k\v/J
^ PRO^°
Workshop Abstracts
Endocrine Disrupters
Program Review Workshop
October 29-31,2002
Research Triangle Park, North Carolina
-------
Endocrine Disrupters Program Review Workshop
Table of Contents
Introduction vii
PLATFORM PRESENTATIONS
Background
Overview of EPA's Regulatory Program 5
GaryE. Timm, James P. Kariya, Leslie W. Touart, L. GregSchweer, Jane S. Smith
Screening and Testing Assays
Mosquitofish as a Model Organism for the Study of Endocrine-Disrupting Chemicals 9
Robert A. Angus, R. Douglas Watson, W. Mike Howell, Ronald L. Jenkins
Identification of Endocrine Disrupters Using a Short-Term Reproduction Assay With the Fathead Minnow 11
GT.Ankley, KM. Jensen, M.D. Kahl, J.J. Korte, E.A. Makynen, M. W. Harming
Environmentally Mediated Endocrine Disruption inEstuarine Crustaceans: Meiobenthos-Based Life-Cycle
Assays for Evaluating Risks of Reproductive and Endocrine Toxicity 12
G. Thomas Chandler, T. Gary, D. Volz, D. Block, J. Ferry, G Scott, J. Quattro, E. Wirth, M. Fulton
Complementary In Vitro and/n Vivo Rodent Assays 13
Kirsten C. Fertuck, Jason B. Matthews, Tim R. Zacharewski
Endocrine Disrupters—Tiered Screening and Testing: Filling Key Data Gaps 14
Leon Earl Gray, Jr., Joseph Ostby, Vickie Wilson, Christy Lambright, Kathy Bobseine, Phillip Hartig,
Andrew Hotchkiss, Cindy Wolf, Johnathan Furr, Mathew Price, Louise Parks, Ralph Cooper,
Tammy Stoker, Susan Laws, Sig Degitz, Kathleen Jensen, M. Kahl, Joseph Korte,
Elizabeth Makynen, Joseph Tietge, Gerald Ankley
TheMedaka: An/n Vivo Model for Detection of Adverse Effects of Endocrine Disrupters 15
David E. Hinton, Seth W. Kullman, Shosaku Kashiwada, David C. Bencic, Yuko Wakamatsu, Kenji Ozato,
Cory S. Koger, Corrine R. Davis, Swee J. Teh
Validation of an/w Vivo Protocol To Evaluate the Effects of Environmental Chemicals on Female Pubertal
Development and Thyroid Function 16
Susan C. Laws, Janet M. Ferrell, Tammy E. Stoker, Jerome M. Goldman, L. Earl Gray, Jr., Ralph L. Cooper
Reproductive and Developmental Screening Protocols for Endocrine Disrupters Using Estuarine Crustaceans 17
Charles L. McKenney, Jr.
Thyroid Axis Inhibition in Xenopus laevis: Development of an Amphibian-Based Screening Assay for
Thyroid Disruption 18
J.E. Tietge, G. W. Holcombe, KM. Flynn, J.J. Korte, R.C. Kolanczyk, PA. Kosian, S.J. Degitz
Novel Cheminformatics and Pattern Recognition Tools Useful for Risk Assessment and Regulatory Control 19
William J. Welsh
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disrupters Program Review Workshop
Developmental Exposures
Short-Term Exposure to an Environmental Mixture of PHAHs: Dose-Additive Effects on Serum Thyroxine 23
KevinM. Crofton, JoanM. fledge, David G Ross, MichaelJ. DeVito
Effects of Endocrine-Disrupting Compounds on Mammary Tissue Development 24
Suzanne E. Fenton, Christine C. Davis, Jennifer Rayner, Geri L. Youngblood, Judy Moon
Analysis of Sensitive Developmental Stages in Birds to EDCs 25
D. Michael Fry, James R. Millam,A.L. Erichsen, L.A. Preston, C.B. Craig-Veit,
A.E. Quigliano, M.R. Viant
Endocrine Disruption in Pubertal Rhesus Monkeys: Growth and Sexual Maturation 26
Mari S. Golub, BillL. Lasley, and Alice F Tarantal
The Effects of In Utero and Lactational Exposure to Genistein or Daidzein on Reproductive Development
inFVB/NMice and on Occurrence of Mammary Tumors inMMTV-weMTransgenic Mice 27
Claude Hughes, Vicki Davis, Firdos Shaikh, Michael Villegas, Arline Ho, Warren Foster
The Michigan PBB Cohort 20 Years Later: Endocrine Disruption? 28
Michele Marcus, Heidi Michels Blanch, Paige Tolbert, Carol Rubin, Alden Henderson,
Lorraine Cameron, Vicki Hertzberg
Effects of Trihalomethanes on Pregnancy Maintenance in Rats 29
Michael G Narotsky, Deborah S. Best, Jerome M. Goldman, Ashley S. Murr, Susan R. Bielmeier
Assessing Consequences of Embryonic Exposure to Methoxychlor: Neuroendocrine and Behavioral Measures 30
Mary Ann Ottinger, Julie Hazelton, Julie Wu, Michael Ruscio, Nichola Thompson,
Michael Quinn, Mahmoud Abdelnabi
Endocrine Disrupters and Testis Development 31
Michael K. Skinner
Effects of Early Developmental Exposure to Endocrine-Disrupting Chemicals on Reproductive Function
in the Adult Male: RabbitModel 32
D.N. Rao Veeramachaneni, Colleen M. Kane, Ty T. Higuchi, Jennifer S. Palmer,
Ginger E. Sammonds, K.-Y. Francis Pau
Exposure, Risk Assessment, and Risk Management
Cross Species Mode of Action Information Assessment for Bisphenol A 35
Susan Y. Ruling, Babasaheb Sonawane
Overview of Intramural Exposure Research Program 36
V. Ross Highsmith, Greg Toth, Lawrence Burns
Development of Molecular Diagnostic Indicators of Exposure to Estrogenic and Androgenic Endocrine-
Disrupting Chemicals 37
D.L. Lattier,A.Miracle, T.V.Reddy, J.L Lazorchak, G.P. Toth
Intramural Research on Risk Management of Wastewater Treatment Sources 38
Marc A. Mills, Gregory D. Sayles, Andrew P. Avel
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disrupters Program Review Workshop
Endocrine-Disrupting Chemicals Related to Feminization of Males 39
Deb das Mukerjee
Overview of the Intramural Risk Management Research Program 40
Gregory D. Sayles, Andrew P. Avel
Field Studies
Frog Deformities: Role of Endocrine Disrupters During Development 43
DavidM. Gardiner, Aristocle Ndayibagira, Felix Grun, David Hoppe, Bruce Blumberg
The Mechanisms and Effects of Endocrine Disruption on Infertility in the Bonnethead Shark on Florida's
Gulf Coast 44
Charles A. Manire, Jim Gelsleichter, LoisE.L. Rasmussen, Enric Cortes, Ian Tebbett,
David L. Hess, Nancy Szabo
Endocrine Disruption in Marine Gastropods by Environmental Chemical Mixtures 46
Patricia D. McClellan-Green, Eva Oberdorster
Effects of PCBsonHerring Gulls in the Field and Chickens in the Laboratory 47
P.M. Anne McNabb
Effects of Exposure to Environmental Estrogens on Reproductive Parameters in a Marine Fish,
Tautogolabrus adspersus 48
Lesley J. Mills, RuthE. Gutjahr-Gobell, Doranne Borsay Horowitz, Gerald E. Zaroogian
Mechanisms of Action
Polymerase Chain Reaction-Differential Display (PCR-DD) Identifies a Subset of Stage-Dependent
and Toxicant-Regulated Genes During Spermatogenesis: The Shark Testis Model 51
Gloria V. Callard, Chunhua Wang
Mode and Mechanism of Action of the Chlorotriazine Herbicides 52
Ralph L. Cooper, Susan C. Laws, Parikshit C. Das, Jerome M. Goldman,
Michael G Narotsky, Tammy E. Stoker
Gestational and Lactational Exposure of Male Mice to Diethylstilbestrol Causes Long-Term Effects on the Testis,
Sperm Fertilizing Ability In Vitro, and Testicular Gene Expression 53
MarkR. Fielden, Robert G Halgren, CoraJ. Fong, Christophe Staub, Larry Johnson,
Karen Chou, Tim R Zacharewski
Development, Application, and Validation of a Sheepshead Minnow Estrogen-Responsive cDNA Macroarray 54
Michael J. Hemmer, Leroy C. Folmar
Interaction of Estrogen and TCDD in an AvianModel 55
Bill L. Lasley, Rebecca Stanton
Masculinization of Invertebrates by Endocrine Toxicants: Mechanisms and Environmental Significance 56
Gerald A. LeBlanc, Meredith P. Goading, Allen W. Olmstead, Xueyan Mu
Dioxin (TCDD) Disrupts Steroid ActioninanEndometriosisModel 57
Kevin G. Osteen and Kay Ion L. Bruner-Tran
The Office of Research and Development's National Center for Environmental Research iii
-------
Endocrine Disrupters Program Review Workshop
Chemical Interference With Non-Genomic Steroid Actions: A Novel Mechanism of Endocrine Disruption 58
Peter Thomas
Environmental Androgens and Anti-Androgens 59
Vickie S. Wilson
POSTER PRESENTATIONS
Screening and Testing Assays
Computational Models for the Rapid Prediction of Ligand Binding Affinities to the Androgen Receptor:
Effects of Mutations on Ligand-Receptor Specificity 65
NiAi, Seong-Jae Yu, Robert Kirk Delisle, William J. Welsh
A Computer-Docking Study of the Binding of Poly cyclic Aromatic Hydrocarbons and Their Metabolites
to the Ligand-Binding Domain of the Estrogen Receptor 66
Katrina W. Brown, Stephen Little, James Rabinowitz
Rainbow Trout Androgen Receptor Alpha and Human Androgen Receptor: Comparisons
in the COS Whole-Cell Binding Assay 67
Mary C. Cordon, L. Earl Gray, Jr., Vickie S. Wilson
Semiquantitative Laser-Scanning Confocal Microscopy (LSCM): Assessing Crustacean Egg
Quality for EDC Screening 68
G Thomas Chandler, David C. Volz
Determination of Structural Requirements for Activation of the Clearance Mechanism of Environmental
Pollutants and Xenobiotic Chemicals by the Pregnane Xenobiotic Receptor: Species-to-Species Extrapolation 69
Vladyslav Kholodovych, William J. Welsh
Utility of In Vitro Assays To Screen in Environmental Mixtures for EDC Activities 70
Christy R. Lambright, L. Earl Gray, Jr., GeraldAnkley, LouJ. Guillette, Edward Orlando, Vickie S. Wilson
Endocrine Disrupters From Combustion and Vehicular Emissions: Identification and Source Nomination 71
Clyde Owens, Brian Gullett, Jeff Ryan, Paul Lemieux, Carolyn Acheson, Michael DeVito,
Christy Lambright, Vickie Wilson, L. Earl Gray, Jr., James Rabinowitz, Sukh Sidhu,
Richard Striebich, Joy Klosterman
Endocrine-Mediated Effects of UV-A Irradiation on Grass Shrimp (Palaemonetespugio) Reproduction:
Implications for EDC Screening 72
David C. Volz, Edward F. Wirth, Michael H. Fulton, Geoffrey I. Scott,
David S. Block, G. Thomas Chandler
Enzyme-Linked Immunosorbant Assays for Detection of Crustacean Vitellin and Ecdysteroids:
Development and Validation for EDC Screening 73
David C. Volz, David S. Block, Adriana C. Bejarano, G. Thomas Chandler
Reproductive Alterations in the Grass Shrimp (Palaemonetespugio) Following Pesticide Exposure 74
Edward F. Wirth, David C. Volz, Michael H. Fulton, Geoffrey I. Scott, G Thomas Chandler
Influence of the Structural Diversity of Data Sets on the Statistical Quality of Three-Dimensional Quantitative
Structure-Activity Relationship (3D-QSAR) Models: Predicting the Estrogenic Activity of Xenoestrogens 75
Seong Jae Yu, Susan M. Keenan, Weida Tong, William J. Welsh
IV The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disrupters Program Review Workshop
Developmental Exposures
Neurophy siological Consequences in Hippocampus as a Function of Developmental Hypothyroidism 79
Mary Gilbert, Li Sui
Retinoids Induce All the Limb Malformations Observed in Wild Populations of Deformed Frogs 80
Aristocle Ndayibagira, Felix Grun, Bruce Blumberg, David M. Gardiner
Low Doses of Estrogenic Chemicals Like Diethylstilbestrol During Development Result in Permanent
Alterations in the Reproductive Tract 81
Retha R. Newbold, Elizabeth Padilla-Banks, Wendy N. Jefferson
Field Studies
Endocrine-Disrupting Chemicals and Thyroid Outcomes 85
Henry A. Anderson, Lawrence P. Hanrahan, Vicky Per sky, Mary Turyk, Dyan Steenport,
Robert Chatterton, Lynda Knobeloch, Livia Navon, the Great Lakes Consortium
Latent Effects of Gestational Exposure to Heptachlor 86
Dean Baker, Ulrike Luderer, Sastry Gollapudi, Haiou Yang
Endocrine-Disrupting Pesticides and Neurodevelopmental Outcomes in Farmworker Children, Salinas Valley 87
Brenda Eskenazi, Asa Bradman, Laura Fenster, Dana Barr, Jonathan Chevrier,
Martha Harnly, Abbey Alkon, Marcy Warner
Persistent Organic Pollutants and Endometriosis Risk 88
Victoria L. Holt, Stephen M. Schwartz, Larry L. Needham, Dana B. Barr
Biogeographic Variation in the Model Organism Palaemonetespugio: Implications for Toxicological Studies 89
Mark A. Roberts, Joseph M. Quattro, G Thomas Chandler
Persistent Organochlorine Compounds, Genetic Susceptibility, and Testicular Cancer Risk 90
Stephen M. Schwartz, Mary L. Biggs, Chu Chen, Larry L. Needham, Dana B. Barr
A Case-Cohort Study of Cryptorchidism, Hypospadias, and Delayed Sexual Maturation in a Dioxin-Contaminated
Region: Chapaevsk, Russia 91
Oleg Sergeyev, Boris Revich, Paige Williams, Susan Korrick, Vladimir Zeilert, MaryM. Lee,
Tatiana Ushakova, Igor Saharov, Larisa Altshul, Russ Hauser
Study for Future Families II: Phthalates in Pregnant Women and Children 92
Shanna H. Swan, Erma Z. Drobnis, Christina Wang, J. Bruce Redmon, Amy Sparks,
Bill L. Lasley, Antonia M. Calafat
Laboratory and Mesocosm Measurements of the Stereoselective Degradation of Endosulfan 93
Spencer S. Walse, John L. Ferry
Multivariate Modeling of the Photolysis of Aqueous Fipronil 94
Spencer S. Walse, Li Kong, John L. Ferry
Exposure to DDT and DDE in Relation to Menstrual Cycle Length Among Laotian Immigrants 95
Gayle Windham, Patrick Mitchell, Myrto Petreas, Diana Lee, BillL. Lasley
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disrupters Program Review Workshop
Mechanisms of Action
Alteration of Calcium-Dependent Cell Signaling as a Potential Mechanism of Endocrine Disruption 99
AbbyD. Benninghoff, Peter Thomas
Elevations of Estradiol in the Cycling and Ovariectomized, Estradiol-Implanted Female Rat by the
Drinking Water Disinfection By-Product Dibromoacetic Acid 100
Jerome M. Goldman, Ashley S. Murr
Testosterone :Fatty Acid Esterification: A Novel Target of Endocrine Disruption Caused by Tributyltin 101
Meredith P. Goading, Gerald A. LeBlanc
HPLC Purification of an Endocrine-Disrupting Retinoid Activity From a Minnesota Lake With a High
Incidence of Malformed Amphibians 102
Felix Grun, Aristocle Ndayibagira, David M. Gardiner, Bruce Blumberg
Differential Binding of Estrogens and Estrogenic Compounds to ERs Alpha, Beta, and Gamma of the
Atlantic Croaker, Micropogonias undulatus 104
Mary Beth Hawkins, Peter Thomas
A Short-Term Dosing Model for Detecting the Effects of Environmental Contaminants on Thyroid
Hormones in the Rat: Effects of Pesticides 105
JoanM. Hedge, David G Ross, MichaelJ. DeVito, Kevin M. Crofton
Impact of Chemical Mixtures on Steroid Metabolism in Mud Snails (Ilyanassa obsoleta) 106
Patricia D. McClellan-Green
Mechanisms of Toxicity of Dioxin-Like Compounds to Primate Ovarian Cells 108
Francisco M. Moran, Peter Lohstroh, Catherine A. VandeVoort, Jiangang Chen,
James W. Overstreet, Alan J. Conley, BillL. Lasley
Developmental Toxicity of Antiecdysteroids in the Crustacean Daphniamagna 109
Xueyan Mu, Gerald A. LeBlanc
Levels of the Neuropeptide Hormone APGWamide Are Elevated inTBT-Dosed Snails and in Snails
Transferred to a Contaminated Site 110
Eva Oberdorster, Patricia D. McClellan-Green
Endocrinology of Sex Determination in a Crustacean and Its Disruption by Biorational Insecticides 112
Allen W. Olmstead, Gerald A. LeBlanc
Cloning and Characterization of Estrogen Receptor a, -Ba, and -Bb in Zebrafish: Differential Expression During
Development and Effects of Steroids and Environmental Chemicals 113
Sarah Rothberg, Apolonia Novillo, Mitsuyo Kishida, Bikem Akten, Gloria V. Callard
Effects of Perfluorooctane Surfonate on Thyroid Hormone Status in the Rat and Mouse 114
Julie Thibodeaux, Roger Hanson, Brian Grey, John M. Rogers, Christopher Lau
Anti-Androgenic Effects of Vmclozolin on Male Rats Are Partially Attenuated by Testosterone Propionate 115
Cynthia Wolf, Joe Ostby, Jonathan Furr, Gerald A. LeBlanc, L. Earl Gray, Jr.
Index of First Authors 117
VI The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disrupters Program Review Workshop
Introduction
Evidence suggests that environmental exposure to some anthropogenic chemicals may result in disruption of endo-
crine systems in human and wildlife populations. A number of the classes of chemicals suspected of causing endo-
crine disruption fall within the purview of the U.S. Environmental Protection Agency's (EPA) mandates to protect
both public health and the environment. Although there is a wealth of information regarding endocrine disrupters,
many critical scientific uncertainties still remain.
In 1996, EPA's Office of Research and Development (ORD) identified endocrine disruption as one of its top six
research priorities and developed a risk-based research approach to address some of these uncertainties. ORD's
research program is based on a peer-reviewed Research Plan published in 1998 (www.epa.gov/ORDAVebPubs/final)
and has three long-term goals:
• Providing a better understanding of the science underlying the effects, exposure, assessment, and risk management
of endocrine disruptors. Research in this area includes determining: dose-response relationships, the effects of
exposure to multiple endocrine disruptors, major sources of exposure, and approaches for managing risks.
• Determining the extent of the impact of endocrine disruptors on humans, wildlife, and the environment. Research
includes determining: what effects are occurring in human and wildlife populations, the chemical classes of
greatest concern, the ambient levels of exposure, and how unreasonable risks can be mitigated.
• Supporting EPA's screening and testing program. ORD is developing needed computational tools as well as in vitro
and in vivo assays in support of the implementation of a screening and testing program for endocrine disruptors,
required by the 1996 Food Quality Protection Act.
ORD's intramural research program is conducted across three national laboratories—National Health and Environ-
mental Effects Research Laboratory, National Exposure Research Laboratory, and National Risk Management Re-
search Laboratory—and through one of its two national centers, the National Center for Environmental Assessment.
ORD's extramural research program is carried out through its other national center, the National Center for Environ-
mental Research, which is responsible for implementing the Science to Achieve Results (STAR) competitive grants
program. EPA's intramural research related to endocrine disruptors has been ongoing for several decades, and was
integrated into a coordinated research program in the mid-1990s. Since 1996, EPA has issued four Requests for
Applications in the area of endocrine disruptors, two of which have been with other federal agencies and with whom
EPA collaborates under the auspices of a Working Group under the Committee on Environment and Natural Re-
sources of the President's National Science and Technology Council.
This Program Review Workshop brings together EPA's intramural and extramural scientists as well as scientists funded by
other federal agencies who are working to address the uncertainties associated with exposure and effects of endocrine-
disrupting chemicals in the environment. EPA uses Program Reviews such as this one to allow EPA and other federal and
non-federal scientists to discuss research progress on topics of major scientific interest to the Agency. The research
reported here is of critical importance to EPA, as it has the potential to strengthen the scientific basis for both assessing the
risk from exposure to endocrine disruptors and developing appropriate risk-management practices to mitigate their effects.
The abstracts in this report are organized into platform presentations and poster presentations, and are presented in
alphabetical order by research category. The research described in this report has not been subjected to the Agency's
required peer review and policy review, and does not necessarily reflect the views of the Agency. Therefore, no
official endorsement should be inferred. Any opinions, findings, conclusions, or recommendations expressed in this
report are those of the investigators who participated in the research and the Program Review Workshop, and not
necessarily those of EPA or the other federal agencies supporting the research.
For further information on EPA's Endocrine Disruptors Research Program, please contact the National Program
Director, Elaine Z. Francis, Ph.D., by telephone at (202) 564-6789, or by e-mail at francis.elaine@epa.gov.
The Office of Research and Development's National Center for Environmental Research vii
-------
Platform Presentations
-------
Background
-------
Endocrine Disrupters Program Review Workshop
Overview of EPA's Regulatory Program
Gary E. Timm, James P. Kariya, Leslie W. Touart, L. Greg Schweer, and Jane S. Smith
U.S. Environmental Protection Agency, Washington, DC
The Food Quality Protection Act of 1996 requires the
U.S. Environmental Protection Agency (EPA) to develop
and implement a screening program to determine whether
pesticides may have estrogenic effects in humans. It and
other laws provide authority to screen for other endocrine
effects, screen other substances that may have an effect
cumulative to an effect of a pesticide, examine drinking
water contaminants to which a substantial population may
be exposed, and screen for effects in wildlife. The pro-
gram must use "appropriate validated test systems and
other scientifically relevant information."
An Advisory Committee convened by EPA found that
there currently were no appropriate validated test systems
for endocrine effects. It recommended general criteria
that a screening program should meet, and suggested a
two-tiered battery of assays which, if validated, could serve
as a screening program. The Advisory Committee also
suggested a scheme for prioritizing chemicals for screen-
ing. EPA's Endocrine Disrupter Screening Program (EDSP)
currently is involved in several activities, most of which
are based on the Advisory Committee's recommendations:
• Validation of assays
• Coordination of regulatory considerations for pesticides
and other chemicals
• International harmonization of screening programs for
endocrine disruption
• Prioritization of chemicals for screening.
Most of the current efforts by the EDSP are directed
towards validation of assays and selection of an appropriate
battery for use. Validation of assays follows principles set
forth by the Interagency Coordinating Committee for the
Validation of Alternative Methods, a committee of U. S. gov-
ernment agencies recently established to coordinate the de-
velopment, validation, acceptance, and harmonization of
alternative toxicological test methods throughout the fed-
eral government.
The Office of Research and Development's National Center for Environmental Research
-------
Screening and Testing Assays
-------
Endocrine Disrupters Program Review Workshop
Mosquitofish as a Model Organism for the Study
of Endocrine-Disrupting Chemicals
Robert A Angus', R Douglas Watson1, W. MikeHowett2, and Ronald L. Jenkins2
1 University of Alabama, Birmingham AL; 2Samford University, Birmingham, AL
In recent years, evidence has been accumulating that
many of the chemicals that humans have introduced into the
environment may act as hormones and have the potential to
disrupt the endocrine systems of wildlife and humans. This
has led to the realization that there is a need for the develop-
ment and validation of in vivo and in vitro test methods to
screen toxicants for endocrine-disrupting activity in verte-
brate and invertebrate species, hi response to that need, this
research group has conducted the necessary background
studies to establish the mosquitofish (Gambusia qffinis) as a
model organism for in vivo screening of substances with
suspected endocrine activity. This small live-bearing fish in
the guppy family is easily kept and bred in aquaria. It is abun-
dant in nature and available from commercial suppliers. Its
natural range includes fresh and brackish waters throughout
the Southeastern United States. Mosquitofish have been in-
troduced into the Western United States and into areas with
moderate climates throughout the world.
Expression of the vitellogenin gene is used to test
chemicals for estrogenic activity. Vitellogenin is an egg
yolk protein normally only expressed in females in re-
sponse to estrogen. The presence of vitellogenin in the
serum of males is evidence of estrogen exposure. This
research group has developed and published a quantita-
tive Western blot assay for mosquitofish vitellogenin for
this purpose. As part of these studies, the extent to which
the vitellogenic response to estrogen exposure is tem-
perature-dependent has been characterized.
A morphological trait is used to test for androgens.
Male mosquitofish have a highly elongated and modified
anal fin (gonopodium) not found in normal females. If a
female is exposed to an androgen, her anal fin will be
induced to develop into agonopodium-like structure (see
Figure 1). The presence of modified anal fins in female
mosquitofish is visible evidence of exposure to an an-
drogenic substance. This research group exposed fe-
male mosquitofish to a variety of known androgens and
characterized the response using digital photography and
computer image analysis techniques.
As part of this project, steroids have been isolated
and identified from the water of a river receiving paper
mill effluent where female mosquitofish are masculin-
ized. Mosquitofish are now being used in laboratory ex-
posure studies to study the effects of those chemicals
on reproductive development and function.
hi addition to their suitability as laboratory test organ-
isms, mosquitofish are useful as sentinel species for the
detection of endocrine disrupters in aquatic environments.
Field studies of mosquitofish have included evaluations of
reproductive anatomy and fitness in: (1) a population in
which females have been masculinized by exposure to
androgens in paper mill effluent, and (2) a population liv-
ing below the outfall of metropolitan wastewater treat-
ment plants to investigate possible evidence of exposure
to estrogens.
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disruptors Program Review Workshop
r
Normal
Female
Masculinized
Female
Figure 1. The anal fins of female mosquitofish exposed to an androgen are induced to develop into
gonopodium-like structures.
10
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disruptors Program Review Workshop
Identification of Endocrine Disruptors Using a Short-Term
Reproduction Assay With the Fathead Minnow
G. T. Ankley, KM. Jensen, M.D. Kahl, J.J. Korte, E.A Makynen, and M. W. Hornung
U.S. Environmental Protection Agency, Duluth, MN
Tests with small fish have been a recommended com-
ponent of virtually every proposed regulatory program
for endocrine-disrupting chemicals (EDCs). An ideal test
would provide information suitable for determination of
whether a chemical has the potential to exert adverse ef-
fects via specific modes/mechanisms of action (MOA) of
concern, as well as supply dose-response data useful for
higher tier risk assessments (e.g., growth, reproduction).
To address these needs, a short-term reproduction assay
was developed utilizing reproductively mature fathead
minnows (Pimephalespromelas). Endpoints evaluated in
the assay include those specific for endocrine-related MOA
(e.g., plasma vitellogenin and steroid concentrations), and
those reflective of general reproductive fitness (e.g., fe-
cundity, fertility, Fl viability). The test has been evaluated
using model EDCs with suspected/known MOA, includ-
ing strong and weak estrogen receptor agonists (estradiol,
methoxychlor), androgen receptor agonists (methyltest-
osterone, trenbolone), androgen receptor antagonists
(vinclozolin, flutamide), and inhibitors of steroid metabo-
lism (fadrozole). The test has consistently and accurately
characterized the different chemicals both with respect to
their reproductive toxicity and through alterations in one
or more endpoints reflective of presumed MOA.
The Office of Research and Development's National Center for Environmental Research
11
-------
Endocrine Disruptors Program Review Workshop
Environmentally Mediated Endocrine Disruption in Estuarine
Crustaceans: Meiobenthos-Based Life-Cycle Assays for Evaluating
Risks of Reproductive and Endocrine Toxicity
G. Thomas Chandler1, T. Gary1, D. Volz1, D. Block1, J. Ferry2, G Scott', J. Quattro3,
E. Wirth1, andM. Fulton1
department of Environmental Health Sciences, Arnold School ofl*ublic Health, department of Chemistry,
and ^Department of Biology, University of South Carolina, Columbia, SC
In the majority of the world's estuaries, persistent con-
taminants of potential endocrine and reproductive toxicity
reside in sediments almost continuously at nonlethal expo-
sure concentrations. Endocrine-disrupting chemicals
(EDCs) can mimic natural hormone action by either ago-
nizing/antagonizing hormonal effects, modifying hormone
receptor structure, or recognizing/blocking hormonal bind-
ing sites. For example, ecdysteroids are known to be im-
portant regulators of molting, embryonic development,
metamorphosis, reproduction, and pigmentation in insects,
crustaceans, and even nematode worms; thus, the endo-
crine-disrupting bisacylhydrazine insecticides exert con-
trol by targeting ecdysteroid receptors.
Vitellogenesis—the process of yolk synthesis—is hor-
monally regulated and potentially altered by EDCs that inter-
fere with proper ecdysteroid action. Thus, the health and
ultimate quality of developing crustacean embryos is abso-
lutely dependent on ecdysterioid balance and the concentra-
tion and nutritional quality of vitellin—the major lipoprotein
produced via vitellogenesis. During the past 3 years, and
from the Environmental Protection Agency's (EPA) Sci-
ence to Achieve Results Program support, meiofaunal-based
sediment ecotoxicology has moved away from "kill 'em
and count 'em" approaches to more sophisticated life-cycle
approaches targeting biochemical and reproductive effects
directly linked to population maintenance (e.g., genetic
change, fertilization success, altered sex ratios, masculin-
ization/feminization, vitellogenesis, egg quality, reproduc-
tion, recruitment, and population growth modeling) under
the mid- to low-level chronic exposures typical of urban-
ized settings.
For the meiobenthos and most invertebrates, the major-
ity of life-cycle screening tools have not been available for
assessing the endocrine/reproductive disrupting potentials
of contaminants. Therefore, this research project built on
prior meiofaunal culturing technologies to develop: (1) a
microplate culturing assay with the benthic copepodAmphi-
ascus tenuiremis in which 18-hour old nauplii (or stage 1
copepodites) are reared individually to reproductive matu-
rity under EDC exposure in hydrophilic (low binding) 96-
well microtiter plates for 10-12 days and then mated; (2) a
spiked-sediment microassay in which nauplii or stage 1
copepodites are cultured to Fj production under continuous
EDC exposure; (3) an in vivo lipovitellin-based semiquan-
titative assay of copepod individual egg/embryo quality us-
ing dual-channel laser-scanning confocal microscopy
(LSCM); (4) two new enzyme-linked immunosorbent as-
says (ELISAs) for individual copepod ecdysone and vi-
tellin; and (5) detection of pesticide-induced genetic change.
Eggs/embryos of A. tenuiremis are small (approxi-
mately 50 um thick) with a clear chorion and yolk vitellin
tightly packaged into vesicles that are easily quantified by
LSCM in a nondestructive manner. In microplate assays,
individually exposed nauplii (or C-I copepodites) can be
monitored through to the copepodite and adult stages via
inverted microscope or LSCM for developmental/repro-
ductive effects. ELISAs allow assessment of specific en-
docrine effects most important to population growth and
maintenance. This project focuses on how these ap-
proaches have been used to evaluate reproductive/endo-
crine effects of a "modern" gamma-aminobutyric acid-
disrupting cyclodiene insecticide, fipronil. Fipronil induces
developmental delays and strong, but reversible, sex-
specific reproductive failures in copepods at environmen-
tally realistic concentrations. Fipronil is receiving rapidly
increased use in the Southeastern United States in craw-
fish/rice-culture, turf grass management, and fire ant/ter-
mite control. In addition to toxicological data and test
methods, this project is providing EPA with pesticide fate,
degradation, and transport information via mesocosm
simulations with laboratory-based salt-marsh ecosystems.
12
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disruptors Program Review Workshop
Complementary In Vitro and In Vivo Rodent Assays
Kirsten C. Fertuck, Jason B. Matthews, and Tim R Zacharewski
Michigan State University, East Lansing, MI
In attempts to identify pollutants, drugs, dietary con-
stituents, and other compounds possessing hormone ago-
nist or antagonist activity, in vitro assays offer many ad-
vantages, including relative simplicity, rapidity, and low
cost. However, only in vivo assays provide the pharma-
cokinetic and pharmacodynamic context necessary to
much more fully predict the compound's effects on hu-
mans or other organisms of concern. An appealing strat-
egy, therefore, involves an initial in vitro screening to iden-
tify a subset of compounds for which in vivo testing is
warranted.
In a first phase of studies aimed at identifying and
characterizing pollutant (anti)estrogens, both cell-free com-
petitive estrogen receptor (ER) binding assays and ER-
transactivation reporter gene assays in human breast can-
cer cells were performed in vitro on a set of related com-
pounds. For parent compounds inducing transactivation but
not receptor binding, key metabolites also were examined
to verify that the metabolites responsible for activity are
those that are formed in vivo by the species of interest.
Using these initial data, one compound, benzo[a]pyrene, was
selected for in vivo testing using the mouse uterotrophic
assay. When both benzo[a]pyrene and major metabolites
were unable to induce either uterine weight or expression of
lactoferrin, highly estrogen-inducible endpoints in the uterus,
amicroarray strategy was adopted. These microarrays, en-
riched in estrogen-regulated genes, do not depend on in-
creases in organ weight and therefore can be used on any
tissue of interest. This ongoing work will be particularly
useful for the detection and characterization of compounds
that interact with the ER only in a subset of tissues and that,
therefore, may be missed in traditional, solely in vitro and
uterine-based testing.
The Office of Research and Development's National Center for Environmental Research
13
-------
Endocrine Disruptors Program Review Workshop
Endocrine Disrupters—Tiered Screening and Testing:
Filling Key Data Gaps
Leon Earl Gray, Jr.1, Joseph Ostby1, Vickie Wilson1, Christy Lambright1, Kathy Bobseine1, Phillip Hartig1,
Andrew Hotchkiss1'2, Cindy Wolf1'3, Johnathan Furr1, Mathew Price1'2, Louise Parks1, Ralph Cooper1,
Tammy Stoker1, Susan Laws1, Sig Degitz4, Kathleen Jensen4, M. Kahl4, Joseph Korte4,
Elizabeth Makynen4, Joseph Tietge4, and Gerald Ankley4
Reproductive Toxicology Division, National Health and Environmental Effects Research Laboratory
(NHEERL), Office of Research and Development (ORD), U.S. Environmental Protection Agency (EPA),
Research Triangle Park, NC; 2EPA/North Carolina State University (NCSU) Cooperative Research
Program and NCSU Zoology Department, Raleigh, NC; 3NCSU Department of Toxicology, Raleigh, NC;
4Mid-Continent Ecology Division, NHEERL, ORD, EPA, Duluth, MN
The U.S. Environmental Protection Agency (EPA) is
developing a screening and testing program for endocrine-
disrupting chemicals (EDCs). High-priority chemicals
would be evaluated in the Tier 1 Screening (Tl S) Battery.
Chemicals determined to be positive in T1S then would be
tested (Tier 2). T1S includes in vitro estrogen receptor
(ER) and androgen receptor (AR) binding and/or gene
expression, an assessment of steroidogenesis, and mam-
malian and nonmammalian in vivo assays.
Using the rat, the uterotropic assay detects estrogens
and antiestrogens, while steroidogenesis, estrogenicity, and
hypothalamic-pituitary-gonadal function are assessed in a
"Pubertal Female Assay." (Anti-)androgens are detected
in the Hershberger Assay (androgen-dependent tissues in
castrate-immature male rats). Fish and amphibian assays
also are being developed. The fathead minnow assay can
identify EDCs displaying several mechanisms of concern,
including AR and ER, AR antagonists, and aromatase in-
hibitors, but the amphibian assay is designed to detect
thyroid-active substances. Several alternative mammalian
in vivo assays have been proposed. Of these, a short-
term pubertal male rat assay appears most promising. An
in wfero-lactational screening protocol also is being evalu-
ated. For Tier 2, the numbers of endocrine-sensitive end-
points and offspring (Fl) examined needs to be expanded.
Consideration also should be given to tailoring T2, based
on the results of Tl. For example, endpoints such as
anogenital distance at birth and nipple/areola retention in
infant rats should be required in testing for androgens and
antiandrogens because they are sensitive, permanent ef-
fects that are highly correlated with malformations and
reproductive organ weight changes later in life. EDCs that
display antithyroid, estrogenic, or antiestrogenic activity
display a different profile of developmental effects, and
these aforementioned endpoints are not sensitive to dis-
ruption by these mechanisms. Tiers 1 and 2 also should
examine relevant mixtures of EDCs. For example, toxi-
cants that induce malformations in androgen-dependent
tissues produce cumulative effects, even when two chemi-
cals act via different mechanisms of action.
14
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disruptors Program Review Workshop
The Medaka: An In Vivo Model for Detection of Adverse
Effects of Endocrine Disruptors
David E. Hinton1, Seth W. Kuttman1, Shosaku Kashiwada1, David C. Bencic1, Yuko Wakamatsu2,
Kenji Ozato2, Cory S. Roger3, Corrine R. Davis3, and Swee J. Teh3
'Duke University, Durham, NC;2 University of California, Davis, CA;
3Nagoya University, Nagoya, Japan
The principal objective of this research is to develop
and validate the use of a short-term in vivo model—the
small fish, medaka (Oryzias latipes)—to identify adverse
effects of exposure to endocrine-disrupting chemicals.
It is possible to determine sex without invasive method-
ology using secondary characteristics or by detection of
phenotypic markers in mutant species. Sex reversal fol-
lows exposure to estrogenic or androgenic hormones.
Calibration of breeding pairs prior to exposure improves
the ability to detect reproductive impairment. The small
body size, coupled with a newly developed, transparent
medaka, diminishes cost and improves whole-body sur-
vey histopathology while extending phenotypic detec-
tion from embryonic to adult life stages. The compressed
life cycle of the medaka makes it possible to perform
multigenerational assays in a reasonable amount of time.
These studies are designed to integrate findings across
levels of biological organization. Molecular and biochemi-
cal studies have addressed the role of genes associated
with steroid biosynthesis and metabolism in endocrine
disruption. Multiple forms of cytochrome P450, critical
for metabolism of numerous endobiotics including test-
osterone, estrogen, and retinoic acid, have been isolated
and characterized. These studies also have addressed
the promotion of tumorigenesis by endogenous estrogen
and xenobiotics with hormone-like activity. Cellular and
tissue analyses first characterized the effects of endo-
crine-disrupting compounds on active, quiescent, and
recrudescing gonads. Results from these studies dem-
onstrated atresia of oocytes, supportive cells, stromal
alteration, and focal necrosis (testis). At the individual
level, the most sensitive developmental stage for adverse
gonadal and/or reproductive effects followed initiation
of exposure immediately after hatching. At the popula-
tion level, using rigidly controlled laboratory conditions,
exposure to mixtures of various endocrine-disrupting
compounds at environmentally relevant concentrations
reduced reproductive success. The use of medaka as a
model for evaluating effects of endocrine disruption at
all levels of biological organization will continue to be
explored.
The Office of Research and Development's National Center for Environmental Research
15
-------
Endocrine Disruptors Program Review Workshop
Validation of an In Vivo Protocol To Evaluate the Effects
of Environmental Chemicals on Female Pubertal
Development and Thyroid Function
Susan C. Laws, Janet M. Ferrell, Tammy E. Stoker, Jerome M. Goldman,
L. Earl Gray, Jr., and Ralph L. Cooper
Reproductive Toxicology Division, National Health and Environmental Effects Research Laboratory,
U.S. Environmental Protection Agency, Research Triangle Park, NC
In response to emerging concerns that environmental
chemicals may have adverse effects on human health by
altering the function of the endocrine system (http://
www.who.int/pcs/), the Food Quality Protection Act man-
dated that the U. S. Environmental Protection Agency (EPA)
develop and implement an endocrine disrupter screening
program (EDSP). Working toward this goal, EPA cur-
rently is implementing a proposed EDSP that is designed
to detect chemicals that alter the estrogen, androgen, and
thyroid systems in humans, fish, and wildlife (http://
www.epa.gov/scipoly/oscpendo/index.htm). Studies cur-
rently are ongoing within the Agency to develop, stan-
dardize, and validate a number of in vitro and in vivo
mammalian and ecotoxicological assays for use in a Tier 1
Screening Battery.
This project focuses on studies that support the vali-
dation of the in vivo protocol to evaluate the effects of
environmental chemicals on female pubertal development
and thyroid function in the juvenile rat. Using a 20-day
dosing regimen that encompasses the critical period of
sexual maturation for the female rat, this protocol has the
ability to detect agents that display antithyroid, estrogenic,
antiestrogenic (estrogen receptor or steroid-enzyme medi-
ated) activity, or alter puberty via changes in luteinizing hor-
mone, follicle-stimulating hormone, prolactin and growth
hormone secretion, or via alterations in hypothalamic func-
tion. Initial studies have identified the more sensitive end-
points, provided an assessment of the robustness of the
protocols with regard to inter- and intra-laboratory and in-
ter-strain sources of variation, and standardized operating
procedures. Subsequent studies have assessed the reliabil-
ity (e.g., ability to be replicated in multiple laboratories) and
the relevance of the protocol (e.g., ability to provide amea-
sure of a specific biological process) using chemicals with
known mechanisms of action. Specific issues pertaining to
study design that could possibly confound the results have
been evaluated. For example, a study to assess the relation-
ship between growth rate and pubertal development dem-
onstrated that all endpoints were unaffected, even in the
presence of a 10 percent reduction in body weight, a limit
set for dose selection based on the maximum tolerated dose.
hi summary, these studies demonstrate that the female pu-
bertal protocol is a reliable and fairly simple screen to detect
endocrine-disrupting chemicals with multiple mechanisms
of action.
16
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disruptors Program Review Workshop
Reproductive and Developmental Screening Protocols for
Endocrine Disruptors Using Estuarine Crustaceans
Charles L. McKenney, Jr.
Gulf Ecology Division, National Health and Environmental Effects Research Laboratory,
U.S. Environmental Protection Agency, Gulf Breeze, FL
The objective of this research is to develop in vivo screen-
ing protocols for endocrine disruption in marine crusta-
ceans, invertebrates of ecological and economic importance.
A series of comparative developmental and reproductive
studies were performed on several species of estuarine crus-
taceans in response to three juvenile hormone agonists
(JHAs) (methoprene, fenoxycarb, and pyriproxyfen). Lar-
val development of the grass shrimp, Palaemonetes pugio,
was greater than two orders of magnitude more sensitive to
disruption by JHAs than was embryonic development.
Fenoxycarb-exposed larvae had significantly altered levels
of ecdysone, the hormone that along with juvenoids is known
to regulate the metamorphic process in decapod crusta-
ceans. For two of the three JHAs under similar static expo-
sure conditions, developing larvae of the xanthid mud crab,
Rhithropanopeus harrisii, exhibited reduced metamorphic
success at lower concentrations than grass shrimp larvae.
These comparative responses suggest that the more
rigidly controlled metamorphic process in crabs is more
sensitive to compounds acting as endocrine disrupters
than is the more plastic metamorphic pattern seen in
shrimp. The final crab larval stage, the megalopa, was
more sensitive to JHA exposure than earlier zoeal stages.
Mud crab larvae exposed to fenoxycarb had reduced bio-
mass and lipid content, particularly triglycerides and free
sterols, at concentrations below which inhibited metamor-
phic success. Concentrations of fenoxycarb, which re-
duced the reproductive capacity in single life-cycle expo-
sures of the estuarine mysid, Americamysis bahia, were
similar to those concentrations that inhibited metamor-
phosis in grass shrimp under similar flow-through expo-
sure conditions. Juvenile mysids were released by exposed
adults and reared through maturation without further ex-
posure; however, they produced fewer young and had
altered sex ratios (reduced percentages of males) at lower
parental-exposure concentrations than directly impacted
parental reproduction.
Because the endocrine glands responsible for regulating
mysid sexual differentiation and reproduction develop dur-
ing larval stages, these transgenerational responses may well
be a product of irreversible effects during developmental
exposures that become apparent following maturation and
initiation of reproduction. These findings suggest the ne-
cessity of at least a two-generational mysid exposure proto-
col for adequately predicting the ecological risk of chemi-
cals acting as endocrine disrupters on crustaceans that func-
tion as the dominant secondary producers in estuarine eco-
systems.
The Office of Research and Development's National Center for Environmental Research
17
-------
Endocrine Disruptors Program Review Workshop
Thyroid Axis Inhibition in Xenopus laevis: Development of an
Amphibian-Based Screening Assay for Thyroid Disruption
J.E. Tietge, GW. Holcombe, KM. Flynn, J.J. Korte, RC. Kolanczyk, P.A Kosian, andS.J. Degitz
Mid-Continent Ecology Division, National Health and Environmental Effects Research Laboratory,
Office of Research and Development, U.S. Environmental Protection Agency, Duluth, MN
In response to the initial Endocrine Disrupter Screen-
ing and Testing Advisory Committee recommendations,
research was conducted on the development of a Xeno-
pus tow's-based tail resorption assay for evaluating thy-
roid axis disruption. These experiments highlighted key
limitations that are associated with relying on tail resorp-
tion as a measure of anti/thyroid activity. The most criti-
cal limitation is that tail tissue of tadpoles in metamor-
phic climax are insensitive to perturbation by agonists/
antagonists. To improve on the initial proposal, this re-
search group conducted experiments comparing the sen-
sitivity of premetamorphic (stage 51) and prometa-
morphic (stage 54) tadpoles to the model thyroid axis
antagonists methimazole (control, 6.25, 12.5, 25, 50,
and 100 mg/L); 6-propylthiouracil (control, 1.25,2.5,5,
10, and 20 mg/L); perchlorate (control, 15.6, 62.5, 250,
1,000, and 4,000 ug/L); and iopanoic acid (control, 23.4,
93.8, 375, 1,500, and 6,000 ug/L). Tadpoles were ex-
posed for a 2-week period, and developmental stage,
thyroid size, and histology were examined at 1 and 2
weeks after exposure.
Methimazole, 6-propylthiouracil, and perchlorate,
which are thyroid hormone synthesis inhibitors, all caused
a concentration-dependent inhibition of the thyroid axis.
Further, these three compounds caused dose-dependent
changes in thyroid gland morphology. These changes were
characterized as reduced colloid, glandular hypertrophy,
and cellular hyperplasia. Treatment failed to affect growth,
even in tadpoles that experienced significant metamorphic
inhibition. As determined from these endpoints, there were
only minor differences in sensitivity observed between
the two stages examined. Iopanoic acid, atype II deiodinase
inhibitor, surprisingly stimulated metamorphosis. These
results indicate that tadpoles in the early stages of meta-
morphosis are sensitive to thyroid axis inhibition and that
development of a short-term, diagnostic amphibian-based
thyroid screening assay shows considerable promise.
18
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disruptors Program Review Workshop
Novel Cheminformatics and Pattern Recognition Tools Useful
for Risk Assessment and Regulatory Control
William J. Welsh
University of Medicine and Dentistry of New Jersey, Robert Wood Johnson
Medical School, Piscataway, NJ
Computer-aided molecular modeling (CAMM), origi-
nally developed for rational drug design, has demonstrated
its utility in a wider range of applications, including envi-
ronmental risk assessment and computational toxi-
cology. Approaches in CAMM can be reduced to two
general strategies: (1) receptor-based modeling, and
(2) ligand-based modeling. The former case refers to
the situation in which the three-dimensional (3D) struc-
ture of the target receptor is available, thus enabling one
to exploit the familiar "lock and key" paradigm. The lat-
ter case refers to the all-too-frequent situation in which
the receptor's 3D structure is not known, thus requiring
one to discern the relationship between molecular struc-
ture and biological activity by inferring information em-
bedded in pertinent structure-activity data, hi ligand-based
approaches, the transformation from data to knowledge
and insight is greatly facilitated by the construction of
quantitative structure-activity relationship (QSAR) mod-
els. Basically, QSAR models employ statistical regres-
sion techniques to correlate variations in the biological
activity of a series of chemicals with variations in their
molecular structure as encoded in calculated structural
features and properties (descriptors). In risk assessment,
QSAR models can serve as powerful tools to screen
large chemical databases in search of potentially hazard-
ous agents, predict the biological activity of these agents
prior to biological testing, and gain insight into their
mechanism of action and structural prerequisites for bio-
logical activity.
To illustrate the utility of both receptor-based and ligand-
based approaches including QSAR models in risk assess-
ment, examples will be taken from ongoing research
projects in the Welsh Laboratory. The presentation will
conclude by introducing several new computational tools
that show great promise for use in many areas, including
risk assessment and computational toxicology. For ex-
ample, the Polynomial Neural Network (PNN) is a pow-
erful iterative neural network algorithm that automatically
produces QSAR models in parametric form [Y = /(X)n],
where the X terms can be linear (n = 1) or nonlinear (n =
2, 3, etc.). The PNN thus combines the best features of
artificial neural networks (i.e., inherent nonlinearity) and
multivariate regression analysis (i.e., analytical equation)
into a single entity. The PNN also is remarkably insensi-
tive to outliers and "noisy" data that often confound nor-
mal regression methods. Second, the "Shape Signatures"
algorithm employs a unique concept based on molecular
shape and electrostatic properties that enables rapid screen-
ing of large numbers of small molecules against each other
or even against a receptor pocket. The utility of the PNN
and Shape Signatures will be demonstrated using examples
associated with the identification and prediction of endo-
crine-disrupting compounds.
The Office of Research and Development's National Center for Environmental Research
19
-------
Developmental Exposures
-------
Endocrine Disrupters Program Review Workshop
Short-Term Exposure to an Environmental Mixture of PHAHs:
Dose-Additive Effects on Serum Thyroxine
Kevin M. Crofton1, Joan M. Hedge1, David G Ross2, and Michael J. DeVito2
'Neurotoxicology and2Experimental Toxicology Divisions, National Health and Environmental Effects
Research Laboratory, Office of Research and Development, U.S. Environmental Protection Agency,
Research Triangle Park, NC
Endocrine disruption from environmental contaminants
has been linked to a broad spectrum of adverse outcomes.
An additional concern about endocrine-disrupting
xenobiotics is the potential for additive or synergistic ef-
fects of mixtures. A short-term dosing model to examine
the effects of environmental mixtures on thyroid homeo-
stasis (TH) has been developed. Prototypic chemicals such
as dioxins, polychlorinated biphenyls (PCBs), and poly-
brominated diphenyl ethers have been shown to alter TH
homeostasis in this model primarily by upregulating he-
patic catabolism of thyroid hormones. Current efforts are
testing the hypothesis that the effects of mixtures of these
chemicals on thyroxine (T4) concentrations can be pre-
dicted by dose-additivity theory. The current study used
28 day-old female Long-Evans rats, orally dosed with vary-
ing concentrations of 18 different planar halogenated aro-
matic hydrocarbons (2 dioxins, 4 dibenzofurans, and 12
PCBs, including dioxin-like and non-dioxin-like PCBs) for
4 consecutive days. Serum samples were collected 24
hours after the last dose. Serum total T4 was measured
via radioimmunoassay. Extensive (7-9 doses/chemical)
dose-response functions were statistically modeled to de-
termine median effective doses using the Hill Equation. A
mixture was custom synthesized with the ratio of chemi-
cals based on environmental concentrations (e.g., relative
totetrachlorodibenzodioxin: tetrachlorodibenzofuran- 1.5x,
PCB 126 - 50x; PCB 153 - 30,000x). Serial dilutions of
this mixture were tested in the 4-day dosing assay. Pre-
dicted outcome, based on the assumption of dose-additiv-
ity, was tested using statistical dose-response modeling.
Preliminary analyses of results suggest that the effects of
the mixture on serum total T4 can be predicted by dose-
additivity. There was no evidence of synergism or an-
tagonism. Future work will expand the mixture to include
chemicals from diverse classes of thyroid disrupters such
as TH synthesis inhibitors.
The Office of Research and Development's National Center for Environmental Research
23
-------
Endocrine Disruptors Program Review Workshop
Effects of Endocrine-Disrupting Compounds
on Mammary Tissue Development
Suzanne E. Fenton1, Christine C. Davis1, Jennifer Rayner1, Geri L. Youngblood1, and Judy Moon2
1 Reproductive Toxicology Division, National Health and Environmental Effects Research Laboratory,
Office of Research and Development, U.S. Environmental Protection Agency, Research Triangle
Park, NC; department of Toxicology, NITR, Korea FDA, Eunpyung-Gu, Seoul, Korea
Breast cancer risk in women is known to be signifi-
cantly influenced by genetics and by prolonged exposure to
estrogen. However, when all known risk factors and char-
acteristics are considered, more than 50 percent of breast
cancer cases remain unexplained. There is a growing body
of evidence indicating that exposures to certain toxic chemi-
cals and hormone-mimicking compounds may contribute
to the development of breast cancer. Several endocrine-
disrupting compounds (EDCs) that act in an estrogenic or
anti-androgenic manner are known to alter rodent mam-
mary gland development. Although epigenetic in nature, these
xenobiotics may hasten development of the gland and in-
crease the incidence of mammary tumors if they signifi-
cantly alter serum estradiol levels, or if they change recep-
tor expression patterns, hormone transport, or metabolism
that results in altered response to endogenous estradiol lev-
els. Nonylphenol is an example of an EDC that this research
group has shown to hasten mammary gland development
following acute in utero exposure.
Delayed development of the mammary gland also can
be caused by in utero exposure to EDCs, resulting in im-
printing, or irreversible effect in the offspring. This type of
delayed glandular development could lead to increased tu-
mor formation due to a shift or enlargement in the window
of sensitivity to carcinogens. For example, a toxicant may
delay mammary development so that undifferentiated or
dividing cells may be present for longer periods of time,
thus rendering the tissue more vulnerable to a subsequent
genotoxic insult. The herbicide atrazine and the polyaromatic
hydrocarbon dioxin are examples of compounds that delay
mammary gland development and increase the potential for
mammary carcinogenesis. A delay in mammary gland de-
velopment has been detected as early as postnatal day 4 and
persists throughout puberty in female rats prenatally ex-
posed to atrazine. Similarly, dioxin exposure on gestation
day 15 causes an irreversible modification in epithelial mi-
gration and branching patterns. Because of altered epithelial
differentiation, terminal end buds are present for a longer
period of time. These multilayered structures are sensitive
to carcinogens, such as 7,12-dimethylbenz[a]anthracene,
and exposure to such agents during this critical window of
susceptibility could lead to increased multiplicity or decreased
latency to tumor formation.
24
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disruptors Program Review Workshop
Analysis of Sensitive Developmental Stages in Birds to EDCs
D. Michael Fry1, James R Millam1, AL. Erichsen1, L.A. Preston1, C.B. Craig-Veit1,
AE. Quigliano2, and M.R Vianf
Department of Animal Science, University of California, Davis, CA; 2Leonard Davis School of Gerontology,
University of Southern California, Los Angeles, CA; Department of Environmental
Toxicology, University of California, Davis, CA
Estrogenic exposure of embryonic Japanese quail (JQ),
newly hatched zebra finch (ZF), or JQ chicks produced de-
velopmental abnormalities of the reproductive tracts, brain,
and behavior. Diethylstilbestrol and estradiol benzoate (E2B)
produced dose-responsive changes of testicular feminiza-
tion when injected into eggs early in incubation, at doses as
low as 10 nmol/g E2B. Oral dosing of ZF chicks on days 5-
11 after hatching with 10-1,000 nmol/g body wt/day E2B
produced a dose response in altered courtship and singing
behaviors in both males and females, expressed when the
birds became adults. Female finches developed the ability to
sing, and male finches exhibited reduced courtship and in-
creased nesting behaviors. The xenobiotic compounds
dicofol, methoxychlor, and endosulfan had little effect on
brain morphology or behavior.
E2B treatment at doses of 1-1,000 nmol/g body wt/
day altered sexually dimorphic brain nuclei with an in-
crease in the nuclear volume of area X. The three highest
doses (10,100, and 1,000 nmol/g) significantly increased
the nuclear area of RA and HVC. Courtship and reproduc-
tive behaviors were altered by postnatal estrogenic expo-
sure. The 100 nmol/g and 1,000 nmol/g E2B, and 100
nmol/g octylphenol (OP) dosed males displayed lower
female-directed song. The 1,000 nmol/g methoxychlor-
dosed males sang more in tests against females. The 10
nmol/g and 100 nmol/g E2B females had significantly higher
incidences of singing. Females in the 100 nmol/g E2B,
1,000 nmol/g E2B, and 1,000 nmol/g OP treatment groups
exhibited masculine behaviors without singing.
Reproductive testing was conducted either in individual
pair cages or in communal cages that permitted self-selec-
tion of mates, N = 5-10 pairs per group. Pairs consisted of
E2B-treated males and females, E2B-treated males paired
with canola-treated females, vice versa, and canola-treated
males and females. Posthatch E2B treatment produced sex-
specific impairments in reproduction that, in some instances,
were additive when both sexes were treated. Egg produc-
tion was reduced and egg breakage was increased in 100
nmol/g E2B-treated male and female pairs. The incidence of
missing eggs was increased in 10 nmol/g E2B-treated male
andfemale pairs. Candledfertilitywasreducedinbothgroups
containing 100 nmol/g E2B-treated males. The number of
hatched chicks was severely reduced in all E2B-treated
groups. These significant treatment effects (all p < 0.05)
show that posthatch E2B treatment profoundly disrupts the
reproductive performance of zebra finches, suggesting that
exposure to estrogens in the wild could impair the repro-
ductive performance of wild populations.
The Office of Research and Development's National Center for Environmental Research
25
-------
Endocrine Disruptors Program Review Workshop
Endocrine Disruption in Pubertal Rhesus Monkeys:
Growth and Sexual Maturation
Mari S. Golub, Bill L. Lasley, and Alice F. Tarantal
California Regional Primate Research Center, University of California, Davis, CA
During puberty, a number of systems (reproduc-
tive, central nervous, skeletal, immune) complete their
final stage of maturation under the influence of estro-
gen. Monkeys, like humans, undergo what some con-
sider to be a unique period of development during ado-
lescence, which makes them important models for un-
derstanding exogenous estrogen effects during this
period. This study evaluated a cohort of pubertal fe-
male monkeys treated with exogenous estrogen; this
report presents data on growth and sexual maturation.
Female adolescent rhesus monkeys (M. mulatto)
(n = 8/group) received daily oral doses of methoxy-
chlor (MXC; 25 or 50 mg/kg/day), diethylstilbestrol
(DES; 0.5 mg/kg/day), or vehicle control for 6 months
preceding and following the expected age at menarche
(30 months) with a subsequent 8-month recovery pe-
riod. Serum exogenous (nonsteroidal) estrogen activ-
ity as determined by an ERa transcription activation
assay was 0, 34, 42, and 60 pg/mL in the control,
MXC25, MXC50, and DES groups at the end of the
treatment period.
The DES group did not gain weight during treat-
ment and did not recover fully. Height growth also was
clearly depressed in the DES group. They weighed 15
percent less than controls and were 8 percent shorter
at the end of the study. Although the DES group gained
very little weight, they did demonstrate a very attenu-
ated "growth spurt" at the same time as the other mon-
keys. The MXC groups lagged behind controls in weight
gain only briefly during the growth spurt. Analysis also
demonstrated a smaller increase in height in the MXC25
as well as the DES group during the treatment period.
Muscle mass, skinfold measures, and body mass index
did not differ by group, indicating symmetrical growth
retardation. The reduced weight gain in the DES group
was associated with both reduced food intake and re-
duced food use efficiency.
Menarche (first occurrence of vaginal bleeding) oc-
curred 6 months later in the DES group than in controls.
Although MXC groups were not significantly delayed,
regressions indicated that higher serum exogenous es-
trogen was associated with later first menses in the co-
hort as a whole. The DES group mensed very infre-
quently during treatment (average 1 day in 6 months),
but all began mensing shortly (average 5±1 days) after
treatment was discontinued, suggesting that menses was
suppressed rather than menarche delayed. Both DES and
MXC led to premature emergence of a secondary sex
characteristic, reddening and swelling of skin, but re-
tarded growth of the nipple also was observed. Uterine
size and endometrial thickness, measured by ultrasound
after the recovery period, were not influenced by treat-
ment.
In conclusion, exogenous estrogens did not accel-
erate pubertal growth and development, but generally
retarded them. Most of the effects on growth and mor-
phology were reversible, but functional consequences
have not been evaluated.
26
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disruptors Program Review Workshop
The Effects of In Utero and Lactational Exposure to Genistein
or Daidzein on Reproductive Development in FVB/N Mice
and on Occurrence of Mammary Tumors
in MMTV-neu Transgenic Mice
Claude Hughes', Vicki Davis2, Firdos Shaikh2, Michael Vdlegas2, Arline Ho2, and Warren Foster3
'Quintiles, Inc., Research Triangle Park, NC, and Duke University Medical Center, Durham, NC;
2Cedars-Sinai Medical Center, Los Angeles, CA; 3McMaster University, Hamilton, ON, Canada
Exposure to estrogens during critical windows of de-
velopment may affect reproductive and mammary devel-
opment. In this study, researchers determined whether in
utero or lactational exposure to the phytoestrogens genistein
and daidzein affect reproductive development and occur-
rence of mammary tumors in mice. For the reproductive
studies, FVB/N dams were treated with diethylstilbestrol
(DES; 0.03 mg/kg/day), daidzein (40 mg/kg/day), or
genistein (4 and 40 mg/kg/day) during pregnancy (gesta-
tionalday 14-18)and/or lactation (birth-weaning). Thedams
and the offspring were fed a semipurified (isoflavone-free)
diet, and all treatments were given to the dams by gavage to
prevent direct exposure of the pups. Endpoints measured at
birth, weaning, and/or sexual maturity (2 months of age) in
the FVB/N offspring included anogenital distance, body
weight, uterine weight, testes weight, onset of puberty, and
estrus cycling. As expected, the potent estrogen DES al-
tered many reproductive outcomes in both male and female
mice. For some of these endpoints, daidzein and genistein
mimicked DES; however, unique or opposite responses
also were evident for each of the phytoestrogens, including
the low dose of genistein. Changes were evident for both
male and female pups with in utero, lactational, and both
exposures to all four treatments compared to the appropri-
ate control group. The window of exposure also was im-
portant because the in utero and lactational exposures often
had different effects. Adult outcomes such as estrous cy-
cling and uterine or testicular weights showed that treat-
ment effects were evident well after the end of the expo-
sure window (birth or weaning). These data indicate that in
utero and lactational exposure to isoflavones can influence
reproductive development of mice.
For mammary tumor studies, wild-type FVB/N dams
were mated with MMTV-neu transgenic males, yielding c-
neu hemizygous offspring. Mammary tumors spontane-
ously arise in this model due to activation of the neu proto-
oncogene. These dams were treated with daidzein (40 mg/
kg/day) or genistein (40 mg/kg/day) during pregnancy (ges-
tational day 14-18) or lactation (birth-weaning), and their
female offspring were monitored for tumor development
until 13 months of age. Neither genistein nor daidzein af-
fected mammary tumor latency or number of mammary
tumors per female via in utero or lactational exposure. Al-
though lactational exposure to daidzein increased the per-
centage of females that remained tumor free at maximum
age (13 months of age), in utero exposure to daidzein or
lactational exposure to genistein reduced the fraction of tu-
mor-free females, hi the MMTV-neu model, in which tu-
mors are induced by the most commonly overexpressed
oncogene in human breast cancer, there appears to be no
protective effect resulting from developmental exposure to
these isoflavones, as has been reported for carcinogen-in-
duced tumors in other animal models. Assessments of rates
of metastases and further analyses of tissues from these
animals are underway.
The Office of Research and Development's National Center for Environmental Research
27
-------
Endocrine Disruptors Program Review Workshop
The Michigan PBB Cohort 20 Years Later: Endocrine Disruption?
Michele Marcus1, Heidi Michels Blanch2, Paige Tolbert1, Carol Rubin2, Alden Henderson2,
Lorraine Cameron3, and Vicki Hertzberg1
'Emory University, Atlanta, GA; 2Centersfor Disease Control and Prevention, Atlanta, GA;
''Michigan Department of Community Health, Lansing, MI
In 1973, inadvertent substitution of a livestock feed
supplement with fire retardant led to widespread con-
tamination of meat and dairy products in Michigan with
polybrominated biphenyls (PBBs), a class of chemicals
lexicologically similar to polychlorinated biphenyls
(PCBs), dichlorodiphenyltrichloroethane (DDT), and di-
oxin, that are suspected of disrupting endocrine function
in humans and wildlife. More than 4,000 individuals with
high likelihood of exposure were subsequently enrolled
into the Michigan PBB Cohort, and serum samples were
analyzed for PBB.
Research on the health of exposed women and their
daughters has revealed a number of interesting findings.
Daughters of highly exposed women who were exposed in
utero and through breastfeeding had an earlier age at me-
narche than unexposed girls. PBB (as well as DDT, PCBs,
and dioxins) are concentrated in fatty tissue and breast milk.
These girls may have been exposed during a critical stage in
development of their endocrine and reproductive systems.
Preliminary analyses also reveal an extremely high rate of
miscarriage among highly exposed daughters who have
entered their reproductive years.
PBB exposure among adult women was not associ-
ated with duration of lactation, benign breast disease, uter-
ine fibroids, hip fractures, or age at menopause. There was
some evidence of an association of PBB exposure with
altered menstrual function among highly exposed women
who had experienced recent weight loss. It is possible that
recent weight loss mobilized the PBB from fat stores, thereby
causing an increase in circulating PBBs.
Future research on this cohort should examine the re-
productive health of male members of the cohort and sons
of exposed parents. Followup of the important findings
among women should include detailed study of pubertal
development, menstrual function, and early pregnancy.
This cohort provides an important opportunity to deter-
mine the human reproductive effects of endocrine-disrupt-
ing chemicals. Because of the large number of individuals
with documented exposure in utero and through breastfeeding,
this cohort represents a rare opportunity to study exposure
during critical periods of development. During these critical
periods, the developing fetus and infant are more susceptible
to the disrupting effects of chemical exposures.
28
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disruptors Program Review Workshop
Effects of Trihalomethanes on Pregnancy Maintenance in Rats
Michael G Narotsky1, Deborah S. Best1, Jerome M. Goldman1, Ashley S. Murr1,
and Susan R Bielmeier2
'Reproductive Toxicology Division, National Health and Environmental Effects Research Laboratory,
Office of Research and Development, U.S. Environmental Protection Agency,
Research Triangle Park, NC;2University of North Carolina, Chapel Hill, NC
In a recent epidemiological study1, consumption of
high concentrations of trihalomethanes (THMs), particu-
larly bromodichloromethane (BDCM), were associated
with an increased risk of spontaneous abortions. This
research group has shown that bromoform and BDCM,
two of the brominated THMs, cause pregnancy loss (i.e.,
full-litter resorption) in F344 rats. In view of the con-
cerns raised by the epidemiological data, BDCM-induced
pregnancy loss is being investigated to determine its mode
of action in rats. This investigation involves several fac-
ets of experimentation: (1) determination of the critical
period of pregnancy sensitive to the effect, (2) evalua-
tion of hormonal profiles, (3) hormonal replacement to
rescue the pregnancy, (4) ex vivo evaluation of critical
tissues, and (5) strain comparisons.
In early work, it was demonstrated that the critical
period for BDCM-induced full-litter resorption is during
the luteinizing hormone (LH) dependent period of preg-
nancy (gestation days 7-10). During this period, LH is
required to maintain luteal secretion of progesterone that,
in turn, is required to maintain pregnancy. In an evalua-
tion of hormonal profiles, it was demonstrated that preg-
nancy loss is indeed associated with decreases in serum
LH as well as progesterone. Furthermore, exogenous
progesterone and human chorionic gonadotropin (hCG),
an LH agonist, were effective in rescuing BDCM-ex-
posed pregnancies. These findings strongly support the
hypothesis that BDCM's mode of action is mediated by
a disruption of pituitary LH secretion. Exogenous GnRH
increased LH levels, suggesting that BDCM's effect is
mediated by GnRH, rather than nonresponsiveness of
the pituitary.
hi view of the epidemiological association of BDCM
and spontaneous abortion, it is important to consider the
possibility that BDCM may disrupt pregnancy mainte-
nance by acting at multiple target sites. It remains un-
clear whether BDCM's effect on pituitary LH secretion
in rats may be analogous to a potential effect on placen-
tal hCG secretion in humans. Further, because hCG and
LH bind to the same luteal receptor, it is important to
consider the hypothesis that BDCM also may decrease
luteal responsiveness to LH/hCG To test this hypoth-
esis, rat corpora lutea (CL) were examined ex vivo for
their ability to secrete progesterone following stimula-
tion with hCG BDCM had no effect on luteal respon-
siveness to hCG; however a "rebound" increase in ex
vivo progesterone secretion was evident in CL previ-
ously exposed to BDCM in vivo. This finding was unex-
pected in view of the reduced serum progesterone levels
observed following BDCM exposure. It is unclear if this
rebound effect may reflect the removal of the CL from a
possible inhibitory influence of BDCM.
Finally, in contrast to the F344 strain, this research
group demonstrated that Long Evans rats, and especially
Sprague-Dawley rats, are remarkably less sensitive to
BDCM-induced pregnancy loss. It is planned to pursue
these differences in strain sensitivity as a research tool
that may provide insights into susceptible subpopulations
as well as BDCM's mode of action.
1 Waller K, Swan SH, DeLorenze Q Hopkins B. Trihalomethanes in drinking water and spontaneous abortion. Epidemiology
1998:9:134-140.
The Office of Research and Development's National Center for Environmental Research
29
-------
Endocrine Disruptors Program Review Workshop
Assessing Consequences of Embryonic Exposure to Methoxychlor:
Neuroendocrine and Behavioral Measures
Mary Ann Ottinger, Julie Hazelton, Julie Wu, Michael Ruscio, Nichola Thompson,
Michael Quinn, and Mahmoud Abdelnabi
Department of Animal and Avian Sciences, University of Maryland, College Park, MD
Endocrine-disrupting chemicals (EDCs) include pes-
ticides, herbicides, and other chemicals that interact with
endocrine systems. This research project focused on meth-
oxychlor (MXC) and other EDCs, as well as on effects of
the estrogenic compounds and on estradiol as a positive
control to assess potential impact of estrogenic EDCs.
The male Japanese quail is exquisitely sensitive to the ef-
fects of exogenous estradiol during sexual differentiation,
resulting in impaired sexual behavior and reduced fertility.
Moreover, exposure of two successive generations af-
fected the untreated offspring.
A series of studies were conducted on Japanese quail,
with exposure to MXC, a widely used pesticide. Expo-
sure was either by egg injection (0, 150, or 300 ug) at
embryonic day 4 or via the diet (0, 0.5, or 5 ppm) in a
two-generation experimental design. Embryonic exposure
resulted in impaired sexual behavior in the adult male Japa-
nese quail. Moreover, long-term effects on hypothalamic
catecholamine systems and on the gonadotropin-releasing
hormone-I system were detected. Dietary exposure did
not affect the parent generation; however, both the Fl
(MXC treated) and F2 (no MXC treatment) offspring
showed effects of MXC exposure. Although fertility and
other general indicators of reproduction did not appear to
be affected, other endpoints critical to reproduction were
impacted by MXC, including plasma steroid hormones,
sexual behavior, and sexual maturation. These data sug-
gest that the embryo is the most sensitive stage for expo-
sure to EDCs, with effects on a subset of reproductive
endpoints.
Bobwhite quail have been a species of choice for as-
sessing effects of toxic compounds in birds. However,
little is known about EDC effects on reproductive end-
points in Bobwhite quail. Therefore, Bobwhite and Japa-
nese quail were simultaneously compared in a two-
generation study, with dietary exposure to MXC (0, 5
ppm, and 10 ppm). The parent generation was raised un-
der short photoperiod, paired, and transferred to long days
(16L:8D) with initiation of treatment. Basic measures of
health and reproduction were monitored, including feed
intake, egg production, fertility, and offspring viability.
Chicks (Fl) were raised on the same diet as their par-
ents and observed for sexual maturation, reproductive be-
havior, and endocrine endpoints. Results showed species
differences in maturation, with Bobwhite quail requiring
3-4 weeks longer to achieve sexual maturity. Neither spe-
cies showed effects of 5 ppm MXC on egg production,
fertility, body weight, feed intake, or chick viability. How-
ever, Japanese quail fed MXC matured more slowly, sug-
gesting that the treatment interfered with activation of re-
production. A separation test was used in young chicks to
assess motivation to rejoin siblings. These data demon-
strate the importance of examining individual responses to
EDCs. hi summary, neuroendocrine and behavioral mea-
sures are likely to provide reliable endpoints that relate to
embyronic EDC exposure. These measures are particu-
larly relevant for consideration in assessing the long-term
impact of EDCs on birds because these endpoints are
sexually dimorphic and organized under the influence of
steroid hormones during embryonic development.
30
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disruptors Program Review Workshop
Endocrine Disruptors and Testis Development
Michael K. Skinner
Center for Reproductive Biology, School of Molecular Biosciences,
Washington State University, Pullman, WA
Endocrine disrupters have been shown to influence
male reproduction by causing abnormal sperm numbers
and fertility. One of the most sensitive periods to endo-
crine disrupter exposure is during embryonic develop-
ment. The objective of the current research is to investi-
gate the mechanism of action of amodel endocrine disrupter
on male reproduction with a focus on testis development.
A rodent model system is used to provide insight into the
mechanistic aspects of endocrine disrupter action. The
model endocrine disrupter tested is methoxychlor, which
has metabolites that are both weak estrogenic and anti-
androgenic compounds. Therefore, this model endocrine
disrupter allows consideration of both estrogenic and anti-
androgenic endocrine disrupter actions. Comparisons with
standard estrogenic and androgenic steroid agonists and
antagonists also are made. The objective is to obtain in-
sight into the molecular, cellular, and physiological actions
of endocrine disrupters on male reproduction. The hy-
pothesis tested is that endocrine disrupters (i.e., meth-
oxychlor) influence embryonic testis development via ste-
roid receptors to interfere with critical growth factor-me-
diated cell-cell interactions that result in abnormal germ
cell differentiation through epigenetic effects (e.g., DNA
methylation), and that this subsequently influences adult
spermatogenesis and is transgenerational through the
germline.
Studies have shown that methoxychlor can affect em-
bryonic testis development at the time of testis morpho-
genesis, and that this causes an increase in germ cell
apoptosis in the adult. Interestingly, observations suggest
this abnormal spermatogenesis is transgenerational, and
preliminary data suggest altered DNA methylation of the
germline may be the epigenetic action of the endocrine
disrupter. Preliminary studies also have demonstrated that
two families of paracrine growth factors directly influ-
ence testis development at the time of methoxychlor ac-
tion. Abnormal testis development and germ cell differen-
tiation caused by endocrine disrupters may be due in part
to inappropriate control of these critical growth factor-
mediated cell-cell interactions. Preliminary studies indi-
cate that the transforming growth factor (TGF) families
are critical for embryonic testis growth. Inhibition of the
TGF factors blocks testis growth and results in abnormal
testis development and potential subfertility. Preliminary
studies also indicate that the neurotropin (NT) family of
factors (i.e., NT3) has a critical role in the morphogenesis
of testis development (i.e., sex cord or seminiferous tu-
bule formation). This non-neuronal action of the NTs when
blocked inhibited normal testis development and morpho-
genesis, which may result in abnormal germ cell differen-
tiation and subfertility in the adult male. Methoxychlor was
found to alter the expression of these growth factors.
Abnormal control of critical testis cell-cell interactions af-
ter treatment with different doses of endocrine disrupters
is anticipated to influence germ cell development and male
fertility.
Information obtained provides insight into how envi-
ronmental toxins (i.e., methoxychlor) may impair male
fertility by adversely effecting gonadal NTs and transform-
ing growth factors. These studies develop a better under-
standing of the mechanistic aspects of how endocrine
disrupters influence reproductive function (i.e., testis
growth and function). The research will be discussed to
extrapolate and provide insight into the impact of endo-
crine disrupters on human development, reproduction, and
health.
The Office of Research and Development's National Center for Environmental Research
31
-------
Endocrine Disruptors Program Review Workshop
Effects of Early Developmental Exposure to Endocrine-Disrupting
Chemicals on Reproductive Function in the Adult Male: Rabbit Model
D.N. Rao Veeramachaneni1, Colleen M. Kane1, Ty T. Higuchi1, Jennifer S. Palmer1,
Ginger E. Sammonds1, and K-Y. Francis Pau2
'Colorado State University, Fort Collins, CO; 2Oregon Regional Primate Research Center, Beaverton, OR
This research project focused on developmental and
long-term reproductive sequelae of prenatal plus infantile
exposure of male rabbits to pesticides that are known to
competitively bind to androgen receptor and thus exert
anti-androgenic activity. Rabbits were chosen as a model
because this species, unlike rodents, has a relatively long
quiescent period of reproductive development before pu-
berty, mimicking that of humans. Moreover, longitudinal
evaluations of endocrine profiles, sexual capacity, and se-
men quality are feasible in rabbits.
Groups (n = 7-10) of Dutch-belted rabbits were treated
daily beginning from gestation day 15 through post-kin-
dling week 4 with p,p'-dichlorodiphenyltrichloroethane
(p,p'-DDT), vinclozolin, or amixture of both. The chemi-
cals were administered orally in corn syrup at 0 (control),
25 (low dose), or 250 (high dose) umol/kg body weight.
Testicular descent to a scrotal-dependent position was
monitored by palpation, beginning at 4 weeks after birth.
One of the 15 low-DDT pups, 4 of the 16 high-DDT
pups, 1 of the 17 low-mixture pups, and 3 of the 18 high-
mixture pups were cryptorchid. Histological evaluation
revealed the presence of atypical germ cells, some resem-
bling carcinoma in situ, in the undescended testes. Begin-
ning at 20 weeks of age, six seminal ejaculates were col-
lected, one every 3-4 days, by using an artificial vagina.
Ability to accomplish ejaculation was recorded by moni-
toring outcome (interest to mount, penile erection, and
time taken to ejaculate) within 3 minutes after introduc-
tion of afemale teaser. Two of the 7 low-vinclozolin groups
and 2 of the 9 low-mixture groups showed absolutely no
sexual interest and never ejaculated. One each of 8 low-
DDT and 8 low-vinclozolin rabbits and 3 out of 10 high-
vinclozolin rabbits failed to ejaculate at least once. The
serum concentrations of luteinizing hormone and testoster-
one in any of the treatment groups did not differ from that
of the control (p > 0.01). However, follicle-stimulating
hormone was consistently lower in vinclozolin-treated rab-
bits (p < 0.05). This indicates that the impairment of an-
drogen-dependent events resulted not from lack of test-
osterone, but possibly from unavailability of the receptor
because of either its downregulation or chronic occu-
pancy by the pesticides or their metabolites.
hi summary:
• p,p'-DDT, but not vinclozolin, caused cryptorchid-
ism. Atypical germ cells that resemble carcinoma in
situ were evident in DDT-induced undescended tes-
tes.
• Developmental exposure to anti-androgenic pesticides
caused sexual dysfunction in the male; lack of sexual
interest and ejaculation failures were more prevalent
in vinclozolin-exposed rabbits.
• Impairment of androgen-dependent events may have
resulted from unavailability or dysfunction of andro-
gen receptor because serum testosterone was unaf-
fected.
These results indicate that exposure of males to anti-
androgenic pesticides during critical periods of develop-
ment impairs reproductive function as adults.
32
The Office of Research and Development's National Center for Environmental Research
-------
Exposure, Risk Assessment, and
Risk Management
-------
Endocrine Disrupters Program Review Workshop
Cross Species Mode of Action Information Assessment
for Bisphenol A
Susan Y. Euling and Babasaheb Sonawane
National Center for Environmental Assessment, Office of Research and Development,
U.S. Environmental Protection Agency, Washington, DC
The goal of this project was to identify mode of action
(MOA) information for Bisphenol A (BPA) across animal
species and to explore how to integrate this information for
the evaluation of ecological and human health risks. This
approach is predicated on the idea that common MOAs are
a function of evolutionary relationships. BPA was selected
because it is a high production volume chemical that can
act as an endocrine-disrupting chemical (EDC) and there
are some effects data in some nonmammalian and inverte-
brate species. BPAMOAinformationfor developmental and
reproductive effects was gathered from the published lit-
erature. BPA in vivo effects data were identified for only 17
species (of the total 9 or 10 million) representing 7 animal
classes: gastropods, crustaceans, insects, amphibians, fish,
birds, and mammals. For invertebrates, the MOA for BPA
could not be determined, hi gastropods, BPA treatment led
to sexual differentiation effects, but data to establish an MOA
were lacking. For the crustaceans, no consistent effects
were observed in daphnids; in copepods, effects were con-
sistent with an estrogen (E) agonist MOA; and for insects,
data were dissimilar from E2 effects and data from an in
vitro assay suggest an ecdysone antagonist MOA.
Although some invertebrate species showed effects
from E or BPA exposure, the significance of the findings
is unclear because the role of E (or E analogs) in normal
invertebrate development is not known. Within the verte-
brates, in vitro studies found that BPA competitively binds
to the estrogen receptor (ER) of reptiles, amphibians, fish,
birds, and mammals at a much lower affinity (250-12,500
times lower) than ethinyl estradiol. For amphibians, some
developmental effects data were consistent with BPA act-
ing as an E agonist, and in vitro data suggest a thyroid
hormone bioavailability MOA. In birds, in ovo BPA treat-
ment led to some significant sexual differentiation effects,
consistent with an E agonist MOA. In fish, BPA in vivo
study findings were consistent with an E agonist, andro-
gen (A) antagonist, an A agonist, steroid hormone bio-
availablity, and/or anonendocrine MOA. Mechanistic data
in fish support BPA treatment leading to an increased E
activity, including E agonism. Mammalian in vivo effects
data, limited to rat and mice, are consistent with an E
agonist, A antagonist, A agonist, altered E bioavailability,
altered A bioavailability, altered thyroid hormone, and/or
altered prolactin hormone MOA; whereas most mecha-
nistic studies support an E agonist and one study supports
a decreased A bioavailability MOA. Human and rodent in
vitro ER-binding data indicate that BPA can act as an E
agonist.
Together, the data indicate that BPA exposure can elicit
effects in vertebrate and invertebrate species at sublethal
concentrations, and the majority of the evidence supports
BPA acting to increase E or E analog activity with mecha-
nistic studies supporting E agonism. Limitations of the
BPA assessment were the small number of species with
data, the focus of most studies on the E MOA, and the
lack of knowledge of the role of hormones in invertebrate
sexual differentiation as well as the role of estrogens in
mammalian male development. Studies to investigate
whether BPA affects alternative or multiple MOAs at dif-
ferent doses and under different conditions are needed.
When BPA MOA and effects data for additional animal
species have been generated, the MOA evolutionary rela-
tionships among species could be used to make predic-
tions of MOA to untested species.
The Office of Research and Development's National Center for Environmental Research
35
-------
Endocrine Disruptors Program Review Workshop
Overview of Intramural Exposure Research Program
V. Ross Highsntith, Greg Toth, and Lawrence Burns
National Exposure Research Laboratory, U.S. Environmental Protection Agency,
Research Triangle Park, NC
The National Exposure Research Laboratory (NERL)
conducts research to develop and validate exposure tools
for characterizing human and ecological exposures to en-
docrine-disrupting chemicals (EDCs). NERL's Methods
Development Research Program includes enhanced ana-
lytical methods for assessing ecological exposures to
alkylphenols and related compounds as well as innovative
DNA microarray techniques to rapidly identify first-order
ecological effects resulting from exposures to estrogenic
compounds. Large-scale exposure field measurement
studies also are being planned throughout the United States.
Several studies have been conducted to assess children's
exposures to EDCs, pesticides, and other organic pollut-
ants. A longitudinal study examining children's aggregate
exposures to EDCs and other pollutants is being planned
for Jacksonville, FL, in Fiscal Years 2003-2005. Ecologi-
cal field studies also are being conducted to characterize
the extent of EDCs and pesticide exposures in eastern,
midwestern, and western ecological watersheds. Expo-
sure modeling research includes enhancements of the
current AGDISP (spray drift) model to include additional
application techniques, secondary volatization, and regional
transport of agricultural pesticides. NERL's primary eco-
logical and human exposure models (PRZM/EXAMS and
SHEDS/ERDEM, respectively) also are being upgraded
for use in characterizing and assessing EDC exposures.
This overview presentation will highlight NERL's current
and future EDC exposure research programs.
36
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disruptors Program Review Workshop
Development of Molecular Diagnostic Indicators of Exposure to
Estrogenic and Androgenic Endocrine-Disrupting Chemicals
D.L. Lattier, A Miracle, T.V. Reddy, J.L. Lazorchak, and GP Toth
Environmental Exposure Research Division, National Exposure Research Laboratory, Office of Research
and Development, U.S. Environmental Protection Agency, Cincinnati, OH
The extent to which humans and wildlife are exposed
to endocrine-disrupting chemicals (EDCs) is an important
focus of environmental research. This work has been di-
rected toward the development of molecular indicators di-
agnostic for exposure to EDCs in freshwater fish. Research
includes the discovery of genes indicative of environmental
exposure in the U.S. Environmental Protection Agency's
(EPA) long-established aquatic toxicological organism, the
fathead minnow (Pimephalespromelas). Novel cDNAs and
gene sequences will be used in DNA microarray analyses
for pattern identification of stressor-specific, differentially
up- and downregulated genes. The methods currently used
to discover genes in this organism, for which few anno-
tated nucleic acid sequences exist, are cDNA subtraction
libraries, differential display, exploiting polymerase chain
reaction (PCR) primers for known genes of other mem-
bers of the family Cyprinidae, and use of degenerate PCR
primers designed from regions of moderate protein homol-
ogy. To date, hundreds of fathead minnow cDNA sequences,
resulting from exposure to estrogenic compounds, have
been isolated by subtractive hybridization cloning, and a
preliminary glass-based DNA microarray has been con-
structed to which gene sequences will be added as further
discovery and characterization proceed.
As the number of genes that will detect stressor-specific
transcriptome changes continues to expand, future micro-
array spot printing of covalently bound DNA will use over-
lapping oligonucleotides as hybridization targets, replacing
the current cDNA platform. This approach will increase
efficiency of hybridization, sensitivity of detection, and aid
in addressing inherent issues of reproducibility. Single or
multiple genes noted as being differentially expressed in
microarray analyses then will be used in separate studies to
measure bioavailable stressors in the laboratory and field.
These analyses will be accomplished by "real time," quan-
titative PCR The expression of some genes (e.g., vitellogenin
in male fish and the androgen receptor gene) that indicate
exposure to estrogens and androgens, respectively, have
been experimentally confirmed in the laboratory and used in
a number of environmental EDC evaluations.
Currently, field and laboratory studies are being con-
ducted, including, but not limited to: (1) mesocosm expo-
sure experiments to measure variation in estrogen bio-
availability as influenced by primary productivity in aquatic
ecosystems, (2) monitoring studies of potential estrogenic
EDCs in effluent samples taken from 50 U.S. sewage treat-
ment plants, (3) exposure and effects studies of androgenic
activity in effluents collected at concentrated animal feeding
operations, and (4) surface water and effluent monitoring
studies in EPA Region 9, subsequent to a recently initiated
cooperative technology transfer program between the Na-
tional Exposure Research Laboratory/Office of Research
and Development and EPA Region 9. Positive results in
these studies present a range of possibilities for those genes
identified by DNA microarray analyses to be critical com-
ponents of toxicity pathways for stressors having various
modes of action. The ability to discriminate bioavailability
of mixture components and their association with adverse
effects, downstream from these early molecular events,
presents new ground to be broken in EDC risk assessment.
The Office of Research and Development's National Center for Environmental Research
37
-------
Endocrine Disruptors Program Review Workshop
Intramural Research on Risk Management of Wastewater
Treatment Sources
Marc A. Mills, Gregory D. Sayles, and Andrew P. Avel
National Risk Management Research Laboratory, U.S. Environmental Protection Agency, Cincinnati, OH
Extensive data in the literature indicate that certain endo-
crine-disrupting chemicals (EDCs) commonly appear in
surface waters. For example, a recently published report
from the U.S. Geological Survey's National Reconnaissance
Study showed that 90 percent and 70 percent of the surface
water locations tested in the United States showed measur-
able levels of steroid hormones or detergent metabolites,
respectively. Other published data sets have revealed that
municipal wastewater treatment plants receive steroid hor-
mones and alkylphenol ethoxylate surfactants, and that in
the plants, the hormones often are not completely destroyed
and the surfactants are biotrans-formed into estrogenic me-
tabolites. Wastewater treatment plants are potential sources
of EDCs to the environment by two routes: aqueous effluent
and the disposal of biosolids (sludge). Amajor portion of the
U.S. Environmental Protection Agency's research program
on risk management of EDCs is focused on understanding
wastewater treatment plants as sources of EDCs to the en-
vironment and, as needed, developing approaches to mini-
mize this source. To address this complex problem, several
projects have been undertaken:
• Fate of EDCs in Wastewater Treatment Processes.
A study is underway to understand the fate of EDCs
in the various unit processes within treatment plants
by constructing and operating two secondary treat-
ment pilot plants (one with aerobic digestion and the
other with anaerobic digestion). To date, the plants are
fully operational, and analytical methods have been de-
veloped for all sampling streams. Steroid hormones
and alkylphenol ethoxylate surfactants are to be intro-
duced to the plant this month.
• Removal of EDCs by Sludge Digesters. As part of a
larger National Risk Management Research Labora-
tory (NRMRL) biosolids research study, this research
project will sample several full-scale sludge digesters
to correlate the ability of digesters to remove steroid
hormones and alkylphenols in sludge as a function of
operational process variables.
• Removal of EDCs by Land Application of Biosolids.
Most commonly, biosolids are disposed of by land
application. Also part of the NRMRL biosolids study,
the ability of land application to treat steroid hormones
and alkylphenols in biosolids will be evaluated for typi-
cal land application techniques.
• Characterizing Wastewater Treatment Plants and
Their Effluents. A national screening study is under-
way by NRMRL, National Exposure Research Labo-
ratory, and the Regional offices, in which up to 50
municipal treatment plant effluents will be screened
for endocrine-disrupting character and related to char-
acteristics of the plant. The endocrine-disrupting char-
acter will be determined by a vitellogenin gene expres-
sion assay and by analytical chemistry for steroid hor-
mones and alkylphenols. To date, 30 effluents have
been assessed.
• Potential of Sediments To Remove EDCs. EDCs that
are not removed by treatment may partition from the
receiving surface water in aquatic sediments. Labora-
tory studies are underway to evaluate the ability of sedi-
ments to transform or destroy steroid hormones and
alkylphenols under various typical redox conditions.
The presentation will summarize the wastewater treat-
ment program including its goals, approaches, and re-
sults to date.
38
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disruptors Program Review Workshop
Endocrine-Disrupting Chemicals Related to Feminization of Males
Debdas Mukerjee
National Center for Environmental Assessment, Office of Research and Development,
V.S Environmental Protection Agency, Cincinnati, OH
Observations in wildlife on the potential of certain en-
vironmental chemicals to modulate endocrine-regulated
processes have led researchers to question whether simi-
lar abnormalities are occurring in humans. Currently, there
is a growing realization that a wide range of anti-andro-
genic endocrine-disrupting chemicals (EDCs) is associ-
ated with feminization of birds, fish, and alligators. Male
fish with DNA code of the Y chromosome and elevated
vitellogenin levels have decreased serum testosterone con-
centrations. Feminization of veterinary animals is not un-
common. Various degrees of male genital abnormalities
from mild to moderate hypospadias, unilateral or bilateral
cryptorchidism, and poorly developed testes are common
in dogs, goats, horses, and other domestic animals.
Parental in utero or perinatal exposure of experimental
animals to EDCs, namely dichlorodiphenyltrichloroethane,
dichlorodiphenyldichloroethylene, di(2-ethylhexyl) phthalate,
flutamide, and vinclozolin, can result in hypospadias, cryp-
torchidism, and other male sexual abnormalities. Epidemio-
logic studies have reported an increased risk of male genital
malformation in children of workers exposed occupation-
ally to pesticides and clustering of cryptorchidism in areas
of intensive agriculture/horticulture where various types of
pesticides are used. A marked increase in the incidence of
hypospadias in the United States, European countries, and
Japan was observed during the 1960s, 1970s, and 1980s at
a time when exposure to anti-androgenic EDC levels in the
environment was high. Hypospadias and cryptorchidism
are occurring approximately at arate of 1 in 125 male births
in the United States, hi humans, the male sexual disorders,
including hypospadias and cryptorchidism, are symptoms
of one underlying entity, the testicular feminization syndrome.
Animal data indicate that the parental or in utero or perinatal
exposure to EDCs can result in various degrees of femini-
zation of male offsprings with testicular feminization ex-
pressed as hypospadias, cryptorchidism with undescended
testis, and related intersex conditions.
To examine whether male sexual abnormalities and
their different degrees of manifestations are associated
with parental, in utero (at different stages of organogen-
esis), or perinatal exposures to anti-androgenic EDCs, the
U.S. Environmental Protection Agency will convene a
workshop in Cincinnati, OH, on December 4 and 5,2002.
The primary objective of this workshop is to determine
the rationale and approach for exposure and risk assess-
ments of environmental anti-androgenic EDCs.
The Office of Research and Development's National Center for Environmental Research
39
-------
Endocrine Disruptors Program Review Workshop
Overview of the Intramural Risk Management Research Program
Gregory D. Sayles and Andrew P.Avel
Office of Research and Development, National Risk Management Research Laboratory,
U.S. Environmental Protection Agency, Cincinnati, OH
This presentation will provide a summary of the risk
management portion of the Office of Research and
Development's endocrine-disrupting chemicals (EDCs)
research program, including its motivation, goals, plan-
ning efforts, and resulting research areas.
In an emerging research area such as EDCs, risk
management (RM) research and risk assessment (ef-
fects, exposure, characterization) research can and should
be conducted in parallel. Early or tentative effects and
exposure information can point to the identification and
development of risk management approaches that may
prevent or control risks with high effectiveness at low
cost. These early risk management approaches may pos-
sess relatively large uncertainty in their effectiveness at
reducing risk because the risk information is relatively
uncertain. However, if risk managers must make a deci-
sion based only on limited risk information, risk manage-
ment approaches with large uncertainties may be more
attractive than no options. As the knowledge of risk
matures, the approaches for risk management become
increasingly certain.
Consistent with this philosophy, the RM research
program began with the development of the Risk Man-
agement Evaluation (RME) of Endocrine Disrupting
Chemicals. This RME provides a snapshot in time of the
current state of risk management (i.e., what currently
available risk management approaches can be adapted to
address EDCs, and what new RM technologies or strat-
egies need development?). This document should be
available to the public by the end of 2002. Because risk
information is continually evolving and risk management
technologies and approaches are continually growing and
maturing, the RME for EDCs will be updated on a regu-
lar basis.
The RME provided a list of research questions that
have been used to plan the RM research program. The
program is characterizing sources of EDCs to the envi-
ronment, developing strategies to minimize these sources,
and investigating strategies to remediate EDCs in envi-
ronmental reservoirs. Specifically, research is being con-
ducted in the following areas:
• Studying wastewater (sewage) treatment as a source
of EDCs, and developing approaches to improve
treatment as necessary.
• Characterizing concentrated animal feeding operation
as sources of EDCs and developing animal waste
treatment approaches as necessary.
• Characterizing combustion processes as sources of
EDCs, and developing improvements to the process
to minimize this source.
• Studying the natural attenuation of alkylphenols in
sediment impacted by wastewater treatment plant
outfalls.
• Evaluating the ability of conventional and advanced
drinking water treatment plants processes to remove
EDCs, and improving these processes if necessary.
• Developing pollution prevention tools that can be used
to evaluate and nominate substitute chemicals for
EDCs currently in use.
• Adaptation of EDC screening bioassays for use in
determining the performance of RM technologies.
40
The Office of Research and Development's National Center for Environmental Research
-------
Field Studies
-------
Endocrine Disrupters Program Review Workshop
Frog Deformities: Role of Endocrine Disruptors
During Development
David M. Gardiner1, Aristocle Ndayibagira1, Felix Grun1, David Hoppe2,
and Bruce Blumberg1
University of California, Irvine, CA;2 University of Minnesota, Morris, MN
There is considerable evidence that wild animals have
suffered adverse consequences from exposure to environ-
mental chemicals that interact with components of the en-
docrine system. The high incidence of deformed frogs in
North America, coupled with the worldwide decline in the
occurrence of amphibian species, suggests that environ-
mental modification is negatively impacting amphibian popu-
lations. Although the cause of amphibian declines in rela-
tively pristine environments remains unknown, there is an
emerging consensus that the increasing prevalence of de-
formed frogs is the result of a waterborne contaminant that
has appeared, or reached a critical concentration, in recent
years. The objective of this research project is to assess the
significance of endocrine disrupters that activate retinoid-
signaling pathways for their role in causing limb develop-
mental deformities in frogs, and to understand their mecha-
nism of action to assess their implications for human health.
This research group analyzed the skeletal dysplasias
observed in severely affected frogs from Minnesota and
identified two classes of common limb abnormalities.
First, supernumerary or absent limbs, suggesting that
the process of limb initiation is being affected. Second,
skeletal abnormalities, including truncated and phocomelic
limbs, suggesting that limb growth and pattern forma-
tion also are being modified, hi the phocomelic limbs,
the skeletal elements are folded back on themselves, such
that the proximal and distal ends of the bone lie adjacent
to one another and the mid-portion of the bone projects
laterally, forming a "bony triangle." Retinoid treatment
at sensitive stages phenocopies deformities in wild frogs.
The effects of treating a range of stages of larval devel-
opment with retinoids was tested, and it was found that
all the deformities that are observed in wild populations
of frogs can be induced by experimental exposure to
retinoids. Using the parameters established in these stud-
ies, a developmental toxicology assay has been devel-
oped to screen the activity of a number of chemicals
known to be present at sites where deformed frogs are
found.
If environmental retinoids are the cause of frog de-
formities, then retinoids will be found at sites where de-
formed frogs are found. To test this hypothesis, hydro-
phobic substances will be extracted from water samples,
and then fractionated by high-performance liquid chro-
matography and tested for their ability to activate the
retinoic acid receptor in transient transfection assays.
Active fractions will be purified to homogeneity as judged
by ultraviolet absorption spectra and then analyzed by
electrospray and electron impact mass spectroscopy for
exact mass determination. Candidate compounds thus
identified then will be retested in the reporter and animal
assay to verify biological activity. Similar activity peaks
have been discovered in water samples from a vernal
pond in Mission Viejo, California, as in the permanent
lake being studied in Minnesota.
hi addition to the laboratory studies, these research-
ers have continued to monitor field sites in Minnesota. At
the Crow Wing County, Minnesota, site, all ranid frog
species have declined during the 3 years of the study.
This past spring, no leopard frog or green frog calls
were heard, and only scattered mink frog calls were
heard. Only one leopard frog juvenile was found at the
site during the entire 2000 season (compared to 562 in
1997, a year of similar sampling effort, for example).
Mink frog capture success also was decreased; only 74
juveniles were captured in 2000, compared to 365 in
1997. The mink frog was the only species with enough
juvenile captures to calculate meaningful malformation
frequencies. The total malformation frequency among
juveniles in 2000 was 18 percent, apparently lower than
1996-1999, when the frequency ranged from 50-75 per-
cent, but still much higher than the working "back-
ground" frequency of 1 percent.
The Office of Research and Development's National Center for Environmental Research
43
-------
Endocrine Disruptors Program Review Workshop
The Mechanisms and Effects of Endocrine Disruption on Infertility
in the Bonnethead Shark on Florida's Gulf Coast
Charles A. Manire1, Jim Gelsleichter1, Lois E.L. Rasmussen2, Enric Cortes3, Ian Tebbett5,
David L. Hess4, and Nancy Szabo5
'Mote Marine Laboratory, Sarasota, FL;2Oregon Graduate Institute, Portland, OR; ^National Marine Fisheries
Service, Panama City, FL; ^Oregon Regional Primate Research Center, Beaverton, OR; ''Analytical Toxicology
Core Laboratory, University of Florida, Gainesville, FL
Previous studies have demonstrated a high frequency
of infertile ova in the uteri of pregnant bonnethead sharks
(see Figure 1) from the Tampa Bay area, a highly industri-
alized region on Florida's Gulf Coast. Whereas infertility
typically is rare in most shark species, its occurrence in
approximately 75 percent of pregnant bonnethead sharks
examined in recent surveys of affected populations sug-
gests that it may have detrimental effects on population
growth. The objectives of this study were to: (1) deter-
mine if infertility observed in these populations is associ-
ated with disruption of the endocrine system; (2) identify
the mechanism(s) underlying the production of infertile
ova; (3) determine if infertility is correlated with levels of
environmental contaminants in shark tissues; and (4) esti-
mate the effects that infertility may have on the rate of
population growth.
To address these objectives, levels of environmental
contaminants, trends in growth and reproduction, and rates
of increase for three separate populations of bonnethead
sharks on Florida's Gulf Coast were investigated. Sharks
were collected from three dissimilar geographical regions:
(1) the Anclote River, a site adjacent to Tampa Bay that is
known to contain high levels of environmental pollutants
and high rates of infertility in resident shark populations;
(2) Apalachicola Bay, a site on the northwest coast of
Florida that possesses moderate levels of environmental
contamination; and (3) Florida Bay, a site that is known to
possess relatively low levels of environmental pollutants
and low rates of shark infertility.
The reproductive competence of mature male and fe-
male sharks was evaluated using a variety of indices, in-
cluding measures of gonadal development, semen quality,
female sperm storage, and fertility. The relationship be-
tween reproductive success and endocrine function was
investigated using measurements of serum steroid con-
centrations. Associations between reproductive success
and levels of environmental contamination were evaluated
by measuring the concentrations of organochlorine pesti-
cides (OCs) and polychlorinated biphenyls (PCBs) in shark
tissues. Lastly, differences in population growth were de-
termined by incorporating estimates of natural mortality,
age at maturity, lifespan, and fecundity (which is influ-
enced by fertility rate) into demographic models for the
three study populations.
The reproductive biology of male bonnethead sharks
from the three study sites did not differ significantly. In
contrast, the ability of female sharks to store spermato-
zoa prior to fertilization was significantly lower in sharks
from the Anclote River. Differences in sperm storage,
fertility and, perhaps, other undetected disparities in fe-
male reproduction appear to be related to differences in
endocrine function, based on low serum concentrations
of 17B-estradiol measured in Anclote River sharks spe-
cifically during the period of sperm storage and vitello-
genesis. An association between high rates of infertility
and levels of contaminant exposure was observed, based
on significantly elevated concentrations of total PCBs
and more than 10 OCs in Anclote River sharks. Differ-
ences in fertility do not appear to have a profound effect
on population growth, based on demographic analyses.
However, the effects of infertility on population growth
are likely to be greater in most other regional shark spe-
cies, which display lower rates of population increase.
44
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disruptors Program Review Workshop
Figure 1. Reproductive tract and embryos from a pregnant female bonnethead shark from the Tampa
Bay region of Florida's Gulf Coast. Several infertile eggs also are present in both uteri.
Embryos are attached to uteri by placental-like attachments, which provide nourishment
during the second half of pregnancy (embryos are dependent on yolk for nourishment prior
to development of these connections).
The Office of Research and Development's National Center for Environmental Research
45
-------
Endocrine Disruptors Program Review Workshop
Endocrine Disruption in Marine Gastropods by Environmental
Chemical Mixtures
Patricia D. McClettan-Green1'2 and Eva Oberddrster3
Department of Environmental and Molecular Toxicology, North Carolina State University,
Raleigh, NC; 2Duke University Marine Laboratory, Beaufort, NC; Department of Biology,
Southern Methodist University, Dallas, TX
The masculinization of female snails, termed imposex,
occurs worldwide in more than 150 species following
exposure to low levels of tributyltin (TBT) (as low as 1
ng/L).1 Females exposed to TBT grow accessory sex
organs (ASO), including sperm ducts, seminal vesicles,
external sperm grooves, and most notably, penises. At
least two mechanisms have been proposed for the devel-
opment of imposex: one involves the abnormal release of
neurohormones that control sexual maturation and repro-
duction in molluscs; the other is based on the vertebrate
model of steroid hormone regulation.
The first model of imposex induction centers on the
concept that TBT acts on the ganglia of snails to alter the
expression of neuropeptide hormones. TBT is thought
to control the expression/activity of penis morphogenic
factor (PMF), a neuropeptide that controls the produc-
tion of male accessory sex organs in gastropods. Injec-
tion of a putative PMF into female mud snails resulted in
a significant induction of imposex. In addition, snails
exposed to TBT contain higher levels of immunodectable
APGWamide (the putative PMF). This laboratory also
has shown that TBT decreases the number of egg cap-
sules laid by female snails. Egg laying is controlled by a
36-amino acid neuropeptide (ELH) produced in the gan-
glia of gastropods. It is possible that TBT functions on a
signal transduction pathway common to both peptide
hormones.
hi the second model of imposex induction, TBT is
proposed to function as a noncompetitive inhibitor of cy-
tochrome P450 metabolism. This enzymatic inhibition
blocks the metabolism/conversion of steroids, ultimately
leading to a buildup of testosterone and the subsequent
development of ASO in female snails.2 Studies in this labo-
ratory have shown that TBT significantly reduces the lev-
els of each steroid (pregnenolone, progesterone, and an-
drostenedione) produced in the mud snail. In addition,
there is an approximately 50 percent reduction in aromatase
activity in snails after exposure to 20 ng/L of TBT for 45
days and in environmentally exposed animals.
From these studies, it is evident that both mechanisms
play arole in the induction of imposex in gastropods, even
at extremely low doses of TBT. It is likely that a positive
feedback loop mechanism stimulated by both processes
functions in the development and maintenance of these
structures. TBT would stimulate the release of PMF to
induce imposex in females, while steroids would act on
the feedback loop to maintain these structures.
'Oberdorster E, McClellan-GreenP. The neuropeptide APGWamide induces imposex in the mud snail, Ilyanassa obsoleta.
Peptides2(m;2l(9):l323-30.
Tent K.Ecotoxicology of organotin compounds. CritRevToxicoll996',26(l):l-lll.
46
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disruptors Program Review Workshop
Effects of PCBs on Herring Gulls in the Field
and Chickens in the Laboratory
EM. Anne McNabb
Virginia Polytechnic Institute and State University, Blacksburg, VA
The objectives of this study were to investigate the
effects of polychlorinated biphenyl (PCB) exposure on
thyroid function in developing Herring gulls from Great
Lakes (GL) sites and in developing chickens from PCB-
dosed eggs. Herring gulls, a fish-eating top-predator, have
been used as a sentinel species to monitor contaminant
effects in the GL for more than 30 years. Because GL
gulls have had thyroid hypertrophy and histopathology,
and because PCBs disrupt thyroid function in laboratory
mammals, it has been speculated that PCB-exposed GL
gulls may have depressed thyroid function despite de-
creases in PCBs and other organochlorines since the 1970s.
This research group assessed organismal thyroid status
based on plasma thyroid hormones (THs), thyroid gland
(TG) function based on TG-TH content, and activation of
the hypothalamic-pituitary-thyroid (HPT) axis based on TG
weight (TG hypertrophy indicates a negative feedback re-
sponse to low circulating THs). Some of these studies have
addressed mechanisms by which thyroid disruption may
occur (enhanced liver glucuronidation of T4, and displace-
ment of THs from plasma-binding proteins) and whether
hormone activation responses may be compensating for
decreases in THs (brain 5' deiodinase activity).
Herring gulls collected from GL in 1998-2000 show
that both pipping embryos and prefledglings from high
PCB sites have severe depletion of TG-TH stores com-
pared to gulls at the reference site. However, in some
cases birds from high PCB sites were able to maintain
thyroid status comparable to that at the reference site.
Embryos and prefledglings differed in their HPT axis re-
sponse to PCB exposure. Adults, sampled at two high
PCB sites and a low PCB site in the GL and the reference
colony in 2001, did not differ in organismal thyroid status
across sites, but those from the high PCB sites had hyper-
trophied TGs and TG-TH depletion. The depletion of TG-
THs in PCB-exposed birds suggests that they have greatly
diminished capacity for thyroid responses in relation to
environmental change. Thus, environmental stressors may
have different effects on gull populations at high PCB sites,
compared with reference sites.
Studies of chicken embryos and chicks hatched from
eggs dosed with PCB 126, PCB 77, or Aroclor 1254 sug-
gest that embryonic exposure to these PCB treatments
has little or no effect on thyroid function. The PCB 126
and 77 doses used caused immune system effects, and
the highest dose caused considerable mortality (evalua-
tions from K. Grasman's laboratory).
Studies of mechanisms of thyroid disruption focused
on PCB effects on the Phase II liver biotransformation en-
zyme uridine diphosphate glucuronosyltransferase (UDP-
GT) and on the displacement of THs from binding pro-
teins. No evidence of hepatic UDP-GT induction by PCB
126 or 77 in chicken embryos or chicks hatched from
PCB-dosed eggs was found. Because UDP-GT induction
resulted in enhanced T4 metabolism and excretion is a key
mechanism by which PCBs alter thyroid function in mam-
mals, these results suggest that PCBs do not have equiva-
lent effects on birds and mammals. Preliminary studies of
GL gull plasma suggest that PCBs may be displacing THs
from their binding proteins at high PCB sites.
The Office of Research and Development's National Center for Environmental Research
47
-------
Endocrine Disruptors Program Review Workshop
Effects of Exposure to Environmental Estrogens on Reproductive
Parameters in a Marine Fish, Tautogolabrus adspersus
Lesley J. Mills, Ruth E. Gutjahr-Gobett, Doranne Borsay Horowitz, and Gerald E. Zaroogian
Atlantic Ecology Division, National Health and Environmental Effects Research Laboratory,
U.S. Environmental Protection Agency, Narragansett, RI
Estradiol (E2), ethynylestradiol (EE2), and estrone (E4)
are steroidal estrogens that are released into the aquatic
environment in sewage treatment effluent. To determine
whether these estrogens could impact reproductive pa-
rameters in a model fish species, actively spawning male
and female cunner (Tautogolabrus adspersus) were ex-
posed in the laboratory by implanting E2 EE2, or E4 sub-
cutaneously in a slow-release matrix. A separate experi-
ment was conducted with each of the three estrogens.
Experiments consisted of four treatments: control (im-
plant matrix only) and three concentrations of estrogen
(0.05, 0.5, or 2.5 mg/kg). Four replicate tanks, each
with three females and one or two males, were used per
treatment. Reproductive success of fish before and af-
ter implantation was assessed through daily measure-
ments of egg production, number of fertile eggs, and
number of viable developing eggs. At the end of a 2-
week exposure period, blood was drawn from each fish
for plasma steroid hormone andvitellogenin analysis. Fish
then were dissected, gonads weighed, and select tissues
preserved for later histopathological analysis.
Egg production prior to the implantation procedure
was not significantly different among the three experi-
ments, averaging about 280 eggs/gram female/day. Egg
viability prior to implantation ranged from 17 to 20 per-
cent, while mean egg fertility ranged from 25 to 52 per-
cent. After implantation, only egg production in the 2.5
mg/kg EE2 treatment was significantly lower than in the
controls. Notable, but not statistically significant, de-
creases in egg production relative to controls were ob-
served in the 2.5 mg/kg E2 treatment, and in the 0.5 and
2.5 mg/kg E4 treatments. Neither mean percent viability
nor percent fertility was significantly different than con-
trols in any estrogen treatment. All estrogen treatments
induced production of the female protein vitellogenin (Vtg)
in males, but EE2 more so than the others. In males from
the highest concentration of EE2, average plasma con-
centration of Vtg (480 mg/mL) was approximately 40
times (12 mg/mL) that in the high E2 treatment and 9
times (53 mg/mL) that in the high E4 treatment. Overall,
these results indicate that short-term exposure of mature
cunner to estrogens induces Vtg production in males
and may decrease egg production in females.
To investigate whether the presence of male Vtg is a
reliable indicator of decreased reproductive success in
mature fish, data on egg production, egg viability, egg
fertility, sperm motility, and male Vtg concentrations from
the 2-week exposure experiments were combined with
results of earlier 8-week exposure experiments. All males,
including two with Vtg levels exceeding 300 mg/mL,
produced motile sperm. Neither percent fertile eggs nor
percent viable eggs produced by reproductively active
fish demonstrated a significant correlation with male Vtg
concentrations. Male gonadosomatic index and average
daily egg production by females showed significant, but
weak, negative linear correlation with male Vtg concen-
trations. Results suggest that male Vtg expression is not
a reliable indicator of male reproductive dysfunction in
adult cunner exposed to estrogens during their repro-
ductive season, at least in relation to their capacity to
produce motile sperm or to fertilize eggs. In some cases,
male Vtg expression may serve as an indicator of re-
duced reproductive function in females exposed to es-
trogens at the same time.
48
The Office of Research and Development's National Center for Environmental Research
-------
Mechanisms of Action
-------
Endocrine Disrupters Program Review Workshop
Polymerase Chain Reaction-Differential Display (PCR-DD) Identifies
a Subset of Stage-Dependent and Toxicant-Regulated Genes
During Spermatogenesis: The Shark Testis Model
Gloria V. Callard and Chunhua Wang
Boston University, Boston MA
Due to a cystic mode of spermatogenesis and a simple
linear arrangement of developing spermatocysts across
the diameter of the testis, the spiny dogfish shark (Squalus
acanthias) is an advantageous model for stage-by-stage
analysis of factors and mechanisms controlling spermato-
genesis. hi previous studies, this research group reported
that estrogen receptor (ER) and androgen receptor (AR)
binding activities are concentrated in the germinal zone
(GZ) and premeiotic (PrM) regions, where spermatocysts
(follicle-like germ cell/Sertoli cell units) are in the stem cell
through secondary spermatogonial stages of development.
Interestingly, cadmium (Cd), an established mammalian
spermatotoxicant, is taken up and retained preferentially
in GZ/PrM regions of shark testis, where it increases the
percentage of germinal clones undergoing apoptosis.
One mechanism by which Cd is reported to exert its
toxic effects is by altering nuclear receptor-mediated gene
transcription. To assess the utility of the shark testis model
for identifying stage-related and toxicant-sensitive genes,
the polymerase chain reaction-different display (PCR-DD)
method of mRNA fingerprinting was applied to tissues col-
lected from control and Cd-injected animals. Poly (A+) RNA
was prepared from staged tissues: GZ; PrM; meiotic (M;
spermatocytes to early spermatids); and postmeiotic (PoM;
elongating to mature spermatids). Five primer sets were
used to obtain a total of 49 stage-dependent and 39 Cd-
responsive bands. Three bands were subjected to further
analyses: sequencing,5' and 3'rapid amplification of cDNA
ends, reverse transcriptase-PCR, and Northern analysis.
One differentially displayed band was highest in GZ/PrM
stages, where it was upregulated fivefold by in vivo Cd
treatment. It was identified as an approximately 400 base
pair fragment of the control region of mitochondria! (mt)
DNA, implying that Cd selectively enhances transcriptional
activity on the H-strand early in spermatogenesis. Of the 12
proteins encoded in mtDNA, the cytochrome oxidase sub-
units have been implicated in caspase activation leading to
apoptosis in somatic cells, and also are stage-dependent and
androgen-responsive in rodent testis.
A second differentially displayed band (PoM > M >
PrM/GZ) decreased after Cd treatment and ultimately
was identified as the shark homolog of BRAP2 (Ace.
#AF421550). BRAP2 is an evolutionarily conserved protein
that regulates cytosol-to-nuclear shuttling of transcription
factors and other nuclear proteins by specific binding to
nuclear localization signals. A third PCR-DD band increased
progressively from GZ/PrM through subsequent develop-
mental stages and was downregulated by Cd. This band
was identified as a shark-specific SI00 subtype (Ace.
#AF421551), one of a large family of Ca++ binding pro-
teins with diverse functions. Although functional studies
are required, results obtained using Cd as an illustrative toxi-
cant provide a starting point for uncovering spermatotoxic
mechanisms and demonstrate the feasibility of PCR-DD as
applied to the shark testis model for identifying toxicant-
sensitive genes, processes, and stages of development.
The Office of Research and Development's National Center for Environmental Research
51
-------
Endocrine Disruptors Program Review Workshop
Mode and Mechanism of Action of the Chlorotriazine Herbicides
Ralph L. Cooper, Susan C. Laws, Parikshit C. Das, Jerome M. Goldman,
Michael G Narotsky, and Tammy E. Stoker
Endocrinology Branch, Reproductive Toxicology Division, National Health and Environmental Effects
Research Laboratory, Office of Research and Development, U.S. Environmental Protection Agency,
Research Triangle Park, NC
The chlorotriazines (atrazine, simazine, etc.) are her-
bicides used extensively in the United States. Degradation
by-products, as well as the parent compounds, have been
detected in surface and ground-water in areas of major
usage. Atrazine and simazine have been shown to cause
an earlier onset of mammary tumors in Sprague-Dawley
rats following long-term dietary exposure (400 ppm or
approximately 22.5 mg/kg/day). A fundamental question
concerning this effect was whether or not the tumors
were induced by a direct action on the mammary gland
tissue itself, or as a consequence of chlorotrazine-induced
hormonal changes that create an endocrine environment
that is conducive for mammary tumor growth. Because
the majority of studies rule out a genotoxic mechanism
for the development of mammary gland tumors, the pri-
mary focus of these studies was to identify the cascade of
endocrine events leading to mammary tumor development.
These researchers found that the primary mode of
action of these herbicides involves a disruption of the hy-
pothalamic control of anterior pituitary function. Specifi-
cally, atrazine inhibits the pulsatile release of gonadotro-
pin-releasing hormone (GnRH) and subsequent luteinizing
hormone (LH) release. A similar decrease in prolactin (prl)
secretion also was identified. Based on the understanding
of the neuroendocrine alterations, it was found that the
chlorotriazines and their primary metabolites disrupt ova-
rian cycles, cause full-litter resorptions, suppress suck-
ling-induced prl release in the lactating dam, and delay
puberty in juvenile male and female rats (no observed ad-
verse effect levels and lowest observed adverse effect
levels have been identified in all studies). The atrazine-
induced suppression of suckling-induced prl release in the
lactating dam (birth to day 4) caused a significant increase
in prostate inflammation (lateral lobes) in the adult male
offspring.
A number of in vivo and in vitro studies were con-
ducted to determine the mechanism(s) involved in the dis-
ruption of the hormonal control of these reproductive pro-
cesses. In vivo, atrazine increases dopamine (DA) and
decreases norepinephrine (ME) in the hypothalamus. These
effects are consistent with the inhibition of the GnRH
pulses and decreased LH and prl secretion. Using undif-
ferentiated pheochromocytoma (PC12) cells, which con-
stitutively synthesize DA and NE, this research found that
the chlorotriazines (and metabolites) alter catecholamine
metabolism, suggesting that these hypothalamic neurons
represent one set of target cells. This project also is evalu-
ating the potential involvement of aromatase, because
changes in this enzyme may represent another cellular
target for the chlorotriazines.
These studies provided a better understanding of the
development of mammary gland tumors in rats and their
relevance to humans. In brief, the ovarian condition in-
duced in the female rat (persistent estradiol secretion) does
not have a likely parallel in women, thus the significance
of these tumors to humans is questionable. However, the
changes observed in central nervous system and pituitary
function raise new concerns about the potential health
effects of these chemicals. Finally, the fact that the me-
tabolites tested were similarly potent in inducing alterations
in reproductive function underscores the need to take these
metabolites into consideration in the overall assessment of
the chlorotriazine herbicides.
52
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disruptors Program Review Workshop
Gestational and Lactational Exposure of Male Mice to
Diethylstilbestrol Causes Long-Term Effects
on the Testis, Sperm Fertilizing Ability In Vitro,
and Testicular Gene Expression
Mark R Fielden1, Robert G Halgren1, Cora J. Fong1, Christophe Staub2, Larry Johnson2,
Karen Chou3, and Tim R Zacharewski1
department of Biochemistry and Molecular Biology, National Food Safety and Toxicology Center, Institute for
Environmental Toxicology, Michigan State University, East Lansing, MI; Department of Veterinary Anatomy
and Public Health, Texas A&M University, College Station, TX; Department of Animal Science, Institute for
Environmental Toxicology, Michigan State University, East Lansing, MI
The objective of this study was to determine the long-
term effects of gestational and lactational exposure to di-
ethylstilbestrol (DES) on testicular growth and histology,
number of Sertoli cells, epididymal sperm count and mo-
tility, sperm fertilizing ability in vitro, and testicular gene
expression using cDNA microarrays and real-time poly-
merase chain reaction (PCR) in B6D2F1 mice on postna-
tal day (PND) 21, 105, and 315. Pregnant females were
gavaged daily with 0,0.1,1, or 10 ug DES in corn oil per
kg of maternal body weight from gestational day 12 to
PND 21. Male neonates were monitored for body weight
and anogenital distance and weaned on PND 21. The tes-
tes from male offspring were examined on PND 21,105,
and 315 for changes in wet weight, histopathology, and
number of Sertoli cells.
Epididymal sperm count, sperm motion parameters,
and sperm fertilizing ability in vitro were measured on
PND 105 and 315. There were no significant changes in
testes weight, and histological examination of the testes
revealed no treatment-related effects. However, stereo-
logical analysis of the testes indicated a significant de-
crease in the number of Sertoli cells per testis in the
high-dose group, which persisted from PND 21 to PND
315 (p < 0.01). Sperm count also was decreased in the
high-dose group, but the decrease was only significant
on PND 315 (p < 0.05). The number and percent of
motile sperm and sperm velocity, linearity, and amplitude
of lateral head displacement were unaffected. By con-
trast, in vitro fertilizing ability of epididymal sperm was
significantly decreased in the high-dose group on both
PND 105 (p < 0.001) and PND 315 (p < 0.05). Microarray
analysis and confirming studies with real-time PCR iden-
tified early and latent alterations in the expression of genes
involved in estrogen signaling (ERoc), steroidogenesis (SF-
1, Cypl7, Cyplla, Star, SR-B1), lysosomal function
(LGP85, Psap), and regulation of testicular development
(Tr2-ll, Inhbc, HoxalO).
The results demonstrate that early exposure to DES
causes long-term adverse effects on testicular development
and sperm function, and that these effects are associated
with changes in testicular gene expression, even long after
the cessation of DES exposure. These results also suggest
multiple mechanisms by which early developmental expo-
sure to estrogen disrupts estrogen signaling, steroidogen-
esis, Sertoli cell function, and testicular development. Sub-
sequent studies also have revealed that gestational and lacta-
tional exposure to the phytoestrogen genistein does not in-
duce the same spectrum of physiological and molecular
effects in the testis as DES, thus suggesting that genistein
acts through a distinct mechanism of action in the testis.
The Office of Research and Development's National Center for Environmental Research
53
-------
Endocrine Disruptors Program Review Workshop
Development, Application, and Validation of a Sheepshead
Minnow Estrogen-Responsive cDNA Macroarray
Michael J. Hemmer and Leroy C. Folmar
Gulf Ecology Division, National Health and Environmental Effects Research Laboratory,
U.S. Environmental Protection Agency, Gulf Breeze, FL
This presentation provides an overview of research
conducted by the Gulf Ecology Division investigating the
effects of endocrine-disrupting chemicals on estuarine fish
species. A series of research studies were initiated to ex-
amine the comparative dose-response characteristics and
potencies of estrogenic chemicals using the sheepshead
minnow (Cyprinodon variegatus) as a small fish model.
These studies required the development of procedures for
measuring hepatic vitellogenin (VTG) mRNA synthesis
and serum VTG levels in male sheepshead minnows in
response to aqueous exposure to natural, pharmaceutical,
and xeno-estrogenic chemicals. The time-course of he-
patic VTG mRNA regulation and VTG plasma accumula-
tion and clearance kinetics also were determined to add a
temporal component to the field application and interpre-
tation of VTG as abiomarker of exposure, hi further studies,
differential display techniques were applied to samples
taken from dose-response studies to discriminate variably
expressed gene fragments between untreated control fish
and fish treated with 17B-estradiol. Information from these
studies was used to develop an estrogen-responsive cDNA
membrane macroarray. Laboratory validation of the
macroarray was accomplished by measuring the hepatic
expression of these genes in male sheepshead minnows
using fish previously exposed to 17B-estradiol, 17oc-ethy-
nyl estradiol, diethystilbestrol, methoxychlor, and p-
nonylphenol. Identical patterns of gene expression were
observed between native ligand, 17B-estradiol, and the four
estrogenic compounds tested, hi addition, intensities of
the gene responses as measured on the macroarrays clearly
followed a dose-response pattern for each chemical tested.
The research described is the first step toward developing
a suite of specific macroarrays for application to chemical
screening and prioritization programs, or as a monitoring
tool to identify chemical contamination of aquatic envi-
ronments.
54
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disruptors Program Review Workshop
Interaction of Estrogen and TCDD in an Avian Model
Bill L. Lasley and Rebecca Stanton
Department of Population Health and Reproduction and Center for Health and the Environment,
University of California, Davis, CA
This research group hypothesized that interactions be-
tween specific xenobiotics and endogenous factors that
control cell function and differentiation can explain both
similar and disparate toxic responses observed in nonmam-
malian species compared to laboratory models. Results from
a series of studies support this hypothesis and show that
the response of tissues from chickens respond differently
to dioxin than do tissues from mammals. Furthermore, these
studies indicate that the differences are largely attributable
to the interaction between dioxin growth factors and hor-
mones, including sex steroids. Previous studies showed
that in rodents, males demonstrate a greater sensitivity to
dioxin-induced decreases in body weight than females, and
that this difference can be reduced by exogenous estrogen
treatment, hi contrast, immature female chickens were more
sensitive than males, and estrogen treatment of immature
male chickens replicated the increased sensitivity of females.
Taken together, these studies suggest that some toxicants
and estrogens act in concert at the level of the cell as modu-
lators for cell growth and differentiation, energy homeosta-
sis, intermediary metabolism, and lipid mobilization in both
mammalian and avian species. More importantly, the re-
sults from these studies help explain how the mechanisms
by which some toxicants interact with endogenous hor-
mones, modulate targeted physiologic processes, and elicit
adverse effects in a manner that is species-, gender-, and
life-stage specific.
Estrogen treatment of male birds resulted in qualita-
tively similar lipid profiles to those of mature laying hens
and estrogen-treated immature hens, thus providing a model
by which to study dioxin-estrogen effects on lipid me-
tabolism in the absence of the energetic needs of egg pro-
duction. Resulting data show that dioxin antagonized sev-
eral effects of exogenous estrogen in male chickens, and
estrogen enhanced 2,3,7,8-tetrachlorodibenzodioxin
(TCDD) toxicity in a tissue-specific manner. Birds treated
with estrogen alone had increased total triacylglyceride
concentrations with specific increases in the A9-desaturase
products 16:ln7, 18:ln7, 18:ln9, and 20:ln9, although
these increases did not occur for birds treated with TCDD
alone or in combination with estrogen. TCDD and estro-
gen plus TCDD treatments increased phospholipid con-
centrations of the diet-derived polyunsaturated fatty acids
18:2n6,18:3n6,20:3n6,18:3n3,and20:5n3, although only
the estrogen plus TCDD group had significantly increased
total phospholipids. TCDD and estrogen plus TCDD treat-
ments decreased total concentrations of A9-desaturase
products and saturated fatty acids, and estrogen treat-
ment alone more specifically decreased concentrations of
several saturated and polyunsaturated fatty acids. These
findings support the hypotheses that differences exist in
the response of different species, age-stage, and genders
to the same toxicant.
As expected, the interaction of dioxin with estrogen
was common to both mammals and birds, but unexpect-
edly, the nature of the interaction in mammals and birds
was quite different. The protective effects of estrogen
observed in mammalian species was not observed in the
chicken model, and estrogen treatment actually augmented
the adverse effects of dioxin in some avian tissues. The
augmenting effect of estrogen on the metabolic effects of
dioxin in chickens is consistent with a more severe effect
of these compounds on reproduction in nonmurine verte-
brates and may explain the failure of laboratory experi-
ments utilizing rodent models to fully replicate adverse
effects observed in the field. More importantly, the spe-
cific adverse effects of dioxin on key lipid mobilization
may explain the wide range of developmental defects that
are observed in egg-laying species compared to the ef-
fects observed in eutherian species.
The Office of Research and Development's National Center for Environmental Research
55
-------
Endocrine Disruptors Program Review Workshop
Masculinization of Invertebrates by Endocrine Toxicants:
Mechanisms and Environmental Significance
Gerald A LeBlanc, Meredith P. Goading, Allen W. Olmstead, andXueyan Mu
North Carolina State University, Raleigh, NC
Field observations and laboratory studies have suggested
that members of many invertebrate phyla are masculinized
as a consequence of exposure to environmental chemicals.
The overall objectives of this research program are to eluci-
date endocrine processes in a mollusk (Ilyanassa obsoletd)
and in a crustacean (Dapknia magna) that control mascu-
linization and identify mechanisms by which environmental
chemicals can disrupt these processes.
The biocide tributyltin (TBT) has caused a pseudo-
hermaphroditic condition (imposex) in many marine snail
populations. The generally viewed mechanism by which
TBT causes imposex is by inhibiting the metabolic con-
version (via aromatase) of testosterone to 17B-estradiol.
Experiments were conducted to establish the relationships
among TBT exposure, elevated testosterone levels, and
imposex. Exposure of female mud snails (/. obsoletd) to
environmentally relevant concentrations of TBT signifi-
cantly elevated both testosterone levels and the incidence
of imposex. No evidence that elevated testosterone levels
were due to the inhibition of testosterone aromatization
was found. Rather, it was discovered that testosterone is
extensively converted to fatty-acid esters in this species,
which serves to maintain homeostasis of testosterone. TBT
suppressed the esterification of testosterone, resulting in
the elevation of free testosterone levels. Field studies dem-
onstrated that fatty esterification is the major determinant
of testosterone levels during the reproductive cycle of the
snail, and that TBT-induced imposex populations of
mudsnails have compromised testosterone-fatty acid es-
terification capabilities.
Having established testosterone homeostasis as a target
for the endocrine disruption in the snail, experiments were
conducted to identify possible masculinizing effects of tes-
tosterone in the crustacean (D. magna). Exposure of ma-
ternal daphnids to testosterone had no effect on the sex of
offspring. Rather, testosterone caused significant develop-
mental abnormalities in progeny. Detailed analyses revealed
that the developmental toxicity associated with testosterone
was primarily due to anti-ecdysteroidal activity of this com-
pound. The fungicide fenarimol also was found to elicit
significant anti-ecdysteroidal activity. Ongoing studies sug-
gest that 4-nonylphenol, propiconazole, and piperonyl bu-
toxide also elicit antiecdysteroidal activity.
An extensive investigation was performed in search of
a hormone that functions as a sex determinant in daphnids.
This group of investigators discovered that the terpenoid
hormone methyl farnesoate programs oocytes to develop
into male progeny. Further, the insecticides methoprene and
pyriproxifen were found to mimic methyl farnesoate and
alter sex ratios of daphnid progeny. Results from this pro-
gram have significantly advanced understanding of the en-
docrinology of masculinization in invertebrates and have
identified susceptible targets of endocrine toxicants that can
significantly alter sex ratios, sexual development, and fe-
cundity of these organisms.
56
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disruptors Program Review Workshop
Dioxin (TCDD) Disrupts Steroid Action in an Endometriosis Model
Kevin G. Osteen and Kaylon L. Bruner-Tran
Vanderbilt University School of Medicine, Nashville TN
Endometriosis is defined as the growth of endometrial
glandular epithelium and stroma at an extra-uterine or ec-
topic site. Ectopic implantation of endometrial tissue en-
tering the peritoneal cavity via retrograde menstruation
requires an invasive event and the biomolecules necessary
for establishment of endometriosis include the matrix
metalloproteinases (MMPs). The MMPs are expressed in
a cycle-dependent fashion, broadly expressed during men-
struation and focally expressed during estrogen-mediated
growth. These enzymes are largely absent during the
progesterone-dominated secretory phase of the menstrual
cycle, and their expression is inhibited by progesterone
treatments in vitro. Ectopic endometriotic lesions have a
decreased ability to respond to progesterone, impairing
the therapeutic benefit of progestins; nevertheless, preg-
nancy can reduce the risk or impact of the disease in
some women. A recent finding in this laboratory suggests
that women with endometriosis have a decreased sensi-
tivity to progesterone, which is associated with an increased
expression of MMPs during the menstrual cycle. This
research project explored whether environmental agents
might impact the cellular mechanism(s) controling endome-
trial MMP expression in women with endometriosis.
The spontaneous development of endometriosis in
primates has been associated with experimental exposure
to 2,3,7,8-tetrachlorodibenzo-/>-dioxin (TCCD), although
whether exposure to TCDD affects the establishment or
progression of this disease in women remains unclear. To
approach this question, culture models were examined in
concert with an experimental model of endometriosis us-
ing human tissue and nude mice. Specifically, this research
group examined if TCDD might impact endometriosis by
disrupting normal steroid-mediated endometrial MMP ex-
pression. Initially, it was demonstrated that the ability of
progesterone to suppress the expression and secretion of
MMPs is reduced following the in vitro exposure of hu-
man endometrium to TCDD. In vitro treatments of hu-
man endometrium with TCDD subsequently blocked the
ability of progesterone to prevent establishment of experi-
mental endometriosis, following injection of tissue into
nude mice. Further studies indicated that TCDD treat-
ment of endometrial tissue in vitro is associated with a
failure of progesterone to induce TGF-B2, a growth fac-
tor necessary for normal MMP regulation.
This project's most recent studies have demonstrated
that TCDD further impairs progesterone action in endome-
trial tissue by suppressing progesterone receptor expres-
sion. Using stromal-epithelial co-cultures, it was found
that TCDD exposure activates an epithelial-dominant path-
way that increases the expression of MMPs through pro-
inflammatory cytokine-mediated decreases in progester-
one receptor. Together, these findings suggest that TCDD
may influence the development of human endometriosis
by decreasing endometrial sensitivity to progesterone. By
several cell-specific mechanisms, TCDD acts to increase
endometrial MMP expression, leading to increased inva-
sive potential in an experimental endometriosis model.
Current efforts seek to explore whether the molecular regu-
lation of MMPs might be disrupted neonatally, prior to the
development of active disease in adults.
The Office of Research and Development's National Center for Environmental Research
57
-------
Endocrine Disruptors Program Review Workshop
Chemical Interference With Non-Genomic Steroid Actions:
A Novel Mechanism of Endocrine Disruption
Peter Thomas
University of Texas at Austin, PortAransas, TX
There is now convincing evidence that, in addition to
the classic genomic mechanism of steroid action via bind-
ing and activation of nuclear steroid receptors, steroids
also act at the cell surface of target tissues to initiate rapid,
nongenomic responses, and that these actions are medi-
ated by steroid membrane receptors. Steroid membrane
receptors and rapid steroid actions, including activation of
intracellular signaling pathways, have been identified in
many tissues, including cardiovascular tissues, brain, pi-
tuitary, bone, kidney, liver, gonads, and gametes. Several
recent studies have shown that nongenomic steroid ac-
tions, like genomic ones, are susceptible to interference
by xenoestrogens. The project goals are to determine the
mechanism, extent, and potential environmental hazards
of this novel type of endocrine disruption. Recent studies
with spotted seatrout and Atlantic croaker have shown
that the mechanism of interference with these nongenomic
steroid actions involves xenoestrogen binding to the ste-
roid membrane receptors thought to mediate these ac-
tions. A variety of xenoestrogens, including kepone and
o,p'-DDD, bind to the oocyte progestin membrane re-
ceptor (mPR) in spotted seatrout and also antagonize
progestin-induced oocyte maturation in an in vitro bioas-
say at concentrations of 10~6 to 10~7 M (equivalent to 20-
40 ppb, a tissue concentration frequently reported in fish
from contaminated environments). Moreover, these
xenoestrogen effects on receptor binding and oocyte matu-
ration in vitro are reversible (i.e., they were not nonspe-
cific toxic effects of the compounds), and both activities
were completely restored after washing. Xenoestrogens
also have been shown to interfere with the nongenomic
action of progestins to increase sperm motility in these
two fish species by binding to their mPRs on sperm.
The finding that the inhibitory action of a hydroxylated
polychlorinated biphenyl (PCB) on sperm motility was par-
tially reversed by addition of excess progestin is consistent
with a receptor-mediated mechanism of xenobiotic action.
Initial experiments show that the binding of several
xenoestrogens (methoxychlor, hydroxylated PCBs, and
dichlorodiphenyltrichloroethane derivatives) to the oocyte
mPR is dependent on localization of the receptor in the
plasma membrane and is related in part to their lipophilicity.
These preliminary results suggest that xenoestrogen inter-
actions with plasma membrane and nuclear steroid recep-
tors are qualitatively different, and that membrane receptor-
mediated steroid actions may be especially susceptible to
interference by lipophilic xenobiotic compounds. Similar
experiments will be conducted with the mPR on sperm and
amembrane estrogen receptor in testes to confirm the broad
applicability of these findings.
The current lack of information on the structures of
any steroid membrane receptors and their steroid binding
sites has prevented the development of experimental and
theoretical approaches to determine their potential interac-
tions with xenoestrogens at the molecular level. This re-
search group recently cloned, sequenced, and character-
ized the seatrout oocyte mPR, the first steroid membrane
receptor whose structure has been determined in any ver-
tebrate species. The seatrout mPR has seven transmem-
brane domains, which is characteristic of G-protein-
coupled receptors (GPCRs). Fourteen similar genes have
been identified and partially characterized in other verte-
brates, including three in humans. Preliminary results in-
dicate that the recombinant proteins produced in a bacte-
rial expression system transfected with the three human
genes also bind progestins and have characteristics typi-
cal of mPRs. In conclusion, a new family of steroid re-
ceptors has been discovered that is structurally unrelated
to nuclear steroid receptors, but instead has characteris-
tics typical of GPCRs. Moreover, evidence has been ob-
tained that xenoestrogens can interact with mPRs belong-
ing to this receptor family. The nature of xenobiotic inter-
actions with both the seatrout and human mPRs will be
investigated at the molecular level.
58
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disruptors Program Review Workshop
Environmental Androgens and Anti-Androgens
Vickie S. Wilson
Reproductive Toxicology Division, National Health and Environmental Effects Research Laboratory,
U.S. Environmental Protection Agency, Research Triangle Park, NC
Androgens play decisive roles in sexual differentia-
tion of the gonads and accessory reproductive tissues
during prenatal and neonatal development. In addition,
androgens influence the acquisition and maintenance of
secondary sex characteristics in adults. This research
focuses on the impact that endocrine-disrupting chemi-
cals (EDCs) can have on androgen action. This research
integrates in vitro, ex vivo, and in vivo studies to pro-
vide mechanistic and dose-response information for risk
assessment on EDCs. EDC action can seriously alter
reproductive development when administered during
critical, sensitive life stages. In utero, perinatal, and pu-
bertal exposures to EDCs can produce severe effects on
both male and female offspring, depending upon the
chemical and its mechanism of action.
Several classes of chemicals have been studied that act
as anti-androgens either by binding to the androgen recep-
tor and interfering with normal androgen action or by inhib-
iting the synthesis of testosterone (e.g., linuron, ketconazole,
vinclozolin, procymidone, dichlorodiphenyltrichloroethane
and its metabolites, and phthalate esters). Recent work also
includes the study of environmental samples and chemicals
with androgenic activity such as 17B-trenbolone (TB), an
anabolic steroid used as a growth promoter in beef cattle.
Studies with TB have been conducted in both in vitro and in
short-term in vivo screening assays. In the rat in utero
screening assay, maternal TB administration increased
anogenital distance and attenuated the display of nipples in
female offspring in a dose-related manner similar to the
published effects of testosterone propionate. Previous studies
have documented that these types of malformations in new-
born and infant rats are not only permanent effects, but also
are highly correlated with serious reproductive malforma-
tions in adults.
As tools in this research, several in vitro assays have
been developed and utilized both for screening chemicals
and as aids for defining mechanism of action. For ex-
ample, cell lines have recently been developed that stably
express either androgen- or estrogen-responsive luciferase-
reporter genes. This research primarily utilizes steroid
hormone receptors from mammalian species such as rat
or human androgen receptors. It is assumed, but largely
unproven, that EDCs will bind steroid receptors from
mammalian and nonmammalian species with similar af-
finity. To test this hypothesis, this group also is attempting
to identify, isolate, and sequence androgen and estrogen
receptor proteins from several nonmammalian species. If
time permits, there also will be a description of plans to
incorporate these receptors into in vitro assays and assess
receptor binding to EDCs across several species.
The Office of Research and Development's National Center for Environmental Research
59
-------
Poster Presentations
-------
Screening and Testing Assays
-------
Endocrine Disrupters Program Review Workshop
Computational Models for the Rapid Prediction of Ligand Binding
Affinities to the Androgen Receptor: Effects of Mutations on
Ligand-Receptor Specificity
NiAi1, Seong-Jae Yu1, Robert Kirk Delisle2, and William J. Welsh1
'University of Medicine and Dentistry of New Jersey, Robert Wood Johnson Medical School,
Piscataway, NJ; 2Accelrys, Inc., Princeton, NJ
Classically considered the male sex hormone, di-
hydrotestosterone is critically involved in numerous physi-
ological processes, hi addition to initiation of male sexual
differentiation and development of male secondary charac-
teristics, dihydrotestoterone has been implicated in blood
pressure regulation, obesity, and bone development in con-
junction with estrogen. For the past several years, there has
been growing concern regarding the influence ofboth natural
and synthetic chemicals in the enviroment on normal physi-
ological processes ofboth wildlife and humans as mediated
by the androgen receptor (AR). Although the full impact of
these chemicals on human health is still uncertain, it is evi-
dent that both basic and applied research into this issue is
essential, hi this present study, the binding affinities of a
series of compounds for both the rat and human AR were
estimated from calculated ligand-receptor binding energies.
A strong correlation (r2 = 0.76) was found between the
computed binding energies for this series of compound and
the corresponding published values of the observed relative
binding affinity for rat AR When these values of the bind-
ing energy were included as an addtional descriptor to build
three-dimensional quantitative structure-activity relationship
(3D-QSAR) models using comparative molecular field analy-
sis, the predictive ability of the models was improved dra-
matically for a series of external test-set compounds not
employedformodel building. These 3D-QSARmodels cur-
rently are undergoing further testing to determine their util-
ity in screening chemical libraries for endocrine-disrupting
compounds that preferentially bind the AR.
Similar binding energy calculations were repeated for a
particular AR variant containing the single-site Thr877Ala
mutation, which has been associated with failure of hor-
monal therapy in the treatment of human prostate cancer.
Comparision of ligand-binding energies revealed a general
trend of enhanced binding affinity for this AR variant over
the wild-type AR, consistent with published experimental
findings that this AR variant exhibits reduced ligand speci-
ficity and is inappropriately activated by progestins, estro-
gen, and even the anti-androgen hydroxyflutamide.
The Office of Research and Development's National Center for Environmental Research
65
-------
Endocrine Disruptors Program Review Workshop
A Computer-Docking Study of the Binding of Polycyclic Aromatic
Hydrocarbons and Their Metabolites to the Ligand-Binding
Domain of the Estrogen Receptor
Katrina W. Brown, Stephen Little, and James Rabinowitz
Molecular Toxicology Branch, Environmental Carcinogenesis Division, National Health and Environmental
Effects Research Laboratory, U.S. Environmental Protection Agency, Research Triangle Park, NC
Polycyclic aromatic hydrocarbons (PAHs) are a class
of ubiquitous, anthropogenic chemicals found in the en-
vironment. In the present study, computational methods
are used to evaluate their potential estrogenicity and the
contribution that chemicals in this class make to envi-
ronmental estrogenicity. Classical docking methods, dock
and affinity, are used to evaluate the potential binding
affinity of chemicals in this class and their metabolites to
the published crystal structures of the ligand-binding do-
main of the estrogen receptor. These methods, with a
molecular mechanics interaction energy scoring function,
were able to place estradiol within a root mean square
deviation of 0.4 Afrom its binding position determined by
x-ray crystallography. The scores obtained in this manner
show wide variation for the PAHs and their metabolites.
They depend on PAH type and the three-dimensional struc-
ture of the metabolites. For example (-)-antibenzo-
[c]phenanthrene diolepoxide is a much better binder in
this model than the (+) enantiomer. Afew dominant modes
of binding have been identified and will be presented.
These results will be compared to the results for known
binders. Semi-empirical quantum mechanical methods
also were used to compute the interaction energy of the
most stable structures obtained from the classical com-
puter-docking experiments. These quantum mechanical
calculations provide a quantitative description of the in-
teraction between the ligand and the receptor, and con-
tain elements that are omitted from the classical scoring
function. The comparison of these results demonstrates
the importance of the nonclassical interaction terms for
molecules that have pi electron systems. K.W. Brown
was funded by EPA/UNC Toxicology Research Program
Training Agreement CT902908 and CT827206 during
the performance of this study.
66
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disrupters Program Review Workshop
Rainbow Trout Androgen Receptor Alpha and Human Androgen
Receptor: Comparisons in the COS Whole-Cell Binding Assay
Mary C. Cardan, L. Earl Gray, Jr., and Vickie S. Wilson
Reproductive Toxicology Division, National Health and Environmental Effects Research Laboratory, Office of
Research and Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Typically, in vitro hazard assessments for the identi-
fication of endocrine-disrupting compounds (EDCs), in-
cluding those outlined in the Endocrine Disrupter Screen-
ing and Testing Advisory Committee Tier 1 Screening
protocols, utilize mammalian receptors. However, evi-
dence exists that fish sex steroid hormone receptors dif-
fer from mammalian receptors both structurally and in
their binding affinities for some steroids and environ-
mental chemicals. Most of the binding information avail-
able to date has been conducted using cytosolic prepara-
tions from various tissues. This research project sought
to compare competitive binding using rainbow trout an-
drogen receptor alpha (rtAR) and human androgen re-
ceptor (hAR) expressed in transfected COS (African
green monkey kidney) cells. In this system, the binding
affinities of individual receptors can be investigated with-
out the potentially confounding effects of other steroid
receptors present in cytosolic tissue extracts. Saturation
ligand binding and Scatchard analysis using [3H]R1881,
a synthetic androgen, revealed a dissociation constant
(By of 0.24 nM for the rtAR. In the same system, a Y^
of 2.27 nM was found for the hAR. Binding studies in
competition with [3H]R1881 were conducted using ste-
roids and a selection of environmental chemicals shown
to bind mammalian AR.
All of the chemicals and steroids studied competed
for binding in both rtAR and hAR. The relative order of
binding affinities of natural and synthetic androgens for
the rtAR was methyltrienolone > trenbolone > 11-keto-
testosterone > dihydrotestosterone (DHT) > testosterone >
androstenedione. The rank order for the hAR was similar,
except that DHT and testosterone had higher affinity than
11-ketotestosterone. Also, it was found that androstenedi-
one bound with lower affinity than what has been re-
ported in the literature by Wells and Van Der Kraak for the
rtAR. Other steroids and anti-androgens, such as proges-
terone, 17B-estradiol, hydroxyflutamide, vinclozolin and
its metabolites Ml and M2, and 2,2-bis-(4-chlorophenyl)-
1,1-dichloroethene also were studied, and their relative
binding order was similar for the two species. Studies
such as these will facilitate the identification of EDCs that
affect many species and support future risk assessment
protocols.
The Office of Research and Development's National Center for Environmental Research
67
-------
Endocrine Disruptors Program Review Workshop
Semiquantitative Laser-Scanning Confocal Microscopy (LSCM):
Assessing Crustacean Egg Quality for EDC Screening
G. Thomas Chandler and David C. Volz
University of South Carolina, Columbia, SC
Laser-scanning confocal microscopy (LSCM) is widely
used by biomedical investigators, but its application to envi-
ronmental toxicology, in particular marine ecotoxicology, is
practically nonexistent. Anew Semiquantitative LSCM ap-
proach is described here for assessing relative yolk quantity
in marine invertebrate embryos (harpacticoid copepods) after
rearing of their parents in polycylic aromatic hydrocarbons
(PAHs). These researchers found that LSCM represents a
powerful, easy, but largely unexplored ecotoxicological tool
for rapidly assessing in vivo effects of endocrine-disrupt-
ing chemicals (EDCs) on crustacean embryo quality and
development.
In this study, the common PAH chrysene (CHRY)
was selected as a model toxicant to investigate the utility
of the lipovitellin-based LSCM egg/embryo quality
screening tool. CHRY has a chemical structure that is
steroidal in nature, particularly when photo-oxidized by
ultraviolet (UV) light to 6-hydroxychrysene. This photo-
oxidation product exhibits anti-androgenic properties in
vitro in mammalian models, and may have endocrine-
active properties in marine invertebrates. In this study, it
was hypothesized that vitellogenesis may be affected in
female copepods (Amphiascus tenuiremis), and that these
effects could be detected in 1 day-old embryos via
fluorogenic labeling with the yolk-specific probe
BODIPY® 505/515 and direct LSCM photomultiplier-
based measurement.
The fluorescent yolk-labeling method described here
was able to stain and detect statistically significant differ-
ences in yolk concentrations in A. tenuiremis eggs from
females exposed to UV and/or CHRY-contaminated sedi-
ments. Control yolk intensities in less than 24-hour-old
embryos of females cultured throughout their lifecycle in
clean sediments were statistically identical with or with-
out UV exposure. In contrast, yolk intensities in less than
24-hour-old embryos of females cultured throughout their
lifecycle in CHRY-contaminated sediments were signifi-
cantly higher in the non-UV exposed 2,500 ng CHRY/g-
sed (67% higher) and UV-exposed 500 ng CHRY/g-sed
(76% higher) treatments. Females exposed to 500 ng
CHRY/g-sed without UV exhibited yolk intensities that were
significantly lower (18%) than UV-exposed females at the
same CHRY concentration, but significantly higher (45%
higher) than both UV and non-UV exposed controls. A
fivefold increase in CHRY concentration (2,500 ng CHRY/
g-sed) in the absence of UV also enhanced yolk deposi-
tion to eggs, but yolk levels were modestly (but signifi-
cantly) lower (6%) than eggs from the UV-exposed 500
ng CHRY/g-sed treatment. CHRY exposure during matu-
ration to female reproductive maturity significantly en-
hanced yolk deposition to eggs/embryos and was strongly
enhanced by UV irradiation. Although the direct mecha-
nism of CHRY-inducedyolk deposition is unknown, CHRY
may exhibit hormonal properties that mimic endogenous
crustacean hormones such as ecdysteroids.
68
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disrupters Program Review Workshop
Determination of Structural Requirements for Activation
of the Clearance Mechanism of Environmental Pollutants
and Xenobiotic Chemicals by the Pregnane Xenobiotic
Receptor: Species-to-Species Extrapolation
Vladyslav Kholodovych and William J. Welsh
University of Medicine and Dentistry of New Jersey, Robert Wood Johnson Medical School, Piscataway, NJ
Increases in the prevalence of certain cancers (e.g.,
breast, prostate, testicular, ovarian) may be related to
interactions between components of the endocrine sys-
tem and environmental molecules known widely as "en-
docrine-disrupting chemicals." However, there is un-
certainty as to the mechanism of the effects induced
by a particular compound and the dose at which the
effect is elicited as a direct result of exposure. There is
equal uncertainty as to what degree the data from wild-
life and laboratory animal model systems can be ex-
trapolated to measure the risk of human exposure to
the same xenobiotic chemicals. Recent evidence shows
that an "orphan" nuclear receptor known as the preg-
nane xenobiotic receptor (PXR) plays a key role in regu-
lation of gene expression of cytochrome P450-3A
(CYP3A) proteins that metabolize a wide variety of
chemicals including toxins, environmental contaminants,
and endogenous compounds such as toxic bile acids
and steroids. Elucidation of the structural features of
PXR required for binding of diverse classes of harmful
endogenous and exogenous compounds is crucial to
understand the clearance pathways and predict (and
thereby avoid) dangerous interactions with the endo-
crine system. The objective of the present study is to
employ computer-aided molecular modeling strategies
as a means of offering guidance into the molecular ba-
sis for commonalities and differences in how humans
and model animals respond to chemical exposure as
mediated by PXR.
Computer-based approaches are proving useful in
risk assessment to help prioritize existing chemicals in
terms of their endocrine-disrupting effects prior to la-
bor-intensive and time-consuming in vitro and in vivo
biological testing. Such approaches also are valuable
for predicting the potential endocrine-disrupting activ-
ity of new chemicals before they enter the environ-
ment. Comparative molecular field analysis (CoMFA)
and related approaches have been used to develop three-
dimensional quantitative structure-activity relationship
(3D-QSAR) models for several compound data sets:
diindolylmethanes, clotrimazole analogues, polychlori-
nated biphenyls, and a combined set of receptor ago-
nists that includes soil contaminants and toxins. Com-
putational homology modeling techniques were em-
ployed to construct a hypothetical 3D structure of
mouse PXR (mPXR from its human orthologue,
hPXR). Examples will be presented that demonstrate
the utility of the hPXR crystal structure and the mPXR
homology model in screening small-molecule ligands
for prediction of their potential harmful effects on the
endocrine system, either rapidly using "docking and
scoring" routines or more rigorously using calculated
ligand-receptor binding energies. QSAR models also
have been developed for these data sets employing two
novel computational tools, the polynomial neural net-
work (PNN) and the volume learning algorithm (VLA).
The PNN and VLA exhibit superiority over traditional
QSAR techniques in terms of their predictive ability.
The Office of Research and Development's National Center for Environmental Research
69
-------
Endocrine Disruptors Program Review Workshop
Utility of In Vitro Assays To Screen in Environmental
Mixtures for EDC Activities
Christy R Lambright1, L. Earl Gray, Jr.1, Gerald Ankley3, Lou J. Guillette2,
Edward Orlando2, and Vickie S. Wilson1
National Health and Environmental Effects Research Laboratory, Office of Research and Development,
U.S. Environmental Protection Agency, Research Triangle Park, NC; department
of Zoology, University of Florida, Gainesville FL; 3Mid-Continent Ecology Division,
U.S. Environmental Protection Agency, Duluth, MN
Female mosquitofish (Gambusia qffinis holbrooki)
downstream from Kraft paper mills in Florida display
masculinization of the anal fin, an androgen-dependent
trait. The first series of studies were designed to deter-
mine if water contaminated with pulp mill effluent (PME)
from the Fenholloway River, FL, displayed androgenic
activity in vitro and to relate this activity to the reproduc-
tive status of female mosquitofish taken from this river.
Eighty percent of the female mosquitofish from the
Fenholloway River were partially masculinized, while
another 10 percent were completely masculinized based
on the number of segments in the longest anal fin ray
(18.0 ± 0.4 versus 28.1 ± 0.9 [p < 0.001]) in a control
river versus Fenholloway River, respectively. In a COS
whole-cell binding assay, all three PME samples displayed
affinity for human androgen receptor (hAR) (p < 0.001).
In addition, PME induced androgen-dependent gene ex-
pression in CV-1 cells (cotransfected with pCMV hAR
and mouse mammary tumor virus [MMTV] luciferase
reporter), which was inhibited by approximately 50 per-
cent by coadministration of hydroxyflutamide, an AR
antagonist. When CV-1 cells were transfected with hu-
man glucocorticoid receptor rather than hAR, PME failed
to significantly induce MMTV-luciferase expression. Fur-
ther evidence of the androgenicity was observed using a
COS cell AR nuclear translocalization assay. PME bound
hAR and induced translocalization of AR into the nucleus.
PME also displayed "testosterone-like" immunoreactiv-
ity in a testosterone radioimmunoassay, whereas water
from the reference sites did not. In summary, water col-
lected downstream of the Kraft mill on the Fenholloway
River contains unidentified androgenic substances whose
presence is associated with masculinization of female
mosquitofish. Currently, PME samples are being frac-
tionated in attempt to identify the androgenic chemicals.
hi a second project, androgenic activity was detected
in feedlot effluent from a feedlot in the Midwestern United
States. However, it has not yet been determined if this
activity arises from natural or synthetic hormones. The
potency of beta trenbolone also was examined for
androgenicity. Trenbolone is an anabolic steroid used to
promote growth in beef cattle found in feedlot waste
water and manure samples. Based on observations of
reproductive alterations in fish in waters receiving feed-
lot effluent, concern has arisen about the presence and
persistence of this hormonally active substance in efflu-
ent-reaching streams near the feedlots. In vitro, beta
trenbolone was a full agonist, approximately as active as
dihydrotestosterone, and this activity was fully inhibited
by the antiandrogen hydroxyflutamide. When examined
in vivo in the Hershberger Assay, trenbolone displayed
selective androgenic receptor-mediated (SARM) activ-
ity, affecting some androgen-dependent tissues much
more than others in a manner suggesting that 5-alpha-
reduction inactivated rather than activated the parent com-
pound. In summary, trenbolone is a potent SARM. Fur-
ther studies are in progress to: (1) determine whether
trenbolone is present in feedlot effluent in concentra-
tions sufficient to induce effects, and (2) characterize
the ability of this chemical to alter vertebrate reproduc-
tion and development.
70
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disrupters Program Review Workshop
Endocrine Disruptors From Combustion and Vehicular Emissions:
Identification and Source Nomination
Clyde Owens', Brian Guttett1, Jeff Ryan1, Paul Lemieux1, Carolyn Acheson2, Michael DeVtto3,
Christy Lambright3, Vickie Wilson3, L. Earl Gray, Jr.3, James Rabinowitz3, Sukh Sidhu4,
Richard Striebich4, and Joy Klosterman4
National Risk Management Research Laboratory, U.S. Environmental Protection Agency (EPA),
Research Triangle Park, NC; 2National Risk Management Research Laboratory, EPA, Cincinnati, OH;
3 Toxicology Division, National Health and Environmental Effects Research Laboratory, EPA,
Research Triangle Park, NC;4University of Dayton Research Institute, Dayton, OH
During the last decade, concerns have been raised
regarding the possible harmful effects of exposure to
certain chemicals that are capable of modulating or dis-
rupting the function of the endocrine system. These
chemicals, which are referred to as endocrine-disrupt-
ing chemicals (EDCs), have the capability to interfere with
the production, release, transport, metabolism, or elimina-
tion of the natural hormones in the body responsible for
the regulation of developmental processes. Recently, ex-
haust samples from combustion and vehicular sources
are being analyzed to provide initial identification of EDCs.
The intent of this screening effort is to provide discerning
evidence for nominating sources for further EDC charac-
terization. Conventional sampling, advanced analytical
methods, and bioassays are being used to provide initial
characterization of these samples for their compound iden-
tity and EDC activity. The intent of this research is to
sample and chemically characterize multiple combustion
sources to determine whether EDCs are emitted from
combustion sources and in what quantity.
The Office of Research and Development's National Center for Environmental Research
71
-------
Endocrine Disruptors Program Review Workshop
Endocrine-Mediated Effects of UV-A Irradiation on Grass Shrimp
(Palaemonetes pugio) Reproduction: Implications
for EDC Screening
David C. Volz1, Edward F. Wirth2, Michael H. Fulton2, Geoffrey I. Scott2,
David S. Block1, and G Thomas Chandler1
University of South Carolina, Columbia, SC; 2National Oceanic and Atmospheric
Administration, Charleston, SC
Little is known concerning the interaction between ul-
traviolet (UV) light and endocrine-disrupting chemical
(EDC) exposure on crustacean reproductive cycles. Per-
turbations to the relationships between reproduction and
UV light in laboratory environments may induce responses
that are attributed incorrectly to toxicant exposure, hi this
study, male and female grass shrimp, Palaemonetes pugio,
were exposed to sublethal concentrations of endosulfan
(ES; 200 ng/L and 400 ng/L) under both white fluores-
cent (WF) and UV-A(315-400 nm) light conditions for 50
days in laboratory bioassays.
Female endocrine (vitellogenin, ecdysteroids, and cho-
lesterol), reproductive (percent gravid, clutch size), and em-
bryo (days to hatch, hatching success, and hatching sur-
vival) responses were assessed. UV exposure alone caused
a significant (more than fourfold) increase in total P. pugio
female egg production over the course of 50 days. Expo-
sure to ES and UV light significantly lowered the percent-
age of gravid females relative to UV controls, whereas ES-
exposed shrimp under WF lighting did not exhibit these
trends. Although higher vitellogenin concentrations and
lower ecdysteroid titers were correlated with increased
female egg production, cholesterol titers only exhibited a
dose-dependent change when exposed to ES. Embryos
from females exposed to UV light had significantly lower
ecdysteroid titers and shorter hatching times, but there were
no differences in embryo vitellogenin concentrations, hatch-
ing success, or hatching survival. These results indicate
that UV-A exposure has apronounced effect on grass shrimp
(P. pugio) reproduction and likely is mediated through 5-
hydroxytryptamine-related neuroendocrine pathways.
This project showed thatUV-Aradiation alone can sig-
nificantly alter chronic endpoints (e.g., reproduction, egg
quality, etc.) typically measured in crustaceans when evalu-
ating xenobiotic effects, hi this study, ES simply countered
UV-mediated stimulatory reproductive effects and either:
(1) induced shunting of female energy away from repro-
duction to xenobiotic metabolism, or (2) interacted with UV-
triggered neurotransmitter signaling. Regardless, subtle ES
effects were only detected under UV-Alight. Because UV-A
has a profound effect on grass shrimp reproductive
biomarkers and responses, and because UV is an environ-
mentally relevant parameter, pesticides with (e.g., fipronil)
and without (e.g., ES) UV degradation potential should be
evaluated under UV exposure. Furthermore, the definition
and scope of endocrine disrupter research should be ex-
panded to incorporate the complex interplay between envi-
ronmental factors and multiple endocrine feedback processes.
72
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disrupters Program Review Workshop
Enzyme-Linked Immunosorbant Assays for Detection of
Crustacean Vitellin and Ecdysteroids: Development
and Validation for EDC Screening
David C. Volz, David S. Block, Adriana C. Bejarano, and G Thomas Chandler
University of South Carolina, Columbia, SC
Little is known regarding the effects of endocrine-
disrupting chemicals (EDCs) on crustacean vitellogenesis
and hormone synthesis. Recently, this laboratory devel-
oped simple, fluorescence-based enzyme-linked immuno-
sorbant assays (ELISAs) for crustacean lipovitellin and
ecdysteroid quantification in 96-well microplates. Although
vitellogenin ELISAs are widely available for assessing ex-
posure to environmental estrogens in fish, quantitative as-
says for crustacean vitellin/vitellogenin are lacking. Addi-
tionally, nonradiometric approaches for crustacean
ecdysteroid quantification have previously been unavail-
able. These ELISAs involve: (1) competing reactions be-
tween antigens and vitellin- or ecdysteroid-specific anti-
bodies, and (2) detection via enzyme-labeled conjugates
and fluorescent substrates. Both assays take only 3 days
from start to finish (approximately 100 samples) and ex-
hibit a high degree of accuracy and reliability.
For the indirect, competitive vitellin ELISA, a vitellin
standard was purified from grass shrimp (Palaemonetes
pugio) embryos (Stage I-II) by immunoaffmity chroma-
tography and lyophilized into 1 ug aliquots. This ELISA
involves competing reactions between coated or free
lipovitellin and crossreactive polyclonal anti-amphipod
(Leptocheirnsplumulosus} vitellin antibodies. After design-
ing four different ELISAs, it was found that maximum sen-
sitivity (> 150 pg vitellin) and reliability (R2 > 0.99) was
approached by simply increasing the primary and second-
ary antibody incubation times. This vitellin ELISA yields a
dynamic standard curve (2-1,000 ng/mL) and low intra-
and interassay variability (less than 10%). Because vitellin
was detected in several crustaceans (e.g., grass shrimp,
amphipods, and copepods), this ELISA may be potentially
useful for other marine crustaceans. The effects of the
gamma-aminobutyric acid (GABA) disrupting insecticide
fipronil on male and female copepod (Amphiascus tenuiremis)
vitellin production were evaluated using this ELISA.
The indirect, competitive ecdysteroid ELISA involves
competing reactions between free ecdysteroids, conju-
gates of 20-hydroxyecdysone (20HE) and horseradish per-
oxidase, and 20HE-specific polyclonal antibodies. This
ecdysteroid ELISA yields a dynamic standard curve (1-
200 femtomoles/well) and can be used for all crustacean
species. This ELISA is not just specific for 20HE, as it
also detects intermediate ecdysteroids such as ecdysone,
3-dehydroecdysone, and 25-deoxyecdysone. For assay
validation, ecdysteroid concentrations were measured in
various life-cycle and developmental stages of the marine
benthic copepod A. tenuiremis and the amphipod L.
plumulosus. Additionally, the effects of the GABA-
disrupting insecticide fipronil on reproductively success-
ful/unsuccessful male and female copepod (A. tenuiremis)
ecdysteroid titers were evaluated using this assay.
The Office of Research and Development's National Center for Environmental Research
73
-------
Endocrine Disruptors Program Review Workshop
Reproductive Alterations in the Grass Shrimp (Palaemonetes
pugio) Following Pesticide Exposure
Edward F. Wirth1, David C. Volz2, Michael H. Fulton1, Geoffrey I. Scott1,
and G. Thomas Chandler2
1 Center for Coastal Environmental Health and Biomolecular Research, National Ocean Service, National
Oceanic and Atmospheric Administration, Charleston, SC; 2University of South Carolina, Columbia, SC
This study examined the effects of sublethal chronic
pesticide exposure on the reproductive output of the grass
shrimp, Palaemonetes pugio. Grass shrimp are an impor-
tant ecological species in estuarine systems and are ex-
posed to anthropogenic contamination through nonpoint
source runoff. Grass shrimp larvae were reared in the
laboratory until sexually mature and exposed to individual
pesticides for between 30 and 40 days. The four pesticides
evaluated were endosulfan (organochlorine insecticide),
chlorpyrifos (organophosphorus insecticide), methoprene
(insect growth regulator), and fipronil (phenylpyrazole in-
secticide). The concentrations for each exposure were as
follows: 200 and 400 ng endosulfan/L, 100 and 200 ng
chlorpyrifos/L, 1 mg methoprene/L, and 100 and 200 ng
fipronil/L. Shrimp were removed from the exposure cham-
bers when visible clutches appeared on the abdomen and
frozen for further analysis. Endpoints analyzed included
the rate at which females became ovigerous in each popu-
lation and the potential reproductive output based on av-
erage clutch sizes. Results indicated that endosulfan sig-
nificantly reduced the rate at which gravid females oc-
curred during exposure in a dose-dependant manner.
Chlorpyrifos results suggested that there may be an ef-
fect on reproduction, although the data did not exhibit a
dose-dependant trend. There was virtually no difference
in the rate of ovigerous females occurring when shrimp
were exposed to either fipronil or methoprene. Work is
continuing to evaluate a variety of physiological endpoints,
including quantification of vitellin, ecdysone, cholesterol,
and total lipids.
74
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disrupters Program Review Workshop
Influence of the Structural Diversity of Data Sets on the Statistical
Quality of Three-Dimensional Quantitative Structure-Activity
Relationship (3D-QSAR) Models: Predicting the Estrogenic
Activity of Xenoestrogens
Seong Jae Yu1, Susan M. Keenan1, Weida Tong2, and William J. Welsh1
'University of Medicine and Dentistry of New Jersey, Robert Wood Johnson Medical School, Piscataway, NJ;
2National Center for Toxicological Research, U.S. Food and Drug Administration, Jefferson, AR
Federal legislation has resulted in the two-tiered in vitro
and in vivo screening of some 80,000 structurally diverse
chemicals for possible endocrine-disrupting effects. To
maximize efficiency and minimize expense, prioritization
of these chemicals with respect to their estrogenic dis-
rupting potential prior to this time-consuming and labor-
intensive screening process is essential. Computer-based
quantitative structure-activity relationship (QSAR) mod-
els, such as those obtained using comparative molecular
field analysis (CoMFA), have been demonstrated as use-
ful for risk assessment in this context, hi general, how-
ever, the CoMFA models developed to predict estrogenic-
ity have been constructed from data sets with limited struc-
tural diversity. In this study, CoMFA models were built
based on biological data for a structurally diverse set of
compounds spanning eight chemical families. This research
group also compared two standard alignment schemes
employed in CoMFA, namely atom-fit and flexible field-
fit, with respect to the predictive capabilities of their re-
spective models for structurally diverse data sets. The
present analysis indicates that flexible field-fit alignment
fares better than atom-fit alignment as the structural di-
versity of the data set increases. Values of log RP, where
RP is equal to relative potency, predicted by the final flex-
ible field-fit CoMFA models, are in good agreement with
the corresponding experimental values for this data set.
By virtue of the structural diversity of the data set chosen
to build them, these three-dimensional QSARmodels should
be effective for predicting the endocrine-disrupting po-
tential of chemicals that span arange of chemical families.
The Office of Research and Development's National Center for Environmental Research
75
-------
Developmental Exposures
-------
Endocrine Disrupters Program Review Workshop
Neurophysiological Consequences in Hippocampus
as a Function of Developmental Hypothyroidism
Mary Gilbert1 and Li Sui2
'U.S. Environmental Protection Agency,2National Research Council, Research Triangle Park, NC
Thyroid hormones are essential for maturation and
function of the mammalian central nervous system. Se-
vere congenital hypothyroidism results in irreversible struc-
tural damage and mental retardation in children. Although
a variety of environmental contaminants have been dem-
onstrated to alter circulating levels of thyroid hormones
(e.g., polychlorinated biphenyls, brominated flame retar-
dants, drinking water contaminants, pesticides), the
neurotoxicological effects associated with such hormone
reductions have not been adequately assessed. Thyrotoxins
such as propylthiouracil (PTU) have been used perva-
sively in studies designed to determine the role of thyroid
hormone in brain development. Although it is well estab-
lished that the hippocampus is a brain region impacted by
hypothyroidism, functional assessment of the neurophysi-
ological integrity of the hippocampus following such treat-
ment is lacking. Moreover, little information is available on
more modest perturbations in thyroid hormones that would
mimic those induced by environmental agents.
This presentation will focus on recent data characteriz-
ing the physiological changes associated with developmen-
tal hypothyroidism induced by PTU. Synaptic transmission
and plasticity in the dentate gyrus of the hippocampus of
adult offspring of pregnant dams treated with A1254 and
PTU are clearly altered, but the pattern of change is distinct
between these two types of treatment. The age and dura-
tion of exposure, the age and site of assessment, and the
relative change in the two main thyroid hormones T3 and
T4 all may contribute to differential patterns of effects ob-
served. Differential effects also are evident between the
CA1 and dentate subregions of the hippocampus. How-
ever, certain similarities also exist, a predominant feature
being the irreversible nature of the developmental insult in a
brain region critical for cognitive function. Exposure limited
to the perinatal period produced alterations in synaptic trans-
mission and plasticity in adulthood, despite elimination of
the contaminant and a return of thyroid hormones to nor-
mal range.
Future work will extend these observations by charac-
terizing more fully the dose-response relationships, examin-
ing behavioral correlates of altered physiology, and evaluat-
ing environmentally significant thyroid-disrupting chemicals
including brominated flame retardants and perchlorate. In-
creased understanding of the long-term consequences of
mild perturbation of thyroid hormones during brain devel-
opment will aid in assessment of the potential health haz-
ards posed by environmental contaminants that interfere
with thyroid hormone function.
The Office of Research and Development's National Center for Environmental Research
79
-------
Endocrine Disruptors Program Review Workshop
Retinoids Induce All the Limb Malformations Observed in Wild
Populations of Deformed Frogs
Aristocle Ndayibagira, Felix Grun, Bruce Blumberg, David M. Gardiner
University of California, Irvine, CA
There is considerable evidence that wildlife and do-
mestic animals have suffered adverse consequences from
exposure to environmental chemicals that interact with
components of the endocrine system. To assess the sig-
nificance of endocrine disrupters that activate retinoid-
signaling pathways for their role in causing limb devel-
opmental deformities in frogs, and to understand their
mechanism of action to assess their implications for hu-
man health, this research project focused on the effects
of disruption of retinoid-sensitive signaling pathways.
These studies demonstrated that all the malformations
observed in wild populations of frogs can be induced by
experimental exposure to retinoids.
The research focused on the effects of (E)-4-[2-
(5,5,8,8-tetramethyl-5,6,7,8-tetrahydro-2-naph-
thalenyl)-l-propenyl]benzoic acid (TTNPB), which
specifically activates the retinoic acid receptor, and led
to the discovery that there are multiple developmental
windows of sensitivity to retinoid exposure. The ex-
perimental phenotypes include normal development,
early embryonic lethality, duplicated limb buds, bony
triangles, and truncated limbs. These phenotypes could
be elicited in response to treatment with retinol palmi-
tate, atRA, and TTNPB. A notable result is that limb
dysplasias were only observed when larvae were treated
during specific stages of limb bud development (stages
50-52 in Xenopus laevis; see Table 1). Exposures at
earlier developmental stages induced malformations in
other organ systems (e.g., craniofacial, axial), but not
in limbs. Treatment at all stages is developmentally toxic
at high doses and long exposures, although an interest-
ing observation is that TTNPB treatment is not acutely
toxic when treatments are after the gastrula stage. The
surviving larvae typically die a few weeks after expo-
sure.
Table 1. Treatment ofXenophus laevis with TTNPB elicits stage-specific limb defects.
TTNPB
None
8x109
4x10*
8x10"
4x107
8x10'
Stage 50
% mortality
12.5
44.4
25
33.3
50
50
% malformed
0
0
0
0
0
0
Stage 51
% mortality
40
70
41.6
58.3
100
100
% malformed
0
0
50
80
0
0
Stage 52
% mortality
10
40
10
11.1
72.7
77
% malformed
0
33.3
66.6
100
100
100
80
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disrupters Program Review Workshop
Low Doses of Estrogenic Chemicals Like Diethylstilbestrol During
Development Result in Permanent Alterations
in the Reproductive Tract
Retha R. Newbold, Elizabeth Padilla-Banks, and Wendy N. Jefferson
Developmental Endocrinology Section, Laboratory of Molecular Toxicology,
National Institute of Environmental Health Sciences, Research Triangle Park, NC
Adverse human health consequences may result from
exposure to chemicals that interact with the endocrine
system, as documented in experimental animals and wild-
life. Because the developing organism is uniquely sensitive
to perturbation by chemicals with hormone-like activity,
the present study addresses whether exposure to these
chemicals during critical stages of differentiation will per-
manently alter the developmental program of tissues, so
that they respond atypically to further stimuli later in life.
Outbred CD-I mice were treated by subcutaneous injec-
tions with diethylstilbestrol (DES; 0001-1,000 ug/kg) dis-
solved in corn oil, or corn oil alone (control) on days 1-5
of neonatal life. Mice were weaned at 17 days prior to
puberty, housed 4 per cage, and challenged with three
daily doses of 17B-estradiol (500 ug/kg) or DES (10 ug/
kg). On the fourth day, uterine weight-body weight ratios
were determined. Uterine tissues were microscopically
evaluated for changes in epithelial cell height and number,
gland number, and induction of estrogen-responsive pro-
teins, including lactoferrin, progesterone receptor, and c-
fos. Neonatal DES exposure resulted in altered uterine
response to estrogen at puberty. Of particular interest was
that the response varied depending on the dose of neonatal
exposure; neonatal exposure to DES 0.01 caused an en-
hanced response to estrogen at puberty as compared to
controls, whereas higher neonatal doses of DES caused
reduced uterine response. To determine if these effects
were permanent, an additional group of mice was neona-
tally treated with DES (0.001-10 ug/kg) and housed until
4-5 months of age. These adult mice were ovariectomized
and challenged 7 days later as described for the immature
mice. The adult mice exhibited similar results (enhanced
uterine wet weight response at the 0.01 dose and a damp-
ened response at the high doses of 1 and 10 ug/kg). Mecha-
nisms responsible for this altered response involved ERoc-
mediated events. This research group concluded that al-
tered uterine responses were permanently imprinted by
developmental exposure to estrogens. Other environmen-
tal estrogens, including genistein, the naturally occurring
phytoestrogen found in soy products, are being tested
and compared to DES to determine if similar altered uter-
ine responses occur.
The Office of Research and Development's National Center for Environmental Research
81
-------
Field Studies
-------
Endocrine Disrupters Program Review Workshop
Endocrine-Disrupting Chemicals and Thyroid Outcomes
Henry A. Anderson1, Lawrence P. Hanrahan1, Vicky Persky2, Mary Turyk2, Dyan Steenport1,
Robert Chatterton2, Lynda Knobeloch1, Livia Navon1, and the Great Lakes Consortium
'Wisconsin Department of Health and Family Services, Madison, WI;2 University of Illinois, Chicago, IL
Each year, nearly 10 million Great Lake Basin resi-
dents consume Great Lake sport fish (GLSF). These
consumers represent the largest, non-occupational ex-
posure to persistent organic pollutants. Polybrominated
diphenyl ethers (PBDEs) recently have been detected in
Great Lakes fish. These compounds are structurally simi-
lar to polychlorinated biphenyls (PCBs), which have been
shown to have negative effects on thyroid function. This
study will examine the hypothesis that PDBEs may act
separately or synergistically with PCBs to impair thyroid
gland and reproductive hormone function. An existing,
well-characterized cohort of 4,206 GLSF consumers will
be the focus of this research.
Cohort members will be interviewed to assess cur-
rent and historic fish consumption, dietary and potential
occupational exposures to PBDEs and PCBs, and endo-
crine health status. Cohort members who have not been
diagnosed with an endocrine disorder will be invited to
donate blood and urine samples and complete a detailed
questionnaire ascertaining medical, GLSF consumption,
occupational, and dietary histories. Pollutants such as
l,l-dichloro-2,2-bis(p-chlorophenyl)ethylene (DDE),
PCBs, PBDEs, as well as thyroid and reproductive hor-
mones will be assayed. Hemoglobin AjC and total serum
lipids also will be measured. The association of endo-
crine measurements with total and congener-specific
pollutants and GSLF consumption will be examined. In-
teractive effects will be assessed.
A pilot study of this cohort showed inverse associa-
tions between PCB levels and T4 and free thyroxine in-
dex (FTI).1 Associations were stronger in women than
in men, with less consistent effects on T3 and thyroid-
stimulating hormone (TSH). In women, the association
of PCB with FTI—but not with T, level—and in men
4
the association with T4 remained significant after adjust-
ment for years eating GLSF. Among male subjects, in-
verse PCB associations were seen with testosterone and
sex hormone binding globulin-bound testosterone after
adjustment for fish consumption.
Steroid hormones were not measured in women. In
women, the number of GLSF meals eaten during the last
year was inversely associated with T4 and FTI, and posi-
tively related to TSH, although only the association with
FTI was significant after adjustment for PCB level. In
men, TSH was negatively associated with GLSF con-
sumption, but this association was not significant after
adjustment for PCB level. DDE was not associated with
thyroid hormone levels.
Prior studies, combined with data from this cohort,
suggest that fish consumption is associated with lower
levels of T4 and FTI. In these preliminary data, the in-
verse associations of PCBs with T4, FTI, and SHBG-
bound testosterone remained significant after adjustment
for fish consumption. Conversely, the association of
number of GLSF meals in the last year with FTI, but not
T4 in women, remained significant after control for PCB
level, suggesting that there may be factors in fish, other
than PCBs, that contribute to the thyroid effects.
PBDEs will be investigated in this study for the po-
tential additive or interactive effects on endocrine func-
tion with other environmental pollutants.
'Persky V, Turyk M, Anderson HA, Hanrahan LP, Falk C, Steenport DN, Chatterton R, Freels S, and the Great Lakes Consortium. The
effects of PCB exposure and fish consumption on endogenous hormones. Environ Health Perspect 2001; 109:1275-83.
The Office of Research and Development's National Center for Environmental Research
85
-------
Endocrine Disruptors Program Review Workshop
Latent Effects of Gestational Exposure to Heptachlor
Dean Baker, Vlrike Luderer, Sastry Gollapudi, and Haiou Yang
University of Calif ornia, Irvine, CA
This research will investigate whether gestational
exposure to the chlorinated cyclodiene insecticide hep-
tachlor permanently alters reproductive and immune func-
tion. The study is based on a well-characterized episode
in which the entire commercial milk supply on the Island
of Oahu in Hawaii was contaminated with heptachlor
epoxide during a 15-month period (1981-82), resulting
in gestational exposure to the offspring of women who
drank cows' milk during that period.
The research project will evaluate two primary hy-
potheses: (1) reproductive function in young adults who
were exposed to heptachlor during gestation will be defi-
cient compared to a control group, and (2) immune sys-
tem function will be altered in the heptachlor-exposed group
compared to the control group. Secondary hypotheses
are: (1) the biological indicators of reproductive and im-
mune function will correlate with quantitative estimates of
heptachlor exposure, and (2) there will be gender-specific
changes in reproductive endocrine and immune function
associated with the heptachlor exposure.
The study will assess biological indicators of repro-
ductive and immune function in 400 young adults who
were in utero on Oahu at the time of the milk contamina-
tion and have resided their whole lives on Oahu. hi addi-
tion, 200 unexposed comparison participants matched for
age and ethnicity will be studied, including 100 long-term
residents of Oahu who were not born on the island and
100 native residents of neighboring islands in Hawaii who
were not impacted by the milk contamination. The study
will involve a representative population of young adults,
born between July 1981 and June 1982, who participated
in a recently completed study of neurobehavioral effects
of this exposure, plus additional young adults recruited
from the neighboring islands.
The study will assess biological indicators of repro-
ductive and immune function. Hypothalamic-pituitary-
gonadal axis function will be assessed by measuring se-
rum reproductive hormone concentration: testosterone in
men, estradiol and progesterone in women, and luteiniz-
ing hormone and follicle-stimulating hormone in both men
and women. In the men, semen samples will be obtained
for determination of sperm quality. The women will be
asked to collect daily first morning urine specimens and
record menstrual histories for 6 weeks. Levels of luteiniz-
ing hormone, estrone-3-glucuronide, and pregnanediol 3-
alpha-glucuronide will be measured in the urine. Cell-me-
diated (Thl) immunity will be evaluated by antibody titer
response to immunization with tetanus and multivalent
pneumococcal vaccine. The proportion of Thl and Th2
type CD4+ cell subsets in peripheral blood will be as-
sessed using in vitro analysis of cytokine expression fol-
lowing activation. Susceptibility of peripheral blood T cells
to activation-induced cell death will be assessed using in
vitro analysis of Fas (CD95) and its ligand (CD95L) ex-
pression, and the percentage of apoptotic T cells assessed
with Annexin V staining, at basal level and following acti-
vation.
The analysis will compare reproductive and immune
function measures between the Oahu-born group and
the two comparison groups controlling for relevant con-
founders. Secondary comparisons among the Oahu-born
population will be made based on individual estimates of
gestational heptachlor epoxide exposure during the milk
contamination episode.
86
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disrupters Program Review Workshop
Endocrine-Disrupting Pesticides and Neurodevelopmental
Outcomes in Farmworker Children, Salinas Valley
Brenda Eskenazi1, Asa Bradman1, Laura Fenster2, Dana Barr3, Jonathan Chevrier1,
Martha Harnly2, Abbey Alkon1, and Marcy Warner1
University of California, Berkeley, CA;2 California Department of Health Services, Berkeley, CA;
3Centersfor Disease Control and Prevention, Atlanta, GA
The primary goal of this study is to determine whether
in utero exposure to endocrine-disrupting (ED) pesticides
is associated with adverse effects on the neurobehavioral
development of offspring of exposed women. The objec-
tives are to: (1) determine whether in utero exposure to
ED pesticides, such as organochlorine and nonpersistent
ED pesticides, is associated with adverse effects on the
neurobehavioral development of children; and (2) identify
population correlates of exposure. To achieve Objective
1, this research project will determine whether levels of
organochlorine pesticides in serum of women collected at
26 weeks gestational age or levels of nonpersistent ED
pesticides measured in maternal urine collected prenatally
(at 13 and 26 weeks gestational age, respectively) are as-
sociated with poorer performance on the Brazelton
neurodevelopmental assessment. At older ages (6,12, and
24 months postnatal), whether children's exposure to pes-
ticides during the prenatal period is associated with poorer
cognition, perception, attention, memory, motor coordi-
nation, and behavioral/emotional adjustment will be as-
sessed using age-appropriate tests standardized and ad-
ministered in Spanish. To achieve Objective 2, how levels
of ED exposure vary by gender and age as well as which
variables best predict exposure (e.g., occupational status,
season, nearby pesticide use, and length of time in the
United States) will be determined.
The study population consists of women and chil-
dren participating in the University of California, Berke-
ley, Center for Children's Environmental Health Research
in the Salinas Valley, CA, examining organophosphate
pesticide exposures to pregnant women and their chil-
dren as well as their association with adverse neuro-
developmental and respiratory effects in the children. It
includes 601 pregnant women who were less than 20
weeks gestation, Medi-Cal eligible, 18 years or older,
and received prenatal care at two community clinics. Of
these, 88 percent prefer to speak Spanish, 85 percent
were born in Mexico, and 53 percent have lived for 5
years or less in the United States. Women born in Mexico
have been found to have elevated levels of organochlo-
rine pesticides in breast milk compared to levels found in
U.S.-born participants. Possible sources of pesticide
exposure during pregnancy included farm fieldwork
(26%), other agricultural work (16%), other agricultural
workers in the household (75%), pesticides used in the
home (40%), and home within 200 feet of afield (15%).
This research demonstrated that the study population
has potentially heavy exposure to ED pesticides in the
Salinas Valley, and has probable previous exposure to
persistent ED pesticides in Mexico.
During Year 1, development was begun on pesticide
use indices based on state pesticide use data, a unique
resource in California. The association of these indices
with exposure biomarkers will be evaluated. For each
participant, the amount of nearby pesticide use is being
computed for the 1- and 2-month periods prior to urine
or serum sample collection. For the year 2000, approxi-
mately 600,000 pounds of potential ED pesticides were
applied in the Salinas Valley. The Centers for Disease
Control and Prevention (CDC) has developed assays for
as many of the nonpersistent pesticides as possible. Cur-
rently, approximately 65 percent of these can be detected
by urine, about 9 percent are detectable in serum only,
and about 26 percent cannot be detected in urine or se-
rum. All maternal urine specimens have been shipped to
the CDC laboratory. Chemical analyses of the maternal
urine samples began in Year 1 and will continue in Year 2.
The CDC laboratory also will initiate measurement of
biomarkers of exposure to 14 persistent organochlorine
pesticides and polychlorinated biphenyls in archived se-
rum samples collected at 26 weeks gestational age dur-
ing Year 2. An extensive literature review was completed
to identify mechanisms of ED for the ED pesticides.
Neurobehavioral assessments have been completed on
newborns and 6- and 12-month olds; and assessment of
24-month olds will continue until March 2003. Medical
record abstraction for pregnant women through deliv-
ery is complete. Forms for pediatric medical record ab-
straction have been developed, and abstraction has been
initiated for 24-month-old children.
The Office of Research and Development's National Center for Environmental Research
87
-------
Endocrine Disruptors Program Review Workshop
Persistent Organic Pollutants and Endometriosis Risk
Victoria L. Holt, Stephen M. Schwartz, Larry L. Needham, and Dana B. Barr
University of Washington and the Fred Hutchinson Cancer Research Center, Seattle, WA
Endometriosis, a disease affecting approximately 5 per-
cent of U.S. reproductive-aged women with chronic pelvic
pain, dysmenorrhea, and infertility, has been linked in epide-
miologic studies to exposures indicating high circulating
estrogen levels. There has been recent public and scientific
concern that endocrine-disrupting chemicals in the envi-
ronment may have estrogenic effects in the body and there-
fore increase endometriosis risk, but results of the few epi-
demiologic studies of this issue have been mixed.
hi this study, the relationship between endometriosis
and exposure to organochlorine compounds and polychlo-
rinated biphenyls is being examined in a large population-
based study. Whether these associations are modified by
polymorphisms in genes involved in estrogen metabolism
will be determined. The study is an ancillary investigation to
a current National Institute of Child Health and Human De-
velopment-funded case-control study of risk factors for
endometriosis that is being conducted within a large health
maintenance organization in western Washington State.
The parent study provides data from in-person inter-
views including reproductive, contraceptive, menstrual, and
behavioral characteristics; dietary questionnaires; anthro-
pometric measurements; pharmacy information; and re-
sults of laboratory analyses of serum for two polymorphic
genes (GSTM1 and COMT) involved in detoxification and
estrogen metabolism. The current study will utilize blood
samples from 300 cases and 600 controls from the parent
study to determine serum levels of total polychlorinated bi-
phenyls (PCBs), PCB congeners, hexachlorobenzene, B-
hexachlorocyclohexane (B-HCH), ?i-HCH, aldrin, hepachlor
epoxide, oxychlordane, trans-nonachlor, p,p'-dichlorodiphe-
nyl dichloroethylene (p,p'-DDE), o,p'-DDE, dieldrin,
endrin, o,p'-dichlorodiphenyl trichloroethane (o,p'-DDT),
p,p'-DDT, and mirex residues as well as two polymorphic
cytochrome p450 genes (1A1 and 1A2). Additionally, 450
urine samples (150 cases and 300 controls) will be tested
for levels of the methoxychlor metabolite 2,2-bis-(p-
hydroxyphenyl)-1,1,1 -trichloroethane (HPTE).
The primary objectives of the study are to determine:
(1) whether the risk of endometriosis is associated with
lipid-adjusted serum levels of the above-listed organochlo-
rine pesticides or urine levels of HPTE; (2) whether the risk
of endometriosis is associated with lipid-adjusted serum lev-
els of PCBs (total PCBs and 35 PCB congeners); and
(3) whether the risk of endometriosis resulting from orga-
nochlorine pesticide or PCB exposure differs among women
with differing CYP1A1, CYP1A2, COMT, and GSTM1
genotypes. The secondary objective is to determine whether
the risk of endometriosis resulting from organochlorine pes-
ticide or PCB exposure differs among women with differ-
ing levels of other exposures affecting estrogen levels. Analy-
ses comparing cases and controls with respect to levels of
these organic pollutants, other hormonal risk factors, and
their interactions with genetic polymorphisms will be con-
ducted to address the specific objectives.
88
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disrupters Program Review Workshop
Biogeographic Variation in the Model Organism Palaemonetes
pugio: Implications for Toxicological Studies
Mark A. Roberts', Joseph M. Quattro1, and G Thomas Chandler2
'Department of Biological Sciences, University of South Carolina, Columbia, SC;
2School of the Environment, University of South Carolina, Columbia, SC
lexicologists often have used "model" marine or-
ganisms such as grass shrimp (Palaemonetes pugio) to
investigate the effects of potential toxins within estua-
rine environments. The species is widely distributed along
the Atlantic Coast and readily available, making it a com-
monly used "sentinel" species target for toxicological
studies. Though well studied in the laboratory, little is
known about the "phylogeography" of the species, or
how geography and species attributes interact to mold
patterns of intraspecific genetic variation. If species are
highly structured geographically, for example in P. pugio,
then conclusions—and potentially management deci-
sions—drawn from studies in one geographic locale
might not be applicable to other areas. That is, it might
be best if toxicological studies are constrained to com-
parisons involving areas with similar phylogeographic
histories. For example, median lethal concentration (LC50)
studies for the pesticide endosulfan from either side of
the Florida Peninsula have produced toxicological data
that are incongruent: susceptibility of shrimp from the
Gulf of Mexico is reported to be one-half that of shrimp
caught along coastal South Carolina (LC50 = 1.31 ug/uL
and 0.66 ug/uL, respectively). However, without knowl-
edge of the background genetic variation between shrimp
populations, a complete understanding of these data are
difficult. The phylogeographic history of P. pugio
throughout most of its range has been determined by
examining mitochondria! DNA sequence variation in 161
individuals representing 21 populations. Phylogenetic
analyses showed a strong phylogenetic split between
populations within the Gulf of Mexico and those along
the Atlantic Coast. Interestingly, there was almost no
divergence within the Atlantic Coast populations. In con-
trast, the Gulf of Mexico populations exhibited fixed ge-
netic differences and strong population-level differentia-
tion, suggesting that differences in life histories between
populations in either basin are highly divergent or that
they have widely divergent biogeographic histories. These
shrimp currently are being analyzed at the nuclear DNA
locus vitellogenin to test for congruence with the mito-
chondrial data that would suggest a longstanding diver-
gence between the two regions. In either case, it is clear
that individuals collected from these two genetically dis-
tinct regions might respond very differently to the same
toxin. Thus, region-specific toxicological assessments
appear to be more appropriate then species-wide gener-
alizations.
These data have allowed for the design of a common
garden experiment to test for susceptibility differences
between regions. Because the two regions do not share
haplotypes, sequence data can serve as an identifying
marker. Shrimp from the Gulf of Mexico and the Atlan-
tic Coast now can be placed into a common tank for
toxicology testing, and identified as to origin afterwards
by sequence data; thus eliminating questions surround-
ing protocol and methodology differences between dif-
ferent LC50 studies.
The Office of Research and Development's National Center for Environmental Research
89
-------
Endocrine Disruptors Program Review Workshop
Persistent Organochlorine Compounds, Genetic Susceptibility,
and Testicular Cancer Risk
Stephen M. Schwartz1, Mary L. Biggs', Chu Chen1, Larry L. Needham2, and Dana B. Barr2
*Fred Hutchinson Cancer Research Center, Seattle, WA; 2National Center for Environmental Health, Centers
for Disease Control and Prevention, Atlanta, GA
The incidence of testicular germ cell carcinoma
(TGCC), the most common malignancy developing in
young men, has increased several-fold since the 1950s.
Experimental and observational studies in animal sys-
tems have raised concern that the increasing rates are
due in part to population-wide, persistent exposure to
endocrine-disrupting compounds from industrial and
agricultural applications. Whether human exposure to
such chemicals is associated with TGCC risk has not
been directly studied.
This research project will determine whether the risk
of TGCC is related to serum levels of persistent orga-
nochlorines, focusing on p,p'-1-1 -dichloro-2,2-&;X/?-
chlorophenyl)ethylene (p,p '-DDE), polychlorinated bi-
phenyls (PCBs), and other compounds (e.g., dieldrin,
hexachlorocyclohexanes, hexachlorobenzene). The study
also will examine whether the risk of TGCC associated
with these compounds is modified by genetic suscepti-
bility to mechanisms through which these compounds
may alter TGCC risk. Specifically, it will determine
whether TGCC risk is related to interactions between:
(1) elevated serum/',/''-DDE and polyglutamine repeat
tract polymorphisms in the androgen receptor (AR) gene;
and (2) elevated serum PCB levels and polymorphisms
in oxidative stress defense enzyme genes (e.g., glu-
tathione S-transferases and superoxide dismutases).
An ancillary investigation to the Adult Testicular Can-
cer Lifestyle and Blood Specimen (ATLAS) Study—a
National Cancer Institute-funded, population-based case-
control study of molecular genetic risk factors for
TGCC—will be conducted. ATLAS funding includes popu-
lation-based case and control ascertainment and recruit-
ment; a detailed in-person interview; blood collection; and
molecular genetic analyses of polymorphisms in andro-
gen synthesis, metabolism, and signaling genes (including
AR). The ancillary study will include approximately 250
cases of TGCC and 750 controls recruited as part of AT-
LAS. Serum samples from cases and controls will be as-
sayed for organochlorine pesticides and PCBs by high-
resolution gas chromatography/isotope dilution high-reso-
lution mass spectrometry, and genotyping for common
polymorphisms in genes involved in oxidative stress de-
fense systems (manganese superoxide dismutase, glu-
tathione S-transferases Al, Ml, PI, and Tl) will be per-
formed. Data on the AR and other androgen pathway poly-
morphisms from the main ATLAS Study will be incorpo-
rated. There will be adequate statistical power to detect
relatively weak overall associations, as well as less than
threefold interaction effects between organochlorine com-
pounds and genetic polymorphisms. The results should
add significant new information to the understanding of
the role, if any, of environmental contaminants in the patho-
genesis of TGCC.
90
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disrupters Program Review Workshop
A Case-Cohort Study of Cryptorchidism, Hypospadias, and Delayed
Sexual Maturation in a Dioxin-Contaminated
Region: Chapaevsk, Russia
Oleg Sergeyev1, Boris Revich2, Paige Williams3, Susan Korrick4, Vladimir Zeilert1, Mary M. Lee5,
Tatiana Vshakova6, Igor Saharov7, LarisaAltshul8, andRuss Hauser8
1 Chapaevsk Medical Association, Chapaevsk, Samara Region, Russia;2 Center for Demography and Human
Ecology of Institute for Forecasting, RAS, Moscow, Russia; ^Department of Biostatistics, Harvard School of
Public Health, Boston, MA; 4Channing Laboratory, Department of Medicine, Brigham and Women's
Hospital, Boston, MA, and Department of Environmental Health, Harvard School of Public Health,
Boston, MA; 5Pediatric Endocrine Division, Duke University Medical Center, Durham, NC; 6Cancer
Scientific Research Center, RAMS, Moscow, Russia; 7Ecologoanalytical Center, Research and
Production Enterprise, Moscow, Russia; Department of Environmental Health,
Harvard School of Public Health, Boston, MA
Chapaevsk, Russia, with a population of 83,000, is
located 43 km southwest of Samara on the Chapaevka
River, which flows into the Volga River. One-half of the
city is occupied by industries that are mostly of the mili-
tary-industrial complex. One of the largest environmental
polluters in Chapaevsk is the Middle Volga Chemical
(S VZH, Himprom) Plant, which produced hexachlorocy-
clohexane (lindane) and its derivatives from 1967 to 1987.
Since then, the plant has produced agricultural pesticides
containing chlorine. By-products of the manufacturing
process include dioxins and furans, which contaminated
the region.
Cryptorchidism and hypospadias are common minor
congenital anomalies of male reproductive tract develop-
ment. Although several risk factors have been identified,
the etiology of both disorders remains largely unknown.
Recently, the possibility that these minor reproductive tract
abnormalities may be related to exposure to environmen-
tal contaminants, such as the hormonally active dioxins,
has been raised. Therefore, a study was designed to in-
vestigate the potential association of dioxin exposure to
Cryptorchidism, hypospadias, and delayed sexual matura-
tion in adolescent-aged boys living in Chapaevsk.
The first phase of the study, conducted from March
to May 1999, consisted of a survey of 10 to 16-year-old
boys in Chapaevsk. Participants who agreed to take part
in the study underwent a physical examination, with par-
ticular attention to the presence of Cryptorchidism and
hypospadias as well as the pubertal stage of maturation.
Among 3,041 age-eligible boys identified via birth and
school records, 2,580 (84.8%) boys were enrolled. Physi-
cal examinations of the children were performed by a pe-
diatric endocrinologist and a urologist. Pubertal matura-
tion was graded according to Tanner Staging for Genita-
lia, and pubic hair with testicular volume was determined
using an orchidometer (Prader beads). Testicular location
and the presence of typical orchidopexy postoperative scars
were noted. The presence of hypospadias was based on
the location of the external urethra! meatus. The children's
birth records were reviewed to identify additional cases of
Cryptorchidism and hypospadias. The second phase of
the study, conducted from October 1999 through May
2000, focused on a subset of 112 boys (14 to 16 years
old) from Phase 1 of the study who were identified by the
presence or absence of Cryptorchidism, hypospadias, and
delayed puberty, along with 134 controls chosen using a
case-cohort design. These 246 children were targeted for
a more thorough assessment, including blood and urine
samples for organochlorine contaminant and hormone
measurements, and administration of a detailed question-
naire on medical history, diet, and lifestyle. Among the
246 children, 221 (90%) mother-child pairs agreed to par-
ticipate, and 208 (85%) of these completed the question-
naire. Blood samples collected from 200 mothers (81%)
and 220 children (89%) were archived for later determi-
nation of dioxins, furans, and polychlorinated biphenyls.
The analysis presented in the poster will be a descrip-
tion of the study population with emphasis on residential
proximity to Himprom and other potential predictors for
dioxin exposure, which includes the residential address of
the mother during pregnancy with index son, the work
history of both parents prior to and during the index preg-
nancy, and children's consumption of foods that were lo-
cally grown (vegetables) or raised (cows, chickens, pigs).
Whether these exposure surrogates are associated with case-
control status for Cryptorchidism, hypospadias, and/or de-
layed sexual maturation is being explored. Maps of the dis-
tribution of participants' residence location relative to the
Himprom factory will be generated using an electronic map
of Chapaevsk, constructed with the use of a geographic
map of scale 1:10,000 and ArcView GIS 3.0.
The Office of Research and Development's National Center for Environmental Research
91
-------
Endocrine Disruptors Program Review Workshop
Study for Future Families II: Phthalates in Pregnant
Women and Children
Shanna H. Swan1, Erma Z. Drobnis1, Christina Wang2, J. Bruce Redman3, Amy Sparks4,
Bill L. Lasley5, andAntonia M. Calafat6
University of Missouri, Columbia, MO; 2Harbor-Vniversity of California, Los Angeles Medical Center, Los
Angeles, CA; University of Minnesota, Minneapolis, MN; ^University of Iowa, Iowa City, IA; ^University
of California, Davis, CA; 6Centers for Disease Control and Prevention, Atlanta, GA
The overall objective of this study, Study for Future
Families II (SFFII): Phthalates in Pregnant Women and
Children, is to assess the extent to which current levels of
phthalates in the environment pose a risk to human repro-
ductive health. To date, there have been no studies of the
effects of in utero phthalate exposure on human develop-
ment. The Centers for Disease Control and Prevention
(CDC) recently reported that levels of monobutyl phthalate
(MBP), a urinary metabolite of dibutyl phthalate, a repro-
ductive animal toxicant, were significantly elevated in young
women in a sample of the U.S. population. This research
group conducted a pilot study of 52 pregnant women from
Missouri and observed measurable levels of the four most
prevalent phthalates in all subjects, although levels of MBP
were somewhat lower than those seen in the CDC sample.
The Study for Future Families (SFFI): U.S. Study of
Semen Quality in Partners of Pregnant Women, conducted
by this research team in four U.S. cities, recruited a total
of 800 pregnant women and their partners by July 2002,
to compare time to pregnancy and semen quality among
cities. Most subjects agreed to be recontacted regarding
participation in followup studies, and the majority of women
provided a urine sample while pregnant. SFFI mothers
who agree to participate and their children constitute the
study population for SFFII. Study centers for SFFII in-
clude three of the four SFFI Centers (Columbia, MO; Los
Angeles, CA; and Minneapolis, MN) as well as a new
center in Iowa City, IA. Pediatric physicians conduct stan-
dardized examinations on each child, including digital pho-
tographs of the breast and genitalia. Buccal smears and
urine are being obtained on all children, and urine samples
collected when infants are less than 6 months of age will
be assayed for follicle-stimulating hormone (FSH).
Because the cost of phthalate screening all mother-
child pairs is prohibitive, 50 boys and 50 girls will be se-
lected as "atypical" if either: (1) they demonstrate definite
or probable genital or breast anomalies, or (2) measure-
ments of their breast and/or genitalia fall in the extremes
of the distributions for these parameters (e.g., for males,
small penile size, excess pubic or breast tissue, decreased
anogenital distance (AGO); for females, small clitoral ra-
tio, decreased AGO, excess breast or pubic fat). These
children will constitute the "case" group. Each case will
be matched (1:1) on sex, gestational age, center, and
ethnicity to a child for whom all measurements fall within
normal limits.
These 200 mother-child pairs will comprise the study
population for a nested case-control study. Urinary phtha-
late metabolite levels will be measured in prenatal and post-
natal maternal samples as well as in samples collected dur-
ing the child's first and second years of life. Phthalate me-
tabolite levels in the mother's prenatal urine will be exam-
ined in relation to outcomes from the pediatric examination,
including genital parameters, amount of breast tissue, and
FSH level. These investigators will seek to identify sources
of phthalate exposure by relating mothers' self-reported use
of phthalate-containing products (e.g., soaps, cosmetics,
teething rings, nipples, and other plastics) at the time of
urine collection to measured phthalate metabolite levels. A
sample of breast milk, or the infant's usual formula, also
will be collected at the first visit and stored for future analy-
ses, together with the previously stored maternal serum
and pre- and postnatal urine samples.
The award was received in January 2002, and much of
the first study year was spent developing study documents
and protocols, which will be summarized in this presenta-
tion (and available on the SFFII Web Site). Pediatric physi-
cians and nurse assistants were centrally trained to perform
the specialized genital examination using a training video
developed for this purpose by study researchers. Recruit-
ment began at most centers in May 2002, and progress as
of October 1,2002, will be presented.
92
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disrupters Program Review Workshop
Laboratory and Mesocosm Measurements of the Stereoselective
Degradation of Endosulfan
Spencer S. Walse and John L. Ferry
Department of Chemistry and Biochemistry, University of South Carolina, Columbia, SC
The pesticide endosulfan is known to biodegrade in
the environment to the stable and bioaccumulative en-
dosulfan sulfate. In the laboratory, these researchers
observed that the stereoisomer hydrolyzed more rapidly
than the form. The rate of hydrolysis was measured in
the laboratory in solution and in a variety of suspensions,
in simulated sunlight, and in the dark. Suspended solids
included sea sand, Ti02, Fe203, a-FeOOH, Laponite®,
and Si02. In all cases, there was no effective photo-
chemical processing, but there was a clear selectivity
for the hydrolysis of B endosulfan over a. This observa-
tion was explained by a more stable transition state for
B-endosulfan that was confirmed with ab initio molecu-
lar orbital calculations (STO-6G) on the anionic interme-
diates of endosulfan hydrolysis. The first hydrolysis prod-
uct was endosulfan diol; a series of hydrolytic and oxi-
dation processes (mechanism unknown) yielded a ter-
minal product, endosulfan hydroxy acid. The chemical
characteristics of this new degradation product are re-
ported. The effect of Stereoselective hydrolysis on en-
dosulfan in the environment was measured in three model
estuarine mesocosms (aqueous volume approximately
300 L, Spartina, and associated sediment dwellers) that
were spiked with a, B, or a mixture of a and B (techni-
cal grade). It was observed that B endosulfan hydro-
lyzed rapidly, so comparatively little was available for
biodegradation and conversion to sulfate. From these
measurements, it was determined that approximately 75
percent of total endosulfan-to-endosulfan sulfate con-
version in estuaries is from biodegradation of the a stere-
oisomer. Future work will focus on the fate of the hy-
droxy acid.
The Office of Research and Development's National Center for Environmental Research
93
-------
Endocrine Disruptors Program Review Workshop
Multivariate Modeling of the Photolysis of Aqueous Fipronil
Spencer S. Walse, Li Kong, and John L. Ferry
Department of Chemistry and Biochemistry, University of South Carolina, Columbia, SC
In this study, the correlation between laboratory and
mesocosm studies of the fate of the phenylpyrazole
insecticide fipronil is reported. A multivariate model of
the fate of fipronil was developed in the laboratory to
predict its degradation rate and product profile in the
presence of sunlight, natural organic matter, bicarbon-
ate, and nitrate. There were several abiotic pathways
available for fipronil degradation in this system, includ-
ing direct photolysis and oxidation by hydroxyl radical,
singlet oxygen, or hydrogen peroxide. However, prod-
uct studies indicated that fipronil was quantitatively
converted to desthio-fipronil, a product that is associ-
ated with direct photolysis alone. It was observed that
natural organic matter acted to decrease the rate of
fipronil degradation, either by competition for photons
or transient oxidants. This model was applied to pre-
dict the fate of fipronil in a series of modular estuarine
mesocosms at the National Ocean Service's Center for
Coastal Environmental Health and Biomolecular Re-
search in Charleston, SC. In these experiments, the
loss of aqueous fipronil (single-dose experiment, with
initial fipronil concentrations of 355 and 5,000 ng/L,
three replicates at each concentration) was monitored
over 28 days. Although the direct photolysis product
was detected, the mass balance was dominated by
fipronil-sulfone and fipronil-sulfide, products that are a
signature of fipronil biodegradation.
Nonetheless, direct photolysis appeared to account
for 3-14 percent of all fipronil loss in the model envi-
ronments, at 10 percent ambient ultraviolet-B (limited
by the structure of the mesocosm). All products were
confirmed by comparison of retention times (gas chro-
matography-electron capture detection) and mass spec-
tra (gas chromatography-ion trap mass spectrometry)
to synthesized standards.
94
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disrupters Program Review Workshop
Exposure to DDT and DDE in Relation to Menstrual Cycle Length
Among Laotian Immigrants
Gayle Windham1, Patrick Mitchell1, Myrto Petreas2, Diana Lee1, and Bill L. Lasley3
California Department of Health Services, Oakland, CA; 2California Environmental Protection Agency,
Berkeley, CA; 3Vniversity of California, Davis, CA
This study was initiated because of concern about
the effects of endocrine-disrupting chemicals (EDCs)
on reproductive capacity in wildlife and laboratory ani-
mals, and the fact that there are few data available in
humans. It was designed to examine the effects of po-
tential EDCs on ovarian function, as measured by the
frequency of menstrual cycle characteristics, in Laotian
immigrants who may have higher exposure from their
home country as well as from sport fish consumption.
Working with local Lao communities, Lao field workers
were hired and trained to recruit women of reproductive
age (18-39) who were born in Southeast Asia and regu-
larly consumed fish. Participants were asked to com-
plete two interviews, provide a blood sample, and collect
first morning urine samples daily during three menstrual
cycles. The urine samples were assayed for metabolites
of estrogen and progesterone, from which several men-
strual cycle parameters were defined. These methods
were similar to those of a larger study conducted among
a general population. The serum was assayed for orga-
nochlorine compounds including pesticides, polychlori-
nated biphenyls, and polybrominated diphenyl ethers.
The target of 50 women completed the study. Urine
was collected during 148 complete cycles, with infor-
mation on an additional 39 cycles that had partial urine
collection. The mean cycle length in complete cycles
was 30.4 days (standard deviation [SD] 7.0). Focusing
on dichlorodiphenyltrichloroethane (DDT) and its me-
tabolite 1,1 -dichloro-2,2-bis(p-chlorophenyl)ethylene
(DDE), these investigators found that all women had
detectable levels, with a mean of 1.7 ppb (SD 3.5) and
20.3 ppb (SD 22.5) respectively, indicating higher body
burdens than currently found in comparable U.S. popu-
lations. The covariates related to mean exposure level
included age, parity, breastfeeding, and time spent in
Thailand, some of which were inter-related. Examining
quartiles of exposure, cycle length was decreased with
increasing exposure in preliminary analyses. Using meth-
ods that account for repeated measures, at the highest
quartile of DDE exposure mean cycle length was de-
creased by 3.5 days (95% confidence interval [CI] -7.6,
0.58) and at the highest quartile of DDT exposure by 4.4
days (CI -8.2, -0.58), compared to the lowest quartile.
Adjusting for lipid level attenuated the decrements only
slightly. Adjusting for other demographic variables at-
tenuated the decrements in cycle length for DDT, but
not DDE. Examining the log of DDE or DDT as a con-
tinuous variable, decreasing cycle length with increasing
exposure was found that was statistically significant for
DDE. There was a decrement in luteal phase length ex-
amining either the quartiles or logged body burdens.
hi summary, these data suggest a possible associa-
tion of DDE/DDT exposure with changes in menstrual
function, but the findings are based on small numbers
and potentially confounded by other exposures. If expo-
sure is associated with shorter cycles, this could reflect
a hormonal pathway. Some laboratory studies have indi-
cated that DDE may interfere with progesterone, which
would be consistent with an effect during the luteal phase,
when progesterone is primarily secreted. The investiga-
tors plan to examine hormone endpoints and the other
EDC exposures as well.
The Office of Research and Development's National Center for Environmental Research
95
-------
Mechanisms of Action
-------
Endocrine Disrupters Program Review Workshop
Alteration of Calcium-Dependent Cell Signaling as a Potential
Mechanism of Endocrine Disruption
Abby D. Benn'mghoff and Peter Thomas
University of Texas at Austin, PortAransas, TX
The importance of calcium-dependent signal trans-
duction in mediating hormonal control of ovarian sex
steroid production has been demonstrated in numerous
vertebrate species, including the Atlantic croaker (Micro-
pogonias undulatus). Additionally, studies have shown
that specific components of calcium-dependent signal
transduction pathways, including calcium channels and
calmodulin, are targets for environmental chemicals such
as the heavy metals cadmium, lead, and mercury as well
as the pesticides lindane, 4-octylphenol, and o,p'di-
chlorodiphenyltrichloroethane (o,p'-DDT). However,
there is little evidence directly linking these actions of
xenobiotic chemicals to adverse endocrine effects, such
as impairment of steroidogenesis. Therefore, the goal of
this research project is to investigate environmental chemi-
cal disruption of calcium-dependent signaling as a po-
tential novel mechanism of endocrine disruption in the
Atlantic croaker.
Primary cell culture. A primary cell co-culture sys-
tem for ovarian theca and granulosa (T-G) cells was
established to conduct specific experiments investigat-
ing the direct actions of environmental chemicals on cal-
cium-signaling pathways. T-G cells were obtained by
enzymatic digestion of mature ovarian tissue and iso-
lated from other cell types by density-gradient centrifu-
gation. These cells adapted well to culture conditions
and produced high levels of testosterone in response to
gonadotropin for at least 5 days. Croaker T-G cells main-
tain their endocrine function in primary culture, and thus
are a useful model for investigating mechanisms of en-
docrine disruption.
Signal transduction pathways. Previous research
on croaker whole ovarian follicles has demonstrated a
clear role for calcium-dependent cell signaling in modu-
lating ovarian steroidogenesis. Therefore, experiments
were conducted on T-G cells to determine if these path-
ways are maintained in culture. Calcium ionophore
A23187 (0.25 uM), a drug that increases cytosolic cal-
cium concentrations, stimulated basal testosterone pro-
duction twofold. Gonadotropin-induced testosterone pro-
duction was attenuated by the voltage-sensitive calcium
channel (VSCC) blocker verapamil (1-10 uM) and the
calmodulin inhibitor W-7 (20 uM). These data show that
increases in cytosolic calcium concentrations via ion
transport through VSCCs and activation of calmodulin
are required to transduce the initial hormone signal and
stimulate steroid production.
Effects of environmental chemicals on steroido-
genesis. Because the chemicals cadmium and o,p'-DDT
have been shown to alter calcium function, preliminary
experiments were conducted to examine their effects on
the endocrine function of T-G cells. Cadmium (10 ppm)
significantly impaired hormone-induced testosterone pro-
duction. Alternatively, treatment with low doses of o,p'-
DDT (0.001-0.01 ppm) stimulated basal steroidogen-
esis. These results demonstrate that a viable and sensi-
tive bioassay has been established for investigating ef-
fects of environmental chemicals.
Effects of xenobiotics on cytosolic calcium. Experi-
ments are ongoing to determine if these chemicals directly
modify calcium-signaling pathways. Epi-fluorescence,
time-lapse microscopy will be used to measure both spa-
tial and temporal cytosolic calcium changes in individual
T-G cells exposed to heavy metals, pesticides, polycyclic
aromatic hydrocarbons, polychlorinated biphenyls, and
other chemicals to determine if they directly alter intracel-
lular calcium concentrations under basal or hormone-stimu-
lated conditions. Due to the ubiquitous nature of calcium
signaling among cell types and organisms, the results of
this study will be applicable to a wide variety of calcium-
dependent processes in many different tissue types and
species. Therefore, this research has the potential to sig-
nificantly contribute to the general understanding of endo-
crine disruption.
The Office of Research and Development's National Center for Environmental Research
99
-------
Endocrine Disruptors Program Review Workshop
Elevations of Estradiol in the Cycling and Ovariectomized,
Estradiol-lmplanted Female Rat by the Drinking Water
Disinfection By-Product Dibromoacetic Acid
Jerome M. Goldman and Ashley S. Murr
Reproductive Toxicology Division, National Health and Environmental Effects Research Laboratory, Office of
Research and Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
The chlorination of municipal drinking water has been
one of the most successful public health interventions to
combat disease. However, a variety of treatment-related
chemicals are generated in the process that have been
suspected of having adverse effects in both humans and
test animal species. Although some positive associations
have been drawn between various disinfection by-prod-
ucts (DBFs) and some forms of cancer, there are indi-
cations that reproductive functions may be compromised
as well. The haloacetic acids are one of the principal
types of DBFs generated by chlorination, and two mem-
bers of this class, dibromoacetic (DBA) and dichloroactic
acids, have been reported to affect sperm production
and gonadal hormonal activity in the male rat. At high
doses, DBA also has been found to induce a persistent
alteration in cyclicity in the female Sprague-Dawley (S-
D) rat, which has been confirmed in the present study
using 14-day oral exposures (0-270 mg/kg). Body
weights during dosing remained unaffected.
Preliminary results suggested that an alteration in es-
tradiol (E2) concentrations may have contributed to this
effect on cycling status. To investigate this possibility fur-
ther, normal-cycling S-D female rats (60-90 days) were
gavaged daily with DBA (0,30,90, or 270 mg/kg in wa-
ter, pH adjusted to 6.8) for 2 weeks. After 8 days of treat-
ment (during which cycles were maintained), all animals
were ovariectomized (OVX), and 3 days thereafter im-
planted subdermally with silastic capsules (6 mm in length)
containing estradiol benzoate (4 mg/mL in sesame oil).
Blood was sampled by tail nick at 24, 48, and 72 hours
postimplant. Capsules containing vehicle only also were
implanted, and blood was sampled at 72 hours for 0 and
270 mg/kg-treated rats. Intact rats with regular 4-day cycles
also were gavaged with DBA (0, 30, 60, or 120 mg/kg)
for 2 weeks as above, and blood samples were taken dur-
ing the last 4 days of treatment. Cycling status was moni-
tored to ensure that normal cycles were maintained at
these dosages. Only those females continuing to exhibit
regular 4-day estrous cycles were retained. For OVX rats,
data showed that by 72 hours, treatment with DBA caused
a distinct dose-related elevation in E2, and that the highest
dose had a mean concentration approximately 2!/2-fold
above the controls. For intact, 4-day cycling rats, E2 con-
centrations remained elevated on the day of estrus for the
two highest dose groups, suggesting that this effect may
have contributed to the emerging disruption in cyclicity
previously observed.
To explore further whether the elevation in E2 ob-
served in the OVX/E2-implanted rats was due to an im-
pairment in E2 metabolism, females were gavaged with
DBA (0 or 270 mg/kg) for 2 weeks. During the final 3
days, they were given phenobarbital (PhB, sodium salt)
either in the drinking water (0.1%) or by intraperitoneal
injection (20 mg/kg) to increase hepatic E2 metabolism.
Neither PhB route caused a lowering of the DBA-induced
E2 elevations, suggesting that the effects were linked to
an alteration in the clearance of the hormone or a suicide
inhibition of hepatic P450 activity. The data also indicate
that although DBA administered over the course of 2 weeks
in adult rats can act as an endocrine-disrupting chemical
at concentrations in excess of those levels permissible for
human exposure, it is conceivable that lower dosages given
over more extended treatment periods could have similar
adverse effects on reproductive function.
100
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disrupters Program Review Workshop
Testosterone:Fatty Acid Esterification: A Novel Target of Endocrine
Disruption Caused by Tributyltin
Meredith P. Goading and Gerald A. LeBlanc
North Carolina State University, Raleigh, NC
Imposex, the development of a penis and/or vas defer-
ens in female gastropods, is regarded as the best example of
endocrine disruption in an invertebrate linked to an environ-
mental pollutant. This masculinization of females is causally
associated with increased testosterone levels resulting from
exposure to the marine biocide tributyltin (TBT). Endocrine
disruption in gastropods is of global concern, with imposex
having been demonstrated in more than 150 species world-
wide. The development of screening approaches to detect
this mode of endocrine disruption is contingent on elucidat-
ing the precise endocrine process that is affected by TBT.
The objectives of this study were to: (1) elucidate major
testosterone biotransformation processes in the mud snail
(Ilyanassa obsolete/) that may be disrupted by TBT, result-
ing in increased testosterone levels; and (2) assess the ef-
fects of TBT on these processes. The study discovered
that testosterone undergoes limited biotransformation in the
mud snail to aromatized, hydroxylated, or oxidoreduced
derivatives. Rather, testosterone is predominantly converted
to apolar esters that are retained by the organisms. These
esters are largely fatty acid conjugates that are produced by
one or more acyl CoA:testosterone acyltransferase enzymes.
It was hypothesized that fatty acid esterification of test-
osterone serves to store excess testosterone in a lipid-soluble
form that could be drawn on as testosterone is needed by
the organism. To test this hypothesis, snails were treated
with various regimens of 4-nonylphenol or testosterone to
decrease or increase, respectively, total testosterone levels.
Total, free (unesterified), and esterified testosterone levels
then were evaluated in individual snails. The amount of es-
terified testosterone present in these organisms increased in
direct proportion to the level of total testosterone, while free
testosterone levels remained relatively constant. Therefore,
in this species, testosterone esterification maintains free tes-
tosterone by removing and storing testosterone when levels
are in excess, and perhaps by providing testosterone when
levels are below normal.
The effects of TBT on this major testosterone ho-
meostatic process next were investigated. Snails were
exposed to environmentally relevant concentrations of TBT,
and free, esterified, and total testosterone levels were as-
sessed. Total testosterone levels in snails were not altered
by TBT; however, free testosterone levels increased with
increasing TBT exposure concentration. Further, the re-
tention of testosterone fatty acid esters was decreased
with increasing exposure concentration of TBT. These
results indicate that TBT elevates free testosterone levels
in snails by decreasing the accumulation of fatty acid es-
ters. A comparative assessment of testosterone homeo-
stasis in snails collected from a tin-impacted and non-
impacted site confirmed that tin- exposed, imposexed snails
retain less testosterone:fatty acid ester and more free tes-
tosterone.
Results from this study have identified a novel target
of endocrine disruption—the fatty acid esterification of
testosterone, which may be responsible for TBT-induced
imposex in snails and also may be of relevance to mam-
mals.
The Office of Research and Development's National Center for Environmental Research
101
-------
Endocrine Disruptors Program Review Workshop
HPLC Purification of an Endocrine-Disrupting Retinoid Activity
From a Minnesota Lake With a High Incidence of
Malformed Amphibians
Felix Grun, Aristocle Ndayibagira, David M. Gardiner, and Bruce Blumberg
University of Calif ornia, Irvine, CA
There is considerable evidence that wildlife and do-
mestic animals have suffered adverse consequences from
exposure to environmental chemicals that interact with
components of the endocrine system. To assess the
significance of endocrine disrupters that activate retin-
oid-signaling pathways for their role in causing limb de-
velopmental deformities in frogs, and to understand their
mechanism of action to assess their implications for
human health, this research project studied the effects
of disruption of retinoid-sensitive signaling pathways.
These studies demonstrated that all the malformations
observed in wild populations of frogs can be induced
by experimental exposure to retinoids. The model that
environmental retinoids is the cause of frog deformities
predicts that retinoids will be found at sites where de-
formed frogs are found. To identify such environmen-
tal retinoids, hydrophobic substances are extracted from
water samples using solid phase extraction. These sub-
stances then are fractionated by high-performance liq-
uid chromatography and tested for their ability to acti-
vate the retinoic acid receptor in transient transfection
assays (increased Relative Luciferase Units; see Figure
1). Active fractions are purified to homogeneity as judged
by ultraviolet absorption spectra and then analyzed by
electrospray and electron impact mass spectroscopy for
exact mass determination. Mass spectra will be used to
derive the compound's molecular formula and molecu-
lar structure.
102
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disrupters Program Review Workshop
! 25 27 29 31 S3 35 37 39 41 43 45 47 49 51 53 55 57 59
C-18 Analytical
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59
Figure 1. Retinoic acid receptor activation by high-performance liquid chromatography fractions of water from
a lake in Minnesota (R.L.U. = Relative Luciferase Units).
The Office of Research and Development's National Center for Environmental Research 103
-------
Endocrine Disruptors Program Review Workshop
Differential Binding of Estrogens and Estrogenic Compounds
to ERs Alpha, Beta, and Gamma of the Atlantic Croaker,
Micropogonias undulatus
Mary Beth Hawkins and Peter Thomas
The University of Texas at Austin Marine Science Institute, PortAransas, TX
These researchers recently described a novel form of
estrogen receptor (ER), ER gamma, in a teleost fish, the
Atlantic croaker. Atlantic croaker ER gamma (acER
gamma) is genetically distinct and has a different tissue
distribution pattern than acER alpha or beta. Phylogenetic
analysis shows that ER gamma is present in other teleost
fish, but was not recognized as such due to lack of evi-
dence for three ERs in these species. ER gamma shares a
high degree of amino acid similarity with other ERs. How-
ever, acER gamma differs from acER alpha and acER
beta in the ligand-binding domain (LED), and these differ-
ences are conserved in all other ER gammas. To deter-
mine whether these amino acid changes have led to dif-
ferences in ligand-binding specificity between the three
ER types, ER fusion proteins were created for bacterial
expression including the E and F domains of acER alpha,
acER beta, and acER gamma. Bacterially expressed fu-
sion proteins for acER alpha, beta, and gamma show spe-
cific, high-affinity binding to [3H] estradiol (E2) with dis-
sociation constants of 0.61± 0.013, 0.40 + 0.006, and
0.38 + 0.059 nM, respectively. The rank orders of bind-
ing to the acER fusion proteins are DES » ICI182 >
TOH > ICI164 > E2 > ZEAR > MOX E >TAM > El >
17BE2 > E3 > 20H E = GEN » RU 486 for acER alpha;
ICI182 > DES > TOH > E2 > ICI164 > GEN > MOX E
> TAM > ZEAR=El > E3 = 17BE2 > 20H E » RU 486
for acERbeta; and E2 > DES > TOH > ICI182 > ICI164
> E3 > GEN > MOX E > ZEAR > El > 17BE2 > RU 486
> TAM > 20H E for acER gamma.
The differences in relative binding affinities that acER
alpha, beta, and gamma fusion proteins show for native
estrogens and other estrogenic compounds suggest that
amino acid changes in the LED may confer different func-
tional properties to the three ER subtypes. These research-
ers currently are investigating the binding characteristics
of xenoestrogens such as kepone, dichlorodiphenyl-
trichloroethane, and polychlorinated biphenyls to the three
ER subtypes. Differential binding of xenoestrogens to ER
subtypes could result in distinct physiological responses
to environmental exposures, depending upon the ER sub-
type population makeup of target organs.
104
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disrupters Program Review Workshop
A Short-Term Dosing Model for Detecting the Effects of
Environmental Contaminants on Thyroid Hormones
in the Rat: Effects of Pesticides
Joan M. Hedge1, David G Ross2, Michael J. DeVtto2, and Kevin M. Crofton1
'Neurotoxicology and Experimental Toxicology Divisions, National Health and Environmental Effects
Research Laboratory, Office of Research and Development, U.S. Environmental Protection Agency,
Research Triangle Park, NC
Recently, a short-term rat-dosing model was devel-
oped to examine the effects of environmental mixtures on
thyroid homeostasis (TH). Prototypic chemicals such as
dioxins, polychlorinated biphenyls, and polybrominated
diphenyl ethers have been tested and shown to adversely
impact both neurological development and TH, primarily
by upregulating catabolism of thyroid hormones by liver
enzymes. Current efforts examined the effects of select
pesticides in this model. Female Long Evans rats (28 days
old) were orally dosed for 4 consecutive days with the
dithiocarbamate fungicide mancozeb (MAN) or the herbi-
cide pronamide (PRO) (0, 3.9, 7.8, 15.6, 31.25, 62.5,
125, 250, 500, 1,000 mg/kg/day), or the dithiocarbamate
fungicide thiram (THI) (0, 6.25, 12.5, 25, 50, 100, 200,
400, 800 mg/kg/day). Serum and liver samples were col-
lected 24 hours after the last dose. Triiodothyronine (T3)
and thyroxine (T4) were measured via radioimmunoas-
says. Hepatic ethoxy-resorufin-0-deethylase (EROD),
pentoxy-resorufin-0-deethylase (PROD), and uridinedi-
phosphate-glucuronsyltransferase were determined in he-
patic microsomal fractions. Liver-to-body weight ratios
(LBR) increased for PRO at doses of > 125 mg/kg/day, for
MAN at > 500 mg/kg/day, and for THI at > 400 mg/kg/
day. PRO, MAN, and THI all produced similar dose-related
decreases in T4, with estimated median effective doses of
approximately 250 mg/kg/day. Potencies for decreases in
T3 were THI > PRO > MAN. Maximal suppression of T3
and T4 was approximately 50 percent and 80 percent at the
highest doses, respectively. Neither MAN nor THI caused
changes in EROD or PROD. These data suggest that this
rodent-dosing model is sensitive to short-term perturba-
tions in thyroid hormones caused by these pesticides.
The Office of Research and Development's National Center for Environmental Research
105
-------
Endocrine Disruptors Program Review Workshop
Impact of Chemical Mixtures on Steroid Metabolism
in Mud Snails (llyanassa obsolete)
Patricia D. McClettan-Green1'2
Department of Environmental and Molecular Toxicology, North Carolina State University,
Raleigh, NC; 2Duke University Marine Laboratory, Beaufort, NC
Imposex snails are found in harbors and near marinas
where they often are exposed to a wide mixture of chemi-
cals. The antifouling compound tributyltin (TBT) has been
causally linked to the induction of imposex in gastropods;
however, the effect of co-exposure to other environmen-
tal contaminants has not been studied. Although TBT is a
cytochrome P450 inhibitor, many environmental contami-
nants such as polycyclic aromatic hydrocarbons (PAHs)
induce the expression of cytochrome P450. Mud snails
(llyanassa obsoletd) were dosed with TBT (20 ng/L) and/
or the model PAH 3-methylcholanthrene (3-MC) (10 nM)
for 45 days. The snails were examined for the induction
of imposex and changes in P450 activity as well as for
effects on steroids biosynthesis.
Control snails had an imposex level of 1.93 percent
with a relative penis size index (RPSI) of 0.005, TBT-
exposed snails exhibited a 40.42 percent imposex induc-
tion with an RPSI of 2.07, 3-MC snails exhibited a 5.05
percent imposex induction with an RPSI of 0.001, and
TBT + 3-MC snails exhibited an 11.95 percent imposex
induction with an RPSI of 0.880. Snails exposed to TBT
exhibited a decrease in P4501A activity (7-ethoxycou-
marin 0-deethylation activity = 0.0003 nmol/min/mg
compared to 0.0055 nmol/min/mg for controls). Snails
exposed to 3-MC exhibited an increase in activity (0.0095
nmol/min/mg), while animals exposed to both TBT and
3-MC had an intermediate level of activity (0.0068 nmol/
min/mg).
Analysis of steroid metabolism indicated that significant
effects occurred in vivo in snails exposed to TBT or 3-MC.
Androstenedione metabolism was measured by the ^O-
release assay. Snails exposed to TBT or 3-MC both exhib-
ited a significant decrease (approximately 50%) in their con-
version of androstenedione to testosterone. Co-exposure of
snails to both contaminants resulted in metabolic activity
similar to the controls (see Figure 1). In addition, other
P450-mediated steroid pathways such as the metabolism
of pregnenolone to 17a-hydroxypregnenolone or progest-
erone to androstenedione were inhibited by exposure to TBT
or 3-MC.
This clearly indicates that although TBT does inhibit
P450 aromatase activity, this inhibition does not account
for increases in testosterone in the exposed animals. 3-
MC also inhibited aromatase activity, yet had no effect on
imposex induction, hi addition, it is quite evident that each
of the P450-mediated steroid pathways was significantly
affected by exposure to the contaminants. Therefore, it is
reasonable to assume that a more diverse mechanism is
responsible for imposex induction.
106
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disrupters Program Review Workshop
Control
TBT
3-MC
TBT + 3-MC
Figure 1. Aromatase activity in mud snails exposed to 20 ng/L TBT or 10 nM 3-MC. Concentration is given as
the mean (n = 10) + standard deviation. Asterisks indicate a significant difference from the controls.
The Office of Research and Development's National Center for Environmental Research
107
-------
Endocrine Disruptors Program Review Workshop
Mechanisms of Toxicity of Dioxin-Like Compounds
to Primate Ovarian Cells
Francisco M. Mordn1, Peter Lohstroh2, Catherine A. VandeVoort3, Tiangang Chen2, James W. Overstreet2,
Alan J. Conley1, and Bill L. Lasley2
Population Health and Reproduction, 2Centerfor Health and the Environment, California Regional Primate
Research Center, University of California, Davis, CA
Endocrine disrupters such as 2,3,7,8-tetra-chloro-
dibenzo-p-dioxin (TCDD) induce abortion in non-human
primates and decrease estradiol (E2) production at the
level of the ovaries. To better understand how this endo-
crine disrupter acts on its target tissue, the effects of
TCDD exposures on steroidogenesis were studied in
human luteinized granulosa cells (hLGC). TCDD inhib-
ited the secretion of E2 without affecting either proges-
terone or 17a-hydroxyprogesterone. It was hypothesized
that the molecular target of TCDD, inhibiting E2 pro-
duction by hLGC, was 17a-hydroxylase/l 7,20-lyase cy-
tochrome P450 (P450cl7), the enzyme catalyzing the
synthesis of androgens. In support of this hypothesis,
TCDD (10 nM) decreased E2 production without alter-
ing either E2 metabolism or levels of aromatase cyto-
chrome P450 (P450arom). The decrease in E2 induced
by TCDD was ameliorated by the addition of androgen
substrates, but not substrates earlier in the metabolic
pathway. An anti-human P450cl7 antisera was devel-
oped, and a direct radiometric assay of the 17,20-lyase
activity of P450cl7 was adapted to further test the hy-
pothesis. Western immunoblot analysis demonstrated that
TCDD treatment of hLGC decreased the expression of
P450cl7 by as much 50 percent. 17,20-lyase was more
than 10-fold lower than P45 Oarom activity and therefore
likely limited E2 synthesis. TCDD induced a 65 percent
decrease in 17,20-lyase activity, but no changes were
seen in either P45 Oarom or in nicotinamide adenine di-
nucleotide phosphate-cytochrome P450 oxidoreductase
(reductase) that supports these microsomal P450 activi-
ties. This research study has concluded that the molecu-
lar target for endocrine disruption of hLGC by TCDD is
P450cl7, acting to specifically decrease the supply of
androgens that are essential for E2 synthesis, and that
the toxic effect does not involve either P45 Oarom or the
redox partner protein reductase.
It is well known that TCDD has a long half-life in
vivo. To test the hypothesis that this endocrine disrupter
can have persistent adverse effects on female reproduc-
tive health, the ovarian function in mature female cyno-
molgus macaques was evaluated more than 1 year after a
single exposure. Urinary estrone conjugates (E1C), preg-
nanediol-3-glucuronide (PdG), and follicle-stimulating
hormone (FSH) were measured in animals following a
single oral exposure (1, 2, or 4 ug/kg BW) to TCDD.
Three out of four animals in the high-dose group revealed
no evidence of menstrual cycles. These noncycling ani-
mals had baseline E1C concentrations without ovulatory
mid-cycle peaks and monotonic PdG profiles. Mean FSH
concentrations during the mid-follicular phase of the me-
dium dose group and during the entire cycle of the high-
dose group were elevated, and the endometria of the
noncycling animals were inactive. These data demonstrate
that a single exposure of TCDD leads to long-term ad-
verse effects on ovarian function in non-human primates,
most likely by blocking estrogen production.
The next steps are to: (1) isolate RNA for a microarray
analysis to identify the gene(s) that may be up- or down-
regulated by the TCDD treatment on hLGC; (2) analyze the
production of transcript for P45 Oc 17 by reverse transcriptase
polymerase chain reaction; (3) examine the rate of tran-
scription of P450cl7 and reductase by nuclear runoff as-
say; and (4) investigate the effects of TCDD on phospho-
rylation of P450cl7 and reductase by incubating with
32P-phosphoric acid followed by immunoprecipitation with
specific antisera.
108
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disrupters Program Review Workshop
Developmental Toxicity of Antiecdysteroids
in the Crustacean Daphnia magna
Xueyan Mu and Gerald A. LeBlanc
North Carolina State University, Raleigh NC
Ecdysteroids, polyhydroxylated steroid hormones,
regulate the molting processes in crustacean and other
arthropod species. These researchers have shown that
ecdysteroids also play an important role in crustacean
embryo development. It was hypothesized that envi-
ronmental chemicals eliciting antiecdysteroidal activity
could be detrimental to crustacean embryo develop-
ment and fecundity. Various chemicals were evaluated
for potential antiecdysteroid activity. Fenarimol, an ag-
riculture fungicide, and testosterone delayed molting of
neonatal daphnids (Daphnia magna) in a concentra-
tion-dependent fashion, and this effect was mitigated
by co-exposure to exogenous 20-hydroxyecdysone.
These results implicated these compounds as having
antiecdysteroid activity. Exposure of either gravid ma-
ternal organisms or isolated embryos to fenarimol or
testosterone resulted in embryo abnormalities that could
be prevented by co-exposure to 20-hydroxyecdysone.
Abnormalities of fenarimol-exposed isolated embryos
were associated with the late stages of development;
while testosterone-exposed isolated embryos exhibited
abnormalities associated with both early and late stages
of development. Fenarimol, but not testosterone, was
found to lower endogenous ecdysone levels. This effect
presumably was due to the known ability of fenarimol
to inhibit steroidogenic enzymes. Conversely, testoster-
one, but not fenarimol, was found to inhibit the action
of ecdysteroids in an ecdysone-responsive cell line.
These results suggest that testosterone exhibits
antiecdysteroid activity by acting as an ecdysone re-
ceptor antagonist.
Analysis of ecdysone levels in the normal developing
embryo revealed that ecdysone levels initially are high
and progressively decrease through mid-development.
Ecdydone levels then increase during the latter stages of
development. It was surmised that the initial pool of ecdys-
one in the embryo is of maternal origin, and the latter
pool results from de novo synthesis in the embryo fol-
lowing organogenesis. Consistent with the demonstrated
modes of actions of fenarimol and testosterone, the in-
hibitor of ecdysone synthesis (fenarimol) elicited effects
on isolated embryos that were associated only with the
pool of newly synthesized ecdysone. The ecdysone re-
ceptor antagonist (testosterone) elicited effects on iso-
lated embryos that were independent of the ecdysone
pool involved. Both fenarimol and testosterone were found
to reduce fecundity of daphnids at exposure concentra-
tions that elicited antiecdysteroid activity. These results
demonstrate mechanisms of action and consequences
of antiecdysteroids. Results also demonstrate the utility
of isolated daphnid embryos to detect antiecdysteroid
activity of chemicals and provide mechanistic informa-
tion regarding the source of this activity. Future studies
will focus on modeling and experimentally demonstrat-
ing the consequences of exposure to combinations of
chemicals that elicit antiecdysteroidal activity through
various mechanisms.
The Office of Research and Development's National Center for Environmental Research
109
-------
Endocrine Disruptors Program Review Workshop
Levels of the Neuropeptide Hormone APGWamide Are
Elevated in TBT-Dosed Snails and in Snails Transferred
to a Contaminated Site
Eva Oberddrster1 and Patricia D. McClellan-Green2'3
'Southern Methodist University, Dallas, TX; 2Duke University Marine Laboratory, Beaufort, NC;
3North Carolina State University, Raleigh, NC
APGWamide, the putative penis morphogenic factor
that normally causes male sex characteristic develop-
ment in breeding male snails, also can induce imposex in
female snails at very low doses (10~16 moles) and in a
short time period (7 days). In this study, snails were
either collected from a relatively pristine estuarine re-
serve and injected subcutaneously with either 20 ng
tributyltin (TBT), 500 ng testosterone, or vehicle (EtOH)
controls. In addition, snails from the pristine location
were caged for 3 months in a site that historically has
100 percent imposex females and males without regress-
ing sex organs. APGWamide immunoreactivity was mea-
sured in snail body homogenates via Western blotting.
Samples were run in duplicate, and an internal standard
was run in all gels. Results are given as percent of the
internal standard. Gels also were run with increasing levels
of APGWamide, and standard curves were generated
showing linearity over a wide range of APGWamide con-
centrations.
The results show that control males had significantly
higher APGWamide levels than control females (see Fig-
ure 1). All TBT-treated animals, whether male, female,
or imposex, had levels of APGWamide similar to control
males and significantly higher than control females. In
testosterone-treated animals, APGWamide levels were
identical to the corresponding vehicle controls. That is,
males had similar levels to control males; females had
similar levels to control females, and imposex animals (n
= 1) also were not different from one another (additional
data analysis ongoing). This indicates that there may be
two different mechanisms of imposex induction, depend-
ing on whether the exposure is TBT or testosterone. In
TBT-dosed animals, APGWamide levels were elevated.
Even in normal TBT-dosed females, levels were elevated,
although imposex was not yet induced. In TBT-dosed
imposex snails, APGWamide levels were higher than in
any other dose group. In testosterone-treated snails, there
were no changes in APGWamide levels, but imposex
was induced. It is possible that testosterone acts via a
different mechanism than TBT. For the transferred ani-
mals, both males and females had significantly higher
levels of APGWamide than their respective controls. It is
clear that APGWamide is upregulated significantly in these
transferred animals, and data analysis for imposex ani-
mals from this transfer is ongoing.
110
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disrupters Program Review Workshop
APGWamide levels in mud snails
300
250
!= 150
I
•a
4)
100
I 50
v / A males
ivv* females
imposex
I
Control
TBT Testosterone Transfers
Figure 1. APGWamide levels in snail homogenates as analyzed by Western Blotting. Control = vehicle injected;
TBT = 20 ng tributyltin; testosterone = 500 ng; transfers = snails transferred from pristine site to high
imposex site. Significance by ANOVA followed by post-hoc Tukey test. Where there are no error
bars, n = 1. Otherwise, n = from 5 to 20. Letters above bars indicate no significant differences.
The Office of Research and Development's National Center for Environmental Research
111
-------
Endocrine Disruptors Program Review Workshop
Endocrinology of Sex Determination in a Crustacean and Its
Disruption by Biorational Insecticides
Allen W. Olmstead and Gerald A. LeBlanc
North Carolina State University, Raleigh, NC
Terpenoid hormones are one of the two major classes
of non-peptide hormones (ecdysteroids being the sec-
ond) found in insects and crustaceans. Juvenile hormone
is the predominant terpenoid hormone of insects, while
methyl farnesoate is produced by crustaceans. These
hormones are similar to the retinoid hormones found in
vertebrates and have major roles in development and re-
production. The objectives of this study were to:
(1) evaluate the role of methyl farnesoate in sex determi-
nation in the crustacean Daphnia magna, and (2) inves-
tigate the ability of insecticides that act as juvenile hor-
mone analogs in target species to disrupt methyl
farnesoate-regulated developmental processes. Chronic
exposure (3 weeks) of small daphnid populations to con-
centrations of methyl farnesoate dramatically increased
the production of male progeny with a lowest observed
effect concentration (LOEC) of 80 nM. Acute exposure
(24 hours) of individual maternal daphnids to methyl
farnesoate during the period of late ovarian oocyte matu-
ration caused these oocytes to develop exclusively into
males. These results indicate that methyl farnesoate is
responsible for programming oocytes to develop into
male offspring.
The insecticidal juvenile hormone analogs methoprene
and pyriproxyfen next were evaluated for their ability to
mimic the effect of methyl farnesoate on sex determina-
tion of offspring. Both compounds increased male prog-
eny production in small populations after chronic expo-
sure with LOECs of 32 nM and 0.39 nM, respectively.
Several non-juvenoid xenobiotics were similarly evalu-
ated and had no effect on sex determination. As with
methyl farnesoate, the juvenile hormone analogs deter-
mined sex of offspring when exposure occurred during
the sensitive period of oocyte development. Daphnids
were exposed to binary mixtures of methoprene-methyl
farnesoate or pyriproxyfen-methyl farnesoate to deter-
mine if these chemicals conformed to a model for con-
centration addition (indicative of same mechanism of
action) or independent joint action (indicative of separate
mechanisms of action). Although both mixtures con-
formed more to the concentration additive model, some
degree of synergism was detected between the insecti-
cides and the crustacean hormone. In conclusion, this
research project found that: (1) the crustacean terpenoid
hormone methyl farnesoate determines sex in daphnids,
(2) biorational insecticides with juvenile hormone activ-
ity can mimic the effects of methyl farnesoate, and (3)
the mechanism of action of this effect is similar for both
the hormone and the insecticides. Future studies will
evaluate the relationship among endogenous methyl
farnesoate levels, reproduction, and sex of offspring as
well as elucidate the mechanism of synergism between
these insecticides and methyl farnesoate.
112
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disrupters Program Review Workshop
Cloning and Characterization of Estrogen Receptor a, -Ba, and -Bb
in Zebrafish: Differential Expression During Development and
Effects of Steroids and Environmental Chemicals
Sarah Rothberg, Apolonia Novillo, Mitsuyo Kishida, BikemAkten, and Gloria V. Collard
Boston University, Boston MA
Estrogen actions on growth, differentiation, and prolif-
eration of a range of tissues, cell types, and life stages are
mediated chiefly through ligand-dependent transcription fac-
tors termed estrogen receptors (ER). Previous studies in
this laboratory demonstrated that estrogen and estrogen-
like chemicals upregulate expression of the predominant
aromatase isoform (P450aromB) in fish brain, and further
identified estrogen response elements in the 5'-flanking re-
gion of the P450aromB-encoding gene (cyp!9B). To obtain
ER for use as reagents in transcriptional analysis and local-
ization studies, and to determine whether estrogenic effects
on expression of P450aromB and other neural genes could
be mediated in part by changes in ER expression, a stepwise
polymerase chain reaction (PCR) cloning strategy was used
to isolate the cDNA(s) encoding ER in zebrafish liver. Three
distinct ER isoforms were identified. Based on sequence
and domain comparisons and phylogenetic analyses, these
were termed ERoc, -Ba, and -Bb to conform to the zebrafish
nomenclature guidelines. Northern analysis revealed that
transcript numbers and sizes are gene- and tissue-specific.
As shown by reverse transcriptase PCR/Southern transfer
analysis, ERoc, -Ba, and -Bb were found to be widely ex-
pressed in neural and nonneural tissues, but relative levels
and ratios of the different subtypes differed by tissue type.
Although mRNAs encoded by all three ER genes were
detectable in unfertilized eggs— signifying maternal trans-
fer—and embryonic transcription had an early onset (12-
24 hour postfertilization, hpf), each had a distinct develop-
mental profile in subsequent stages (24-68 hpf) and was
differentially affected by exposure of embryos to estro-
gens, androgens, and known or suspected endocrine
disrupters (e.g., bisphenol-A, 2,3,7,8-tetrachlorodibenzo-
p-dioxin, polychlorinated biphenyls, atrazine). It was con-
cluded that reported endocrine-disrupting effects of chemi-
cals in laboratory species, and effects of pollutant mixtures
on reproduction and development of wildlife, could be due
in part to altered patterns of ER expression rather than di-
rect binding to existing ER. Because 0-48 hpf embryos
were used as a screening system in these initial studies,
further research is required to determine chemical effects
on ER at other life stages and in specific tissue types.
The Office of Research and Development's National Center for Environmental Research
113
-------
Endocrine Disruptors Program Review Workshop
Effects of Perfluorooctane Sulfonate on Thyroid Hormone
Status in the Rat and Mouse
Julie Thibodeaux, Roger Hanson, Brian Grey, John M. Rogers, and Christopher Lau
Reproductive Toxicology Division, National Health and Environmental Effects Research Laboratory, Office of
Research and Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
Perfluorooctane sulfonate (PFOS), a compound used
in the manufacture of surfactants and insecticides, is a
ubiquitous environmental contaminant that recently has
been found to be a developmental toxicant in laboratory
rodents. The focus of this work is to ascertain if alterations
of thyroid status are involved in PFOS toxicity. Timed-
pregnant, Sprague-Dawley rats were given 1, 2, 3, or 5
mg/kg/day PFOS/K+ by oral gavage from gestational day
(GD) 2-21; controls received 0.5 percent Tween-20 ve-
hicle (1 mL/kg). Blood samples were collected from the
dams on GD 7, 14, and 21 and from the pups throughout
postnatal development. Serum thyroid-stimulating hormone
(TSH), total and free T4, and T3 levels were determined by
radioimmunoassay at all doses in the dams and up to the
3 mg/kg dose in the offspring. Total and free T4 as well as
T3 levels in PFOS-treated dams were significantly lower at
all time points in comparison to controls, but the TSH levels
were not affected. Hypothyroxinemia was observed in the
PFOS-exposed pups during postnatal development. De-
creased rates of body weight gain in the dams during gesta-
tion and the offspring during postnatal development also
resulted from this exposure. Because thyroid hormone lev-
els are known to fluctuate with pregnancy, this study was
extended to nonpregnant adult rats. Ninety day-old male
and female rats received either vehicle control or 5 mg/kg
PFOS daily, and were sacrificed after 3 or 20 days of chemi-
cal exposures. Trunk blood was collected, and serum thy-
roid hormone levels were determined.
As seen in the pregnant cohort, the results indicated a
reduction of T4 in the PFOS-treated rats (45% or 15% of
controls after 3 or 20 days of exposure, respectively) with
an absence of feedback elevation of TSH. In fact, the
TSH levels of the nonpregnant rats appeared to be lower
in the 5 mg/kg group than in controls, at both time points.
Additionally, under similar exposure conditions, thyrotoxic
effects of PFOS also were seen in the pregnant mouse,
although the extent of this hormonal imbalance is less than
the rat. Thus, these results suggest that PFOS can be
regarded as an environmental endocrine disrupter. How-
ever, the underlying mechanisms for its particular profile
of hormonal disruption and the potential alterations of
physiological functions related to thyroid imbalance war-
rant further investigation.
114
The Office of Research and Development's National Center for Environmental Research
-------
Endocrine Disrupters Program Review Workshop
Anti-Androgenic Effects of Vinclozolin on Male Rats Are Partially
Attenuated by Testosterone Propionate
Cynthia Wolf1'2, Joe Ostby1, Jonathan Furr1, Gerald A. LeBlanc2, and L. Earl Gray, Jr.1
Reproductive Toxicology Division, National Health and Environmental Effects Research Laboratory,
U.S. Environmental Protection Agency, Research Triangle Park, NC; Department of Environmental
and Molecular Toxicology, North Carolina State University, Raleigh, NC
Sexual differentiation to the male phenotype is de-
pendent on activation of the androgen receptor (AR) by
androgens during fetal development. The AR inhibitor
vinclozolin (V) administered to the rat dam on gesta-
tional days (GD) 14-19 compromises masculine devel-
opment of the male offspring, resulting in nipple forma-
tion, cleft phallus, ectopic testes, and reduced prostate,
seminal vesicle, and levator ani/bulbocavernosus (LA/
BC) weights. Testosterone propionate (TP) administered
to the dam during the same gestational period does not
affect the male offspring, but masculinizes female off-
spring. Because both androgenic and anti-androgenic
chemicals exist in the environment and the effects of
combinations of chemicals are unknown, the study of
combinational exposure to these chemicals is important
for risk assessment.
In the current study, this research project sought to
determine whether co-administration of TP with V at-
tenuates the action of V on sexual development of the
male. Sprague-Dawley rats were dosed on GD 14-19
with corn oil (vehicle; 2.5 mL/kg; oral gavage), TP (1
mg/0.1 iL/rat; subcutaneous), V (200 mg/kg; oral gav-
age), or V+TP Male offspring were monitored through-
out life and necropsied on postnatal day (PND) 170-
186. Litter size on PND 2 was reduced significantly only
by V+TP (5.6; p < 0.001), although sex ratio was not
affected in any treatment group. Consistent with previ-
ous results, V reduced anogenital distance (AGO) on
PND 2 (p < 0.0001); induced nipples (mean n = 11.68);
induced malformations such as cleft phallus (95%), vagi-
nal pouch (84%), and ectopic testes (59%); and reduced
prostate, seminal vesicle, and LA/BC weights in male
offspring. TP alone had no effect on any of these end-
points. Co-administration of TP with V significantly re-
duced the number of nipples induced by V (n = 9.50; p <
0.005 compared to V) and reduced the incidence of mal-
formations induced by V (cleft phallus = 75%, p < 0.01;
vaginal pouch = 48%, p < 0.05; ectopic testes = 12.5%,
p < 0.01 compared to V). Conversely, V+TP failed to
reverse the reduction in AGO or restore weights of ven-
tral prostate, seminal vesicle, and LA/BC reduced by V
It was concluded that co-administration of TP with V at
the doses used attenuates some of the anti-androgenic
effects induced by V during fetal development in male
rat offspring. Future work may investigate the differ-
ences in protein expression in these tissues that may be
responsible for the difference in response.
The Office of Research and Development's National Center for Environmental Research
115
-------
Endocrine Disrupters Program Review Workshop
Index of First Authors
Ai,N.,65
Anderson, H.A., 85
Angus, R.A., 9
Ankley, G.T., 11
Baker, D., 86
Benninghoff, A.D., 99
Brown, K.W., 66
Callard,G.V.,51
Cardon, M.C., 67
Chandler, G.T., 12, 68
Cooper, R.L., 52
Crofton, K.M., 23
Eskenazi, B., 87
Euling, S.Y., 35
Fenton, S.E., 24
Fertuck, K.C., 13
Fielden, M.R., 53
Fry, D.M., 25
Gardiner, D.M., 43
Gilbert, M, 79
Goldman, J.M., 100
Golub, M.S., 26
Gooding, M.P., 101
Gray, Jr., L.E., 14
Grim, F, 102
Hawkins, M.B., 104
Hedge, J.M., 105
Hemmer, M.J., 54
Highsmith,V.R.,36
Hinton, D.E., 15
Holt,V.L.,88
Hughes, C, 27
Kholodovych, V., 69
Lambright, C.R., 70
Lasley, B.L., 55
Lattier, D.L., 37
Laws, S.C., 16
LeBlanc, GA., 56
Manire, C.A., 44
Marcus, M., 28
McClellan-Green, P.O., 46, 106
McKenney, Jr., C.L., 17
McNabb, P.M., 47
Mills, L.J., 48
Mills, M.A., 38
Moran, P.M., 108
Mu, X., 109
Mukerjee, D., 39
Narotsky, M.G, 29
Ndayibagira,A., 80
Newbold, R.R., 81
Oberdorster, E., 110
01mstead,A.W., 112
Osteen, K.G, 57
Ottinger, M.A., 30
Owens, C., 71
Roberts, M.A., 89
Rothberg, S., 113
Sayles, GD., 40
Schwartz, S.M., 90
Sergeyev, 0., 91
Skinner, M.K., 31
Swan, S.H., 92
Thibodeaux, J., 114
Thomas, P., 58
Tietge, J.E., 18
Timm, GE., 5
Veeramachaneni, D.N., 32
Volz, D.C., 72, 73
Walse, S.S., 93, 94
Welsh, W.J., 19
Wilson, V.S., 59
Windham, G, 95
Wirth, E.P, 74
Wolf, C, 115
Yu, S.J., 75
The Office of Research and Development's National Center for Environmental Research
117
-------
wEPA
United States
Environmental Protection Agency
Office of Research and Development
National Center for Environmental Research (8701R)
Washington, DC 20460
EPA/600/R-02/065
October 2002
www.epa.gov/ncer
Official Business Only
Penalty for Private Use
$300
------- |