaction scheme of both types could be observed at any site (Table 3Go). E2-EQs in HEK 293 cells were within the pM (pg/l) range with highest values for both ERs at sites A and B. Thus, at site A statistically significant values ranged from 0.002 ± 0.0006 nM (0.7 ± 0.2 ng/l) to 0.008 ± 0.002 nM (2.1 ± 0.7 ng/l) E2-EQs for ER{alpha} in May 01 and Nov 00, respectively. In contrast, values of site A samples remained below 0.003 ± 0.001 nM (0.9 ± 0.4 ng/l) E2-EQs within the ERß system. Site B showed low levels of E2-EQs for ER{alpha} up to 0.001 ± 0.00001 nM (0.3 ± 0.03 ng/l), whereas ERß values increased from 0.003 ± 0.002 (0.85 ± 0.52 ng/l) in Mar 01 to 0.01 ± 0.001 nM (2.7 ± 0.4 ng/l) E2-EQs in Sep 01 (Table 3Go). At site C, no statistically significant elevated E2-EQs could be observed in both ER{alpha} and ß tests at any investigated season, if compared to the negative controls (Table 3Go).


View this table:
[in this window]
[in a new window]
 
TABLE 3 Estrogenic Activity of Solid Phase Extracted Water Samples from the Two Sewage Treatment Plants (A, B) and the River Rhine (C) in the HEK 293 Cell Line ({alpha} and ß) Expressed as 17ß-Estradiol (E2) Equivalents in nM (ng/l)
 
Vitellogenin-mRNA Induction in Primary Rainbow Trout Hepatocyte Cultures
Induction of VTG-mRNA could be observed in primary rainbow trout hepatocyte cultures after exposure to extracted water samples from all investigated sites (Table 4Go). Although results could not be evaluated statistically, they show clear tendencies in estrogenic activities with respect to both the investigated sites and the seasons. At sites A and B, E2-EQ values from samples in Nov 00/Mar 01, and May 01 were about 4.5–7 times higher than in Sep 01 displaying values of up to 21 ± 0.04 nM (57.3 ± 10.2 ng/l) and 0.2 ± 0.1 nM (54.7 ± 26.9 ng/l) for sites A and B, respectively. At site C, no season-dependent changes could be observed with values ranging from 0.03 ± 0.01 nM (9.8 ± 1.3 ng/l) to 0.04 ± 0.01 nM (7.2 ± 2.4 ng/l) E2-EQs in May and Sept 01, respectively.


View this table:
[in this window]
[in a new window]
 
TABLE 4 Estrogenic Activity of Solid Phase Extracted Water Samples from the Two Sewage Treatment Plants (A, B) and the River Rhine (C) in Primary Rainbow Trout Hepatocytes Expressed as 17ß-Estradiol (E2) Equivalents in nM (ng/l)
 
Vitellogenin-mRNA Induction in Male and One-Year-Old Rainbow Trout Exposed in Vivo
Vitellogenin-mRNA induction in rainbow trout was highest at site A, followed by site B, and lowest at site C (Fig. 2Go). Contrary season-dependent effects could be observed at sites A and B: Highest values of induced VTG-mRNA could be observed at site A in Nov 00 and Mar 01 (3373 ± 1958 and 3175 ± 1789 pg VTG-mRNA/µg total RNA, respectively), whereas in Sep 01 only 225 ± 38 pg VTG-mRNA/µg total RNA could be detected. Site B yielded lowest values in Mar 01 (about 3 ± 4 pg VTG-mRNA/µg total RNA) and raised to 619 ± 188 pg VTG-mRNA/µg total RNA in Sep 01. Immature 1-year-old male rainbow trout exposed at site C (river Rhine) did not show any induction of VTG-mRNA. VTG-mRNA levels in exposed 1-year-old females were always within the range of those of control females, ranging from 18.4 ± 4 to 46 ± 38 pg VTG-mRNA/µg total RNA for controls and exposed females respectively. In Mar 01, VTG-mRNA induction could not be observed in any control or exposed females (data of those control females are not included in Fig. 2Go).



View larger version (17K):
[in this window]
[in a new window]
 
FIG. 2. Vitellogenin (VTG)-mRNA induction in male and female rainbow trout (Oncorhynchus mykiss) after a two-week exposure to three investigated sites (A and B: STP effluents, n = 5; C: river Rhine at Worms, n >= 5) during various seasons from November 2000 (Nov 00) to September 2001 (Sep 01). Induction of VTG-mRNA was highest (up to 3373 ± 1958 pg/µg total RNA) at site A in Nov 00 and Mar 01, and decreased in Sep 01. At site B, highest activities could be observed in Sep 01 (619 ± 188 pg/µg total-RNA). At site C, no induction or reduction in VTG-mRNA expression could be observed in one-year-old male or and female rainbow trout, respectively. *p < 0.05 from control (one way ANOVA, Dunn’s post-hoc test). Different lettering indicates statistically differences between sites within a given season (p < 0.05; one way ANOVA, Dunn’s post-hoc test).

 
Chemical Analysis of Steroidal Estrogens, Fecal- and Phytosteroids, Resorcyclic Lactones, and Flavonoids in Representative Water Samples
Chemical analysis of water samples revealed the presence of steroidal estrogens, fecal- and phytosteroids, resorcyclic lactones, and flavonoids at any site (Table 5Go). At site A, concentrations of steroidal estrogens were 19 ng/l of E1, 5.6 ng/l of E2, and 1.5 ng/l of EE2, resulting in a total concentration of 16.1 ng/l for all steroids measured. The total amount of fecal- and phytosteroids, resorcyclic lactones, and flavonoids was about 1.2 µg/l (Table 5Go). At site B, the total amount of steroidal estrogens was lower, with values of 1.2 ng/l for E1 and 1.0 ng/L for E2 (Table 5Go). In parallel, with values of about 278 ng/l the total amount of fecal- and phytosteroids and resorcyclic lactones was lower at site B, if compared to site A (Table 5Go). Flavonoids, however, were below the detection limit at this site. In river Rhine water, steroidal estrogens, for example E2, were present in the water sample at a concentration of 3.9 ng/l, whereas E1 and EE2 were below the detection limits of chemical analysis. Furthermore, 120 ng/l cholesterol and 130 ng/l stigmasterol could be identified.


View this table:
[in this window]
[in a new window]
 
TABLE 5 Concentrations (ng/l) of Estrogens, Fecal- and Phytosteroids, Resorcyclic Lactones, and Flavonoids in STP Effluents and a Rhine Water Sample
 
The steroidal estrogens estradiol-17-valerate and mestranol, the fecal- and phytosteroids cholestan, cholestanone, lanosterole, and progesterone, the resorcyclic lactones and flavonoids biochanin A, coumestrol, enterodiol, enterolactone, equol, formononetin, genistein, isoliquiritigenin, matairesinol, naringenin, phloretin, {alpha}-zearelanole, zearalanone, and zearalenone could not be detected in any water sample.


    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIAL AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Estrogenic activities could clearly be detected in STP effluents from the two municipal STPs investigated at the Rhine-Neckar-triangle from November 2000 (Nov 00) until September 2001 (Sep 01) using combined in vitro and in situ bioassays. At the river Rhine at Worms, estrogenic activity of water samples was limited to in vitro bioassays.

Overall, there was a good correlation between results obtained from the in situ (rainbow trout in vivo) and in vitro experiments (YES, HEK 293 cells, primary hepatocytes) as well as the chemical analysis of representative water samples. Although primary rainbow trout hepatocytes data were not statistically evaluated, a tendency in reaction similar to those observed for the other in vivo and in vitro experiments could be observed. Comparing the estrogenicity of the three investigated sites, site A (STP 1 effluent) clearly displayed highest activity, followed by site B (STP 2 effluent) and C (river Rhine; Table 6Go).


View this table:
[in this window]
[in a new window]
 
TABLE 6 Relative Sensitivities of in Vitro and in Situ Bioassays to Detect Estrogenic Activity in Environmental Water Samples
 
In fact, highest E2-EQ levels in in vitro bioassays could be observed in primary rainbow trout hepatocytes, with however, values definitely higher than those calculated from chemical analysis, followed by the YES (similar values) and the ER luciferase assay (lower values). If compared to E2-EQs calculated from the chemical analysis, values from the in vitro assays differ by up to one order of magnitude, which could be due to the fact that estrogenicity of STP effluents and surface waters vary within days and months (Rodgers-Gray et al., 2000Go). Nevertheless, E2-equivalents found in STP A and B of both the primary rainbow trout hepatocytes and the YES were within the range of those observed in a previous study using male channel catfish exposed to municipal sewage effluent (23–123 ng/l E2-EQs) and corresponding YES tests (21–147 ng/l E2-EQs; Tilton et al., 2002Go). In contrast, results from another study using extracted water samples from various STP effluents sampled across southwest Germany and the MCF-7 human breast cancer cell line indicated lower E2-EQs (about 5.6 ng/l; Körner et al., 2000Go) than those observed in the recent study.

There was an apparent decrease in the sensitivity to estrogenic activity across the in vitro biotests applied with primary rainbow trout hepatocytes > YES > HEK 293 cell line. It is well known that biotransformation as well as bioactivating processes of chemicals including potential estrogens take place in primary fish hepatocytes and thus enhance their ability to detect potential estrogens (Braunbeck, 1992Go; Scholz et al., 1998Go). Thus, the higher sensitivity of primary hepatocytes might be explained by differential bioactivation of estrogens in the detection systems used.

The relatively low levels of E2-EQs in HEK 293 cells might be due to the relatively high dilution rates of the extracts tested. Given the spectrum of substances analyzed chemically, the differences in reactivities of the two subtypes investigated ({alpha} and ß; see Kuiper et al., 1998Go) could not be attributed to the specific chemicals. At present, it cannot be resolved whether these discrepancies were caused by variations within the test systems or whether they were caused by unknown (unanalyzed) endocrine active chemicals, for example, potential anti-estrogens or androgens with different modes of actions.

The estrogenic responses in the biotests used (particularly the in situ methods) may have been influenced by water temperature and mean flow rates at the different sampling periods. While E2 and EE2 concentrations were rather similar in STP effluents from September 97, January and October 1998 (Huang and Sedlak, 2001Go), a more in-depth study by Routledge et al. (1998)Go into the time course of steroid levels over several months clearly documented much higher levels of steroidal estrogens during winter months from November 1997 to March 1998, if compared to summer months. This phenomenon might explain the decrease in VTG-mRNA expression in exposed rainbow trout at site A from November 00 to September 01. With respect to site B, dilution effects are likely to be responsible for the rather low levels of estrogenic activity in exposed fish during the winter time at higher dilution rates. In fact, several previous studies have indicated a reduction in estrogenic activity in parallel to an increase in the dilution factor of STP effluent in both laboratory and field experiments using Japanese medaka (Oryzias latipes; Metcalfe et al., 2001Go) and roach (Rutilus rutilus; Rodgers-Gray et al., 2000Go, 2001Go).

The failure to detect any induced VTG-mRNA in 1-year-old male and female rainbow trout exposed to Rhine river water might be attributed to the age since fish were sexually inactive, although sexual development had been completed at this age. However, it seems more likely that either the low estrogenic activity as could be measured in the in vitro biotests or other chemicals acting as anti-estrogens, androgens, etc. interfere with VTG-mRNA induction in exposed fish at site C. Moreover, dilution effects most likely reduced the estrogenic activity in the river water below detection limits (mean flow rate: about 1500–2000 m3/s, Rhine at Worms versus 5000 m3/day at STP 2). Therefore, further contributors to the overall endocrine activity upstream the river must be assumed; in fact, other municipal STPs as potential sources for steroidal estrogens and a paper mill as a potential source for phytoestrogens) are located 10 and 15 km upstream. Thus, androgenic effects such as those observed in a previous study using guppy (Poecilia reticulata) and eelput (Zoarces viviparous) exposed to paper mill effluents (Larsson et al., 2000Go, 2002Go) cannot be excluded. This clearly documents an advantage of in vivo systems, since all in vitro assays would inevitably fail to detect this. Unfortunately, no fish were caged upstream of site C at this time to further corroborate conclusions.

Taking into account previous studies, which indicated the presence of steroidal estrogens at biologically active lower ng/l-ranges in the river Rhine (E1 up to 2.9; E2 up to 2.9 and EE2 up to 4.3 ng/L; Belfroid et al., 1999Go; Ternes et al., 1999cGo), estrogenic effects in fish could not be excluded at this time. Such effects might be even more pronounced by extended exposure of local fish in the river, as could be demonstrated for roach exposed to different STP effluent concentrations for various exposure durations (Rodgers-Gray et al., 2000Go). According to Panter et al. (2000)Go, fluctuations in the concentrations of estrogenically active chemicals within relative short periods of time do not necessarily reduce estrogenic impact on fish.

Given the results from the chemical analysis of steroidal estrogens from sites B and C, it may be assumed that the measured fecal- and phytosteroids, resorcyclic lactones, and flavonoids also contributed to the overall estrogenic activity of the water samples. Several studies indicated that phytoestrogens, with estrogenic activities 103–104 times lower than that of the natural steroid E2, are biologically active at the µg/l range (Dubé and MacLatchy, 2001Go; Tremblay and Van der Kraak, 1999Go). Thus, the concentrations of phytoestrogens measured in the present study could well be biologically relevant. Moreover, it cannot be excluded that other chemicals such as alkylphenols and bisphenols might influence the estrogenicity of water samples. Nevertheless, previous studies showed that most estrogenic activity, for example, of STP effluents, could be attributed (up to 90%) to steroidal estrogens (Desbrow et al., 1998Go). However, these studies used the YES assay to determine this, which exclusively identifies ER agonists. Estrogenic activity may occur through a whole variety of indirect mechanisms such as disruption of E2 degradation/synthesis and biotransformation. All of these potential compounds were not considered, but will still cause estrogenic activity in vivo. This is why in vivo and in vitro assays should always be applied in parallel to determine the identity of non-ER agonist chemicals that are "estrogenic." Especially androgenic, antiestrogenic, and antiandrogenic effects that might be present in surface waters (Nimrod and Benson, 1996Go; Thomas et al., 2001Go) will normally not be identified in in vitro systems when used alone.

The correlation between results from the chemical analyses in the present study with from other studies at other locations (0.4–220 ng/l for E1, 0.5–19 ng/l for E2, 0.5–7.5 ng/l for EE2 in municipal effluents; Belfroid et al., 1999Go; Rodgers-Gray et al., 2000Go; Ternes et al., 1999bGo,cGo) suggests that steroidal estrogens regularly occur in STP effluents and rivers (e.g., small rivers with lower dilution effects) and cause effects in biological systems. An impact of estrogenically active chemicals on fish population might thus be a more general problem in European freshwater systems. As a consequence, the presence of a broad spectrum of hormonally active chemicals in STP effluents and surface waters (Bätscher et al., 1999Go; Gülden et al., 1997Go; Mellanen et al., 1996Go; Tavera-Mendoza et al., 2002Go; Tremblay and Van der Kraak, 1999Go; Tyler et al., 2000Go; Willingham and Crews, 1999Go) requires a combination of both biological and chemical detection systems.

The present study clearly documents that in situ exposure of rainbow trout is capable of identifying potential estrogenic effects compounds in STP effluents; this in vivo result could be replicated using primary rainbow trout hepatocytes. Both YES and HEK 293 cells are less potent in their ability to detect estrogenic effects showing only about 50 and 10% of E2-EQs, if compared to primary rainbow trout hepatocytes. Furthermore, the presence of estrogenic activity in both STP effluents and river Rhine water indicates the risk of endocrine disruption in native fish populations. Further investigations are needed to quantify this risk of hormonal disruption of feral fish species.


    ACKNOWLEDGMENTS
 
The authors wish to express their thanks to the staffs of all the treatment plants, to the public authorities at the respective sites, and to Dr. P. Diehl and his colleagues at the Rheingütestation Worms for excellent cooperation. Thanks are due to Dr. P. van der Saag, NIOB, The Netherlands, for providing the HEK 293 cell line. This project was funded in part by the German Environmental Protection Agency (UBA), Deutsche Bundesstiftung Umwelt (DBU), and a personal grant to S.P. and T.K. by the DBU.


    NOTES
 
1 To whom correspondence should be addressed. Fax: 0049-6221-546162. E-mail: spawlows{at}zoo.uni-heidelberg.de. Back


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIAL AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Bätscher, R., Studer, C., and Fent, K. (1999). Stoffe mit endokriner Wirkung in der Umwelt. Eidgenösische Anstalt für Wasserversorgung, Abwassereinigung und Gewässerschutz (EAWAG) und Bundesamt für Umwelt, Wald und Landschaft (BUWAL), Bern.

Belfroid, A. C., Van der Horst, A., Vethaak, A. D., Schäfer, A. J., Rijs, G. B. J., Wegener, J., and Cofino, W. P. (1999). Analysis and occurrence of estrogenic hormones and their glucuronides in surface water and waste water in The Netherlands. Sci. Total Environ. 225, 101–108.[CrossRef][ISI][Medline]

Beresford, N., Routledge, E. J., Harris, C. A., and Sumpter, J. P. (2000). Issues arising when interpreting results from an in vitro assay for estrogenic activitiy. Toxicol. Appl. Pharmacol. 162, 22–33.[CrossRef][ISI][Medline]

Birnbaum, L. S. (1995). Developmental effects of dioxins. Environ. Health Perspect. 103, 89–94.

Braunbeck, T. (1992). Isolated Fish Hepatocytes and Permanent Fish Cell Lines in Cytotoxicity TestsAn Alternative to Fish Hepatocytes in Vivo? Springer, Vienna, New York.

Carlsen, E., Giwercman, A., Keiding, N., and Skakkebaek, N. K. (1995). Declining semen quality and increasing incidence of testicular cancer: Is there a common cause? Environ. Health Perspect. 103, 137–139.

Desbrow, C., Routledge, E. J., Brighty, G. C., Sumpter, J. P., and Waldock, M. (1998). Identification of estrogenic chemicals in STW effluent. 1. Chemical fractionation and in vitro biological screening. Environ. Sci. Technol. 32, 1549–1558.[CrossRef][ISI]

Dubé, M. G., and MacLatchy, D. L. (2001). Identification and treatment of a waste stream at a bleached-kraft pulp mill that drepressed a sex steroid in the mummichog (Fundulus heteroclitus). Environ. Toxicol. Chem. 20, 985–995.[CrossRef][ISI][Medline]

Fry, D. M., and Toone, C. K. (1981). DDT-induced feminization of gull embryos. Science 213, 922–924.[ISI][Medline]

Giwercman, A., Carlsen, E., Keiding, N., and Skakkebaek, N. E. (1993). Evidence for increasing incidence of abnormalities of the human testis: A review. Environ. Health Perspect. 101, 65–72.[ISI][Medline]

Guillette, Jr., L. J., Gross, T. S., Masson, G. R., Matter, J. M., Percival, H. P., and Woodward, A. R. (1994). Developmental abnormalities of the gonad and abnormal sex hormone concentrations in juvenile alligators from contaminated and control lakes in Florida. Environ. Health Perspect. 102, 680–688.[ISI][Medline]

Gülden, M., Turan, A., and Seibert, H. (1997). Substanzen mit endokriner Wirkung in Oberflächengewässern. Bundesministerium für Umwelt, Naturschutz und Reaktorsicherheit–Wasserwirtschaft.

Huang, C.-H., and Sedlak, D. L. (2001). Analysis of estrogenic hormones in municipal wastewater effluent and surface water using enzyme-linked immunosorbent assay and gaschromatography/tandem mass spectometry. Environ. Toxicol. Chem. 20, 133–139.[CrossRef][ISI][Medline]

Islinger, M., Pawlowski, S., Hollert, H., Völkl, A., and Braunbeck, T. (1999). Measurement of vitellogenin-mRNA in primary cultures of rainbow trout hepatocytes in a non-radioactive dot blot/RNase protection assay. Sci. Total Environ. 233, 109–122.[CrossRef][Medline]

Jobling, S., and Sumpter, J. P. (1993). Detergent components in sewage effluent are weakly oestrogenic to fish: An in vitro study, using rainbow trout (Oncorhynchus mykiss) hepatocytes. Aquat. Toxicol. 27, 361–372.[CrossRef][ISI]

Körner, W., Bolz, U., Süßmuth, W., Hiller, G., Schuller, W., Hanf, V., and Hagenmeier, H. (2000). Input/output of estrogenic active compounds in a major municipal sewage plant in Germany. Chemosphere 40, 1131–1142.[CrossRef][ISI][Medline]

Körner, W., Spengler, P., Bolz, U., Schuller, W., Hanf, V., and Metzger, J. W. (2001). Substances with estrogenic activity in effluents of sewage treatment plants in southwestern Germany. 2. Biological analysis. Environ. Toxicol. Chem. 20, 2142–2151.[CrossRef][ISI][Medline]

Kuiper, G. G. J. M., Lemmen, J. G., Carlsson, B., Corton, J. C., Safe, S. H., van der Saag, P. T., van der Burg, B., and Gustafsson, J. A. (1998). Interaction of estrogenic chemicals and phytoestrogens with estrogen receptor beta. Endocrinology 139, 4252–4263.[Abstract/Free Full Text]

Larsson, D. G. J., Hällman, H., and Förlin, L. (2000). More male fish embryos near a pulp mill. Environ. Toxicol. Chem. 19, 2911–2917.[ISI]

Larsson, D. G. J., Kinnberg, K., Sturve, J., Stephensen, E., Skön, M., and Förlin, L. (2002). Studies of masculisation, detoxification, and oxidative stress responses in Guppies (Poecilia reticulata) exposed to effluent from a pulp mill effluent. Ecotoxicol. Environ. Saf. 52, 13–20.[CrossRef][ISI][Medline]

Lemmen, J. G., van den Brink, C. E., Legler, J., van der Saag, P. T., and van der Burg, B. (2002). Detection of oestrogenic activity of steroids present during mammalian gestation using oestrogen receptor {alpha}- and oestrogen receptor ß-specific in vitro assays. J. Endocrinol. 174, 435–446.[Abstract/Free Full Text]

Meerts, I. A. T., Letcher, R. J., Hoving, S., Marsh, G., Bergman, A., Lemmen, J. G., van der Burg, B., and Brouwer, A. (2001). In vitro estrogenicity of polybrominated diphenyl ethers, hydrocylated PBDEs, and polybrominated bisphenol A compounds. Environ. Health Perspect. 109, 399–407.[ISI][Medline]

Mellanen, P., Petänen, T., Lehtimäki, J., Mäkelä, S., Bylund, G., Hombolm, B., Mannila, E., Oikari, A., and Santti, R. (1996). Wood-derived estrogens, studies in vitro with breast cancer cell lines and in vivo in trout. Toxicol. Appl. Pharmacol. 136, 381–388.[CrossRef][ISI][Medline]

Metcalfe, C. D., Metcalfe, T. L., Kiparissis, Y., Koenig, B. G., Khan, C., Hughes, R. J., Croley, T. R., March, R. E., and Potter, T. (2001). Estrogenic potency of chemicals detected in sewage treatment plant effluents as determined by in vivo assays with japanese medaka (Oryzias latipes). Environ. Toxicol. Chem. 20, 297–308.[CrossRef][ISI][Medline]

Nimrod, A. C., and Benson, W. H. (1996). Environmental estrogenic effects of alkylphenol ethoxylates. Crit. Rev. Toxicol. 26, 335–364.[ISI][Medline]

Panter, G. H., Thompson, R. S., and Sumpter, J. P. (2000). Intermittent exposure of fish to estradiol. Environ. Sci. Technol. 34, 2756–2760.[CrossRef][ISI]

Pawlowski, S., Islinger, M., Völkl, A., and Braunbeck, T. (2000). Temperature-dependent vitellogenin-mRNA expression in primary cultures of rainbow trout (Oncorhynchus mykiss) hepatocytes at 14 and 18°C. Toxicol. in Vitro 14, 531–540.[CrossRef][ISI][Medline]

Rodgers-Gray, T. P., Jobling, S., Kelly, C., Morris, S., Brighty, G., Waldock, M. J., Sumpter, J. P., and Tyler, C. R. (2001). Exposure of juvenile roach (Rutilus rutilus) to treated sewage effluent induces dose-dependent and peristent disruption in gonadal duct develpoment. Environ. Sci. Technol. 35, 462–470.[CrossRef][ISI][Medline]

Rodgers-Gray, T. P., Jobling, S., Morris, S., Kelly, C., Kirby, S. J., Janbakhsh, A., Harries, J. E., Waldock, M. J., Sumpter, J. P., and Tyler, C. R. (2000). Long-term temporal changes in the estrogenic composition of treated sewage effluent and its biological effects on fish. Environ. Sci. Technol. 34, 1521–1528.[CrossRef][ISI]

Routledge, E. J., Sheahan, D., Desbrow, C., Brighty, G. C., Waldock, M., and Sumpter, J. P. (1998). Identification of estrogenic chemicals in STW effluent. 2. In vivo responses in trout and roach. Environ. Sci. Technol. 32, 1559–1565.[CrossRef][ISI]

Routledge, E. J., and Sumpter, J. P. (1996). Estrogenic activity of surfactants and some of their degradation products assessed using a recombinant yeast screen. Environ. Toxicol. Chem. 15, 241–248.[ISI]

Scholz, S., Braunbeck, T., and Segner, H. (1998). Viability and differential function of rainbow trout liver cells in primary culture: Coculture with two permanent fish cells. In Vitro Cell. Develop. Biol. Animal 34, 762–771.[ISI][Medline]

Spengler, P., Körner, W., and Metzger, J. W. (2001). Substances with estrogenic activity in effluents of sewage treatment plants in southwestern Germany. 1. Chemical analysis. Environ. Toxicol. Chem. 20, 2133–2141.[CrossRef][ISI][Medline]

Sumpter, J. P. (1995). Feminized responses in fish to environmental estrogens. Toxicol. Lett. 82/83, 737–742.[CrossRef]

Tavera-Mendoza, L., Ruby, S., Brousseau, P., Fournier, M., Cyr, D., and Marcogliese, D. (2002). Response of the amphibian tadpole Xenopus laevis to atrazine during sexual differentiation of the ovary. Environ. Toxicol. Chem. 21, 1264–1267.[ISI][Medline]

Ternes, T. A., Brenner-Weiß, G., Eggert, T., Müller, J., Kirschhöfer, F., Nusser, M., Wilken, R.-D., and Obst, U. (1999a). Method for the analysis of phytoestrogens and mycoestrogens in water using liquid chromatography-electrospray ionization/tandem mass spectrometry (LC-MSMS). Vom Wasser 93, 255–263.

Ternes, T. A., Kreckel, P., and Mueller, J. (1999b). Behaviour and occurrence of estrogens in municipal sewage treatment plants—II. Aerobic batch experiments with activated sludge. Sci. Total Environ. 225, 91–99.[CrossRef][ISI][Medline]

Ternes, T. A., Stumpf, M., Mueller, J., Haberer, K., Wilken, R.-D., and Servos, M. (1999c). Behaviour and occurrence of estrogens in municipal sewage treatment plants—I. Investigations in Germany, Canada and Brazil. Sci. Total Environ. 225, 81–90.[CrossRef][ISI][Medline]

Thomas, K. V., Hurst, M. R., Matthiessen, P., and Waldock, M. J. (2001). Characterisation of estrogenic compounds in water samples collected from United Kingdom estuaries. Environ. Toxicol. Chem. 20, 2165–2170.[CrossRef][ISI][Medline]

Tilton, F., Benson, W. H., and Schlenk, D. (2002). Evaluation of estrogenic activity from a municipal wastewater treatment plant with predominantly domestic input. Aquat. Toxicol. 61, 211–224.[CrossRef][ISI][Medline]

Toppari, J., Larsen, J. C., Christiansen, P., Giwercman, A., Grandjean, P., Guillette, L. J., Jr., Jégou, B., Jensen, T. K., Jounnet, P., Keiding, N., et al. (1996). Male reproductive health and environmental xenooestrogens. Environ. Health Perspect. 104, 741–776.[ISI][Medline]

Tremblay, L., and Van der Kraak, G. (1999). Comparison between the effects of the phytosterol ß-sitosterol and pulp and paper mill effluent on sexually immature rainbow trout. Environ. Toxicol. Chem. 18, 329–336.[ISI]

Tyler, C. R., Beresford, N., Van der Woning, M., Sumpter, J. P., and Thorpe, K. (2000). Metabolism and environmental degradation of pyrethroid insecticides produce compounds with endocrine activites. Environ. Toxicol. Chem. 19, 801–809.[CrossRef][ISI]

Willingham, E., and Crews, D. (1999). Sex reversal effects of environmentally relevant xenobiotic concentrations on the Red-Eared Slider Turtle, a species with temperature-dependent sex determination. Gen. Comp. Endocrinol. 113, 429–435.[CrossRef][Medline]

Zhan, J., Fahimi, D., and Völkl, A. (1997). Sensitive nonradioactive dot blot/ribonuclease protection assay for quantitative determination of mRNA. Biotechnol. 22, 500–505.





This Article
Abstract
FREE Full Text (PDF)
All Versions of this Article:
75/1/57    most recent
kfg162v1
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (7)
Disclaimer
Request Permissions
Google Scholar
Articles by Pawlowski, S.
Articles by Braunbeck, T.
PubMed
PubMed Citation
Articles by Pawlowski, S.
Articles by Braunbeck, T.