Microsoft word - sonzogni 05-bep-01 wr04r004.doc

Occurrence of Estrogenic Endocrine Disruptors in Groundwater
Wisconsin State Laboratory of Hygiene Investigators: William Sonzogni PhD, Director Jocelyn Hemming, PhD, Assistant Scientist Miel Barman, Environmental Toxicologist Steve Geis, Chemist Supervisor July 1, 2004-June 30, 2006 This project was supported, in part, by General Purpose Revenue funds of the State of Wisconsin to the University of Wisconsin System for the performance of research on groundwater quality and quantity. Selection of projects was conducted on a competitive basis through a joint solicitation from the University and the Wisconsin Departments of Natural Resources; Agriculture, Trade and Consumer Protection; Commerce; and advice of the Wisconsin Groundwater Research Advisory Council and with the concurrence of the Wisconsin Groundwater Coordinating Council. TABLE OF CONTENTS
List of Figures and Tables . Page 3
Project Summary . Page 4
Introduction . Page 6
Procedures and Methods . Page 7
Results and Discussion . Page 8
Conclusions and Recommendations . Page 11
References . Page 11
Appendix A . Page 13
Appendix B . Page 13

LIST OF FIGURES AND TABLES
Figure 1 Escreen results from Facility L1109. Page 9
Figure 2 Escreen results from Facility P1105 . Page 9
Figure 3 Escreen results from Facility W1112 . Page 10
Figure 4 Escreen results from Facility B1029. Page 10
Figure 5 Escreen results from Facility J1118 . Page 10
Figure 6 Escreen results from septic effluents . Page 11
Table 1 QC Sample Results . Page 11
Table 2 Well information . Page 13

Project Summary

Occurrence of Estrogenic Endocrine Disruptors in Groundwater Project ID: WR04R004 Wisconsin State Laboratory of Hygiene Investigators: William Sonzogni PhD, Director Jocelyn Hemming, PhD, Assistant Scientist Miel Barman, Environmental Toxicologist Steve Geis, Chemist Supervisor July 1, 2004-June 30, 2006 Background:

Concern has emerged about a group of trace organic compounds identified in the aquatic environment
which might affect reproduction and development of wildlife species and humans due to endocrine
disruption (Colborn et al., 1996; Roefer et al., 2000). Studies in recent years have documented a high
occurrence of these endocrine disrupting compounds (EDCs) in aquatic ecosystems which may have
serious implications for groundwater quality (e.g., Kolpin et al., 2002). As the hydraulic demand on
groundwater supplies increases, resulting in greater groundwater draw downs, the potential for EDC
contaminated surface water input to groundwater increases, especially in areas where high capacity wells
are located near surface waters. An additional route for entrance of EDCs into groundwater is through the
leaching of septic system effluents. Areas such as non-sewered subdivisions may have an increased
susceptibility to contamination of the groundwater by EDCs.
Compounds with endocrine disrupting activity include both anthropogenic chemicals produced
industrially (such as surface active agents, pesticides, polychlorinated biphenyls (PCBs), plasticizers, food
additives, birth control pills, herbal supplements and cosmetics) and natural occurring compounds (such
as sex steroids, plant-produced estrogens and heavy metals) (National Research Council, 1999). EDCs are
used in large quantities by consumers and industry. Domestic and industrial wastewater and agricultural
run-off are recognized as the major sources of EDCs. Due to their physical-chemical properties and
partial resistance to biotransformation, EDCs have been detected not only in wastewater effluents, but
also at low concentrations in surface and groundwaters used as a source for water supply, and at very low
concentrations in tap water samples (Stumpf et al., 1996; Ternes, 1999; Baronti et al., 2000).
Objectives:
1.
To test high capacity municipal water supply wells located near surface waters impacted by industrial and municipal effluents will be tested for estrogenic EDCs. To sample wells constructed to monitor groundwater close to non-conventional small scale on site waste disposal systems will be tested. The breast cancer cell line (MCF-7) assay (E-screen) will be used to evaluate groundwater samples and septic samples for estrogenic activity.
Methods:

High capacity wells that may be impacted by nearby surface waters were selected from five Wisconsin
communities. Associated drinking water and WDNR personnel were enlisted to perform sample
collection. Each well and associated surface water were sampled four times per year to evaluate seasonal variability. A total of 21 samples were collected from traditional and non-conventional septic systems throughout Southeast Wisconsin. Soil pore water and groundwater samples were collected from lysimeters and monitoring wells installed beneath and adjacent to two of these systems. Water samples were passed through a C18 solid phase extraction disk (EmporeTM ). Compounds were eluted from the disks and concentrated to 1.5 ml. MCF-7 breast cancer cells were exposed to the extract.
Cell proliferation was measured after five days of exposure using the SRB colorimetric protein assay.

Results and Discussion:

All surface waters contained some levels of estrogenic EDC activity. Water from high capacity wells did
not contain any measurable estrogenic EDC activity. Estrogenic activity was detected in 20 of 21 septic
effluent samples, although concentrations were markedly reduced in systems utilizing either sand
filtration or aerobic pretreatment as compared to traditional systems. Although low levels of activity have
been detected in soil water directly beneath one septic system, no estrogenic activity was found in
groundwaters in this study.
Conclusions/Implications/Recommendations:

There appears to be no infiltration of estrogenic endocrine disrupting chemicals from the surface waters
into the associated ground waters. Advanced pretreatment technologies (aerobic, sand filtration) appear
to be quite effective at removing estrogenic compounds from septic effluent. Additional removal of
EDCs occurs in unsaturated soils beneath septic leach fields; no EDCs were detected in groundwater
beneath two systems without advanced pretreatment.

Related Publications:
Publication in progress: A manuscript based on the results from the septic effluents is in preparation in
collaboration with Jeff Wilcox and will be submitted to the Journal of Environmental Quality.
Key Words:
Endocrine disruptors, high capacity wells, septic, non-sewered subdivisions, E-Screen.

Funding:
University of Wisconsin System portion of the Wisconsin Groundwater Research Program through the
Water Resources Institute.
INTRODUCTION

During the past decade, concern has emerged about a group of trace organic compounds identified in the
aquatic environment which might affect reproduction and development of wildlife species and humans
due to endocrine disruption (Colborn et al., 1996; Roefer et al., 2000). Studies in recent years
documented a high occurrence of these endocrine disrupting compounds (EDCs) in aquatic ecosystems
which may have serious implications for groundwater quality (e.g., Kolpin et al., 2002). As the hydraulic
demand on groundwater supplies increases, resulting in greater groundwater draw downs, the potential for
EDC contaminated surface water input to groundwater increases, especially in areas where high capacity
wells are located near surface waters. An additional route for entrance of EDCs into groundwater is
through the leaching of septic system effluents. Areas such as non-sewered subdivisions may have an
increased susceptibility to contamination of the groundwater by EDCs. The proliferation of non-
conventional small scale on site waste disposal systems in rural areas may also result in the entrance of
EDCs to groundwater. It is expected that such non-conventional systems will usually be installed in
vulnerable groundwater settings, such as areas of shallow bedrock or high water tables, where
conventional on-site septic systems would not be allowed.
Compounds with endocrine disrupting activity include both anthropogenic chemicals produced
industrially (such as surface active agents, pesticides, polychlorinated biphenyls (PCBs), plasticizers, food
additives, birth control pills, herbal supplements and cosmetics) and natural occurring compounds (such
as sex steroids, plant-produced estrogens and heavy metals) (National Research Council, 1999). EDCs are
used in large quantities by consumers and industry. Domestic and industrial wastewater and agricultural
run-off are recognized as the major sources of EDCs. Due to their physical-chemical properties and
partial resistance to biotransformation, EDCs have been detected not only in wastewater effluents, but
also at low concentrations in surface and groundwaters used as a source for water supply, and at very low
concentrations in tap water samples (Stumpf et al., 1996; Ternes, 1999; Baronti et al., 2000).
The objectives for this project were:
1.
To test high capacity municipal water supply wells located near surface waters impacted by industrial and municipal effluents for estrogenic EDCs. To sample wells constructed to monitor groundwater close to non-conventional small scale on site waste disposal systems as allowed by recently enacted Comm 83 rules for construction of septic systems and/or wells which already are suspected of having septic influences. We used the breast cancer cell line (MCF-7) assay (E-screen) to evaluate >60 groundwater samples for estrogenic activity (breast cancer cells grow in the presence of estrogen or compounds that mimic estrogen). High capacity wells expected either through modeling or direct empirical evidence to be impacted by nearby surface waters were chosen in consultation with DNR and USGS hydrologists. Drinking water personnel from selected communities were enlisted to perform sample collection. Each facility and associated surface water was sampled four times per year to evaluate seasonal variability. We sampled wells from five Wisconsin communities that utilize high capacity wells for their drinking water. Septic samples were collected from a housing subdivision that utilized unique septic systems. Approximately 31 samples were collected, either from monitoring wells or from cooperative homeowners. Water samples were passed through a C18 solid phase extraction disk (EmporeTM ). Compounds were then eluted from the disks and concentrated to 1.5 ml. MCF-7 breast cancer cells were then exposed to the extract. Cell proliferation was measured after five days of exposure using the SRB colorimetric protein assay. The WSLH has successfully developed the E-screen assay for use on water samples, and find it to be a highly effective and sensitive tool for detecting estrogenically active substances.
PROCEDURES AND METHODS

High capacity wells were sampled from five different Wisconsin municipalities along with their nearby
surface water. All surface water samples were taken upstream from the well sites. Specific information on
each sampling site can be found in Appendix B. Well water from two of the municipalities was further
treated before distribution to the public. These post-treatment plant waters were also collected and tested.
Each municipality was sampled four times to assess seasonal variation. A total of 80 well samples, 12
post treatment drinking waters, and 24 surface waters were collected and assayed for endocrine disruption
activity. Collections of these waters occurred in November and December 2004, February and March
2005, May and June 2005, and August and September 2005.
Also, 21 septic-effluent samples were collected at 15 different residences in southeastern Wisconsin
between April 2005 and February 2006. The types of septic systems included six aerobic treatment units,
seven sand filters, and eight without any secondary pretreatment (Wisconsin Mound). Samples were
collected before and after pretreatment from six of the systems using sand filtration. Aerobic treatment
units and single-pass sand filters are used routinely in Wisconsin and elsewhere when site conditions are
not considered suitable—due to shallow bedrock, high seasonal saturation, or limited infiltration
capacity—for traditional systems. All samples were collected using peristaltic pumps and new Teflon
tubing. In addition, soil-water and groundwater samples were collected beneath and immediately
adjacent to two septic systems that utilize Wisconsin Mound distribution using previously-installed
lysimeters and monitoring wells.
Water from the high capacity wells, nearby surface waters and septic systems was collected in 1 liter
amber glass bottles. Samples were shipped overnight on ice to the laboratory. Samples (volumes of 1-
2L) were extracted through a C18 disk, which was then eluted with solvents to capture the estrogenic
chemicals. The extraction solvents were transferred into ethanol for a final volume of 1.5 ml. The extraction method was evaluated continually throughout the study using blanks, spikes, and duplicates. To ensure that the sampling method was not adding activity, sampling blanks were run using Type I water. Method and matrix spikes using 17β-estradiol were analyzed to ensure the extraction method was recovering the suspected EDCs. Interferences due to sample constituents, especially organic matter, were evaluated using matrix spikes. Duplicates were used to evaluate the consistency of the extraction procedures and assay. All extracts were tested using the in vitro breast cancer cell proliferation assay known as the E-Screen. This assay utilizes the MCF-7 breast cancer cell line. The breast cancer cells proliferate in response to the presence of estrogen or chemicals that mimic estrogen. MCF-7 cells were grown in Dulbecco's Modified Eagle's Medium with 5% fetal calf serum at 37oC and 6.5% CO2 and were sub-cultured every 7 days. To begin the assay, MCF-7 cells were trypsinized, counted with flow cytometry and plated into 24-well tissue culture plates at approximately 25,000 cells per well. Twenty-four hours after seeding, the media was removed and experimental media was added. The experimental media was Dulbecco's Modification of Eagles Medium without phenol red. The 5% fetal bovine serum used in this media is stripped of steroids with a charcoal dextran (CD) stripping procedure (referred to as CD-media). For plates containing the standards, the MCF-7 cells were exposed to 15 concentrations of estradiol ranging from 1 x 10-13 to 1 x 10-8 M. All treatments were done in quadruplicate and three plates were used to obtain a complete standard curve. Four control wells (CD-media only) were included on each plate. For plates on which samples were assayed, 0.1 ml of the water extract (in ethanol) was added to 9.9 ml of CD-media. A 50% dilution series was then made for a total of 5 concentrations. Each concentration was applied to four wells. In one of those wells, 1 x 10-9 M estradiol was added as a positive control. Again, each plate contained 4 control wells. After five days, cell proliferation was measured by the SRB protein assay: Cells were fixed with 10% tricholoracetic acid, rinsed and allowed to dry. The SRB dye was added to each well, followed by a rinse with 1% glacial acetic acid and allowed to dry. The dye was resuspended with 10mM Tris buffer at pH 10.5. The samples were read at 515 nM with a Molecular Devices microplate reader. To determine the estradiol equivalents (Eeq) of the samples, the standard curve was fit with a 4-parameter logistic equation, calculated by the Softmax PRO v. 2.6 analytical software package for the microplate reader. Estradiol equivalents for each sample was calculated by inserting the absorbance readings from the sample into the equation derived from the standard curve. Negative and positive wells were evaluated to ensure that the assay is running correctly. Low growth in the positive control wells may indicate the presence of substances that impede the normal growth of the cells. The limit of detection (LOD) was determined to be 0.02 ng/L Eeq and the limit of quantification (LOQ) was 0.04 ng/L Eeq. RESULTS AND DISCUSSION

Two high capacity wells from five different municipalities were sampled four separate times. Well water
from two of the municipalities was further treated before distribution to the public. These post-treatment
plant waters were also collected and tested. Nearby surface water samples were also collected. Eighty
well samples, twelve post treatment samples and twenty-four surface water samples were collected
(Figures 1 through 5). Septic sampling occurred three times, with a total of 21 septic-effluent, four
monitoring-well, and six soil-lysimeter samples being collected. Quality assurance samples include five
matrix spikes, four travel blanks, three duplicates, two lab water blanks and two lab water spikes. In all,
121 samples have been analyzed for this study. The first round of high capacity well sampling had five of
the six surface waters showing estrogenic activity which ranged from 0.05-0.09 ng/L with the sixth site
having no estrogenic activity, while all of the well samples showed no estrogenic activity. The second
round of sampling, for the high capacity wells had all six of the surface waters exhibiting estrogenic
activity at an elevated rate from the first sampling. This activity ranged from 0.04-0.91 ng/L. The well
waters from this round of sampling indicated activity levels that ranged from no detect (< 0.02 ng/L) to
less than the LOQ (< 0.04 ng/L). The third round of sampling for the high capacity wells had all six
surface waters with estrogenic activity ranging from 0.05-0.24 ng/L, while all of the well samples were
below LOD. The fourth round of sampling for the high capacity wells had all six surface waters showing
estrogenic activity which ranged from 0.07-0.66 ng/L. One of the sites showed estrogenic activity in both
their well waters and post treatment waters. These activity levels ranged from <LOQ to 0.05 ng/L. The
rest of the well waters did not exhibit estrogenicity. The levels of activity found in these samples are at or
near the level of quantification (0.04 ng/L) and may be a result of sampling error or variation in the assay.
Seasonality may have played a role in the increase in estrogenic endocrine disruptor activity in the second
round of sampling. This round of sampling occurred during February and March, which follows a long
period of cold temperatures. Biological activity of surface waters may lower as the temperature
decreases. This phenomenon would explain the higher activity level if bacterial degradation of chemicals
was happening at a slower rate.
Estrogenic activity in the 21 septic effluent samples ranged from <LOQ to 192.5 ng/L (Figure 6). The highest estrogenic activity (192.5 ng/L) was detected in a malfunctioning aerobic treatment unit. Excluding this data point, activity in samples collected following sand filtration and aerobic activity ranged from 0.058-3.81 ng/L (mean=0.76 ng/L) and 0.07-6.06 ng/L (mean=1.6 ng/L), respectively. Activity in eight samples with no advanced pretreatment—including the samples collected prior to sand filtration—ranged from 11.6-50.0 ng/L with a mean of 23.7 ng/L. These results indicate that concentrations of estrogenically-active substances can be significantly reduced in septic systems utilizing advanced pretreatment. As a comparison, a national survey of eight municipal wastewater treatment plants in which the E-screen was used to evaluate removal during secondary treatment indicated estrogenic activity ranging from 17 to 95 ng/L Eeq after primary treatment and 0.62 to 7.9 ng/L Eeq after secondary treatment (Drewes et al., 2005). Thus, removal appears fairly similar between municipal treatment and systems with advanced pretreatment.
Of the two systems that were instrumented with soil lysimeters and monitoring wells, one system showed
effluent activity at 25.50 ng/L while the corresponding soil-water samples had no activity. The second
system had an activity level of 19.57 ng/L in the effluent while activity in the three soil-water samples
ranged from 0.28-0.38 ng/L. No estrogenic activity was detected in any of the groundwater samples.
Figure 1
Round 1 (December) Round 4 (September) Sa mple type
Round 1 (December) Round 4 (September) ng/L 0.80
(
s
nt
le 0.60
Sample type
Round 1 (December) Round 4 (September) 0.00 0.00 0.00 0.04 0.00 0.00 0.00 0.05 Sa mple type
Round 1 (November) Round 2 (February) Sa m ple type
Round 1 (December) Round 4 (September) iva
u
q 0.40
l E
Sa mple type
Figure 6 Total estrogenicity in 21 septic-effluent samples
Note that sample 16 was collected from a system that was not functioning properly.

Table 1 QC Sample Results

sample QC
sample QC
sample QC
sample QC
Sampling
Sampling
Sampling
Sampling
Sampling
Lab Spike
Duplicate
Lab Blank
Travel Blank
Round Round
Recovery
recovery
recovery
Round Round
1st ND 1st ND 2nd 95 2nd 85 2nd 104
1st ND 2nd ND 3rd 88 2nd 81 3rd 127
2nd ND 3rd ND
3rd 77 4th 71

CONCLUSIONS AND RECOMMENDATIONS
With the exception of one sample, all the surface waters tested in this study exhibited estrogenic activity
although this activity was at very low levels (all < 1 ng/L Eeq). When samples from high capacity wells
that had the potential to be contaminated by the surface waters were tested, the vast majority of samples
were below the detection limits of the assay. Therefore there is no evidence that the surface water is
infiltrating into the groundwater. With respect to the pretreatment of septic effluent, both sand filtration
and properly functioning aerobic treatment resulted in markedly reduced estrogenic activity in the
samples relative to septic effluents with no pretreatment. However, estrogenic activity was reduced from
untreated septic effluent in the soil column. Samples collected with a lysimeter indicated greater than
98% removal and removal below detection levels in samples collected from the monitoring wells.

REFERENCES
Baronti, C., Curini, R., D'Ascenzo, G., Di Corcia, A., Gentili, A. and Samperi, R. (2000). Monitoring
Natural and Synthetic Estrogens at Activated Sludge Sewage Treatment Plants and in a Receiving River
Water. Environmental Science & Technology 34, 24, 5059-5066.
Colborn, T., Dumanoski, D., and Myers, J. P. (1996). Our Stolen Future: are we threatening our fertility,
intelligence, and survival? A scientific detective story. New York, Dutton.
Drewes, J.E., Hemming, J., Ladenburger, S.J., Schauer, J., and Sonzogni, W. (2005). An Assessment of
Endocrine Disrupting Activity Changes during Wastewater Treatment through the Use of Bioassays and
Chemical Measurements. Water Environment Research, 77,1, 12-23.
Kolpin, D.W., Furlong, E.T., Meyer, M.T., Thurman, E.M., Zaugg, S.D., Barber, L.B. and Buxton, H.T.
(2002). Pharmaceuticals, Hormones, and Other Organic Wastewater Contaminants in U.S. Streams, 1999-
2000: A National Reconnaissance. Environmental Science and Technology, 36, 1202-1211.
National Research Council (1999), Hormonally Active Agents in the Environment. National Academy
Press, Washington, D.C.
Stumpf, M, Ternes, T. A., Haberer, K., and Baumann, W. (1996), Nachweis von natuerlichen und
synthetischen Oestrogenen in Klaeranlagen und Fliessgewaessern. Vom Wasser 87, 251-261.
Ternes, T. A., Kreckel, P., and Mueller, J. (1999), Behavior and occurrence of estrogens in municipal
sewage treatment plants – II. Aerobic batch experiment with activated sludge. The Science of the Total
Environment 225, 91-99.
Ternes, T. A., Stumpf, M., Mueller, J., Haberer, K., Wilken, R.-D., and Servos, M. (1999), Behavior and
occurrence of estrogens in municipal sewage treatment plants – I. Investigations in Germany, Canada and
Brazil. The Science of the Total Environment 225, 81-90.
APPENDIX A: AWARDS, PUBLICATIONS, REPORTS, PATENTS AND PRESENTATIONS
Presentation:
Wilcox, J.D., J.M. Bahr, C.J. Hedman, and K.R. Bradbury, 2006, Investigation of source concentrations
and transport of hormones and pharmaceuticals beneath on-site wastewater treatment systems: GSA
Annual Meeting Abstracts with Programs, vol. 38.

Barman, M., Sonzogni, W., Hemming, J., Geis, S. 2005, Occurrence of Estrogenic Endocrine Disruptors
in Groundwater. Midwest Regional Society of Environmental Toxicology and Chemistry. Platform
presentation April 2005, Madison WI.
Publication in progress:
A manuscript based on the results from the septic effluents is in preparation in collaboration with Jeff
Wilcox and will be submitted to the Journal of Environmental Quality.

APPENDIX B:
Table 2 Well Information
Facility Code
Well distance
from surface
water (ft)
Well depth (ft)
Municipal
discharges
Industrial
discharges
NA = information not provided by municipality

Source: http://www.wri.wisc.edu/Downloads/PartnerProjects/FinalReports/Final_WR04R004.pdf