Uncertainties for Endocrine Disrupters: Our View on Progress

George P. Daston*, Jon C. Cook{dagger} and Robert J. Kavlock{ddagger},1

* Miami Valley Laboratories, The Procter & Gamble Company, P.O. Box 538707, Cincinnati, Ohio 45253; {dagger} Pfizer Global Research and Development, Eastern Point Road, Groton, Connecticut 06340; and {ddagger} National Health Environmental Effects Research Laboratory, Office of Research and Development, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711

Received February 14, 2003; accepted March 31, 2003


    ABSTRACT
 TOP
 ABSTRACT
 Question 1: What Effects...
 Question 2: What Are...
 Question 3: What Are...
 Question 4: Do Our...
 Question 5: What Extrapolation...
 Question 6: What Are...
 Question 7: How and...
 Question 8: What Are...
 Question 9: How Can...
 Conclusion
 REFERENCES
 
The hypothesis that hormonally active compounds in the environment—endocrine disrupters—are having a significant impact on human and ecological health has captured the public’s attention like no other toxicity concern since the publication of Rachel Carson’s Silent Spring 1962. In the early 1990s, Theo Colborn and others began to synthesize information about the potential impacts of endocrine-mediated toxicity in the scientific literature (Colborn and Clement, 1992Go) and the popular press (Colborn et al., 1997Go). Recognizing the possibility of an emerging health threat, the U.S. Environmental Protection Agency (EPA) convened two international workshops in 1995 (Ankley et al., 1997Go; Kavlock et al., 1996Go) that identified research needs relative to future risk assessments for endocrine-disrupting chemicals (EDCs). These workshops identified effects on reproductive, neurological, and immunological function, as well as carcinogenesis as the major endpoints of concern and made a number of recommendations for research. Subsequently, the EPA developed a research strategy to begin addressing the recommendations (EPA, 1998aGo), and the federal government as a whole, working through the White House’s Committee on the Environment and Natural Resources, increased funding levels and coordinated research programs to fill the major data gaps (Reiter et al., 1998Go).

In parallel with these research efforts that were attempting to define the scope and nature of the endocrine disruptor hypothesis, the U.S. Congress added provisions to the Food Quality Protection Act (FQPA) and the Safe Drinking Water Act of 1996 to require the testing of food-use pesticides and drinking water contaminants, respectively, for estrogenicity and other hormonal activity. These bills were enacted into law, giving the EPA the mandate to implement them. The EPA, with the help of an external advisory committee, the Endocrine Disruptor Screening and Testing Advisory Committee (EDSTAC), determined that other hormonal activity should include androgens and compounds that affect thyroid function, and expanded the mandate to include all chemicals under EPA’s jurisdiction, potentially including the 70,000 chemicals regulated under the Toxic Substances Control Act (Endocrine Disruptor Screening and Testing Advisory Committee [EDSTAC], 1998Go). EDSTAC recommended an extensive process of prioritization, screening, and testing of chemicals for endocrine-disrupting activity, including a screening battery that involves a combination of at least eight in vitro and in vivo assays spanning a number of taxa (EDSTAC, 1998Go). What started out as a hypothesis has become one of the biggest testing programs conceived in the history of toxicology and the only one that has ever been based on mechanism of action as its premise.

As we pass the 10th anniversary of the emergence of the endocrine disruptor hypothesis, it is useful to look back on the progress that has been made in answering the nine questions posed as data gaps in the EPA’s research strategy (EPA, 1998aGo)—not only to see what we have learned, but also to examine whether the questions are still appropriate for the goal, what gaps remain, and what directions should be emphasized in the future.


    Question 1: What Effects Are Occurring in Populations?
 TOP
 ABSTRACT
 Question 1: What Effects...
 Question 2: What Are...
 Question 3: What Are...
 Question 4: Do Our...
 Question 5: What Extrapolation...
 Question 6: What Are...
 Question 7: How and...
 Question 8: What Are...
 Question 9: How Can...
 Conclusion
 REFERENCES
 
Undoubtedly, the most important question to answer is whether indeed EDCs are having effects in either humans or wildlife populations. The core of the hypothesis—that there are compounds that have the potential to affect hormonal status adversely, leading to abnormal development, reproductive dysfunction, and some cancers—is not in doubt. There is a wealth of laboratory-based studies that delineate the range of effects that can be induced following exposures at various life stages. But for natural populations, the situation is not so obvious. As noted by the recent state-of-the-science assessment by the World Health Organization (International Programme on Chemical Safety [IPCS], 2002Go), the clearest available evidence is derived from studies in wildlife, including reproductive and immunological effects in marine mammals living in environments contaminated by organochlorines (PCBs, DDE); eggshell thinning and embryonic abnormalities in birds exposed to DDT and PCBs, respectively; induction of the estrogen biomarker vitellogenin in fish living near a number of effluent point sources from sewage treatment plants or pulp mills; and perhaps most convincingly, imposex in marine molluscs exposed to tributyltin. Nevertheless, there remains uncertainty about cause-and-effect relationships on a broader population basis, and there remains a need to better document the relationship between ambient exposure levels and effects on wildlife at the population level and the impact of pollution relative to other stressors, such as habitat destruction.

In humans, the evidence for direct cause-and-effect relationships from environmental exposures remains generally lacking, with the effects of PCBs on neurological development among the strongest possibilities (IPCS, 2002Go). The diethylstilbestrol (DES) case is clearer, but here the exposure was to pharmacological levels given to prevent miscarriage. DES produced reproductive tract malformations in sons and daughters and vaginal tumors in daughters who were prenatally exposed (for a review, see Mittendorf, 1995Go). For perspective, most other agents that have been implicated as endocrine disrupters are much less potent than DES and are present at lower levels. What is at issue, then, is the notion that, by inference, much lower exposures to agents with weaker activity are responsible for widespread effects on development, reproduction, and the production of certain types of cancer. This has been the central question of a number of reviews, including the National Academy of Sciences report, Hormonally Active Agents in the Environment (National Research Council, 1999Go). The Academy’s panel was split on the general hypothesis. The reasons for the lack of consensus are briefly discussed in the Executive Summary of the report; principal among them were the significant limitations and uncertainty in the database on observed effects and their potential etiologies.

There has been considerable retrospective analysis and reanalysis of secular trend data for cancers and birth defects that may have a hormonal etiology (e.g., Jorgensen et al., 2002Go; Paulozzi et al., 1997Go), as well as for sperm counts (e.g., Fisch et al., 1996Go; Swan et al., 1997Go). Unfortunately, too little of the secular trend data has been critically pursued with etiological studies. There are many alternative explanations for the trends that are at least as plausible as endocrine disrupters in the environment; these include improvements in medical surveillance (e.g., for breast cancer) and changing medical practice (e.g., a possible association between increased use of progesterone in the first trimester to forestall miscarriage and hypospadias [Silver, 2000Go]).

Clearly, then, one of the biggest unanswered questions is that of the magnitude of the endocrine disrupter problem. This question can be answered only by epidemiology and ecoepidemiology approaches that fully conform to criteria controlling for bias and that are aimed at determining causation. Such studies will be time-consuming and expensive but are necessary if we are to determine whether the secular trend data for diseases such as breast cancer, testicular cancer (particularly in Denmark), hypospadias (in the U.S.), and others are attributable, in whole or in part, to endocrine disrupters. It is worth noting that some critical analyses have been done, and these will be helpful in addressing the question. In the last decade, a number of studies have evaluated the levels of environmental agents in banked tissue samples and compared them with outcomes such as breast cancer, concluding that there was no association; these studies will provide a useful start for answering this question (Adami et al., 1995Go; Krieger et al., 1994Go). But even here, it is not clear what the critical life stage is that should be targeted for the exposure assessment (Birnbaum and Fenton, 2003Go). Continuing follow-up of DES daughters as they enter the stage of life during which breast cancer incidence increases will also be useful in addressing the question of whether prenatal exposure is a prerequisite for increased risk (Hatch et al., 1998Go).


    Question 2: What Are the Chemical Classes and Their Potencies?
 TOP
 ABSTRACT
 Question 1: What Effects...
 Question 2: What Are...
 Question 3: What Are...
 Question 4: Do Our...
 Question 5: What Extrapolation...
 Question 6: What Are...
 Question 7: How and...
 Question 8: What Are...
 Question 9: How Can...
 Conclusion
 REFERENCES
 
To date, most of the published in vivo efforts to identify endocrine-disrupting potential have focused on a relatively small subset of chemicals, including persistent organochlorines (primarily DDT, PCBs, and bisphenol A [BPA]), methoxychlor, alkylphenols, phthalates, anti-androgenic fungicides, chlorotriazine herbicides, phytoestrogens, and a few pharmaceutical agents. Given the impetus of the FQPA, EDSTAC, and the current EDMVS (Endocrine Disruptors Methods Validation Subcommittee—see http://www.epa.gov/scipoly/oscpendo/index.htm), there have been considerable efforts placed on developing, standardizing, and validating screening tests for the identification of chemicals with endocrine-disrupting potential (e.g., Goldman et al., 2000Go; Gray et al., 2002Go; Kanno et al., 2001Go; O’Connor et al., 2002Go; Stoker et al., 2000Go) that can be applied to a wider set of chemicals.

The EDSTAC recommendations for screening were comprehensive in scope: Evaluate all known mechanisms of action for compounds that interfere with estrogen, androgen, and thyroid hormone function and that are representative of a number of vertebrate taxa (EDSTAC, 1998Go). Known mechanisms are not limited to receptor agonism or antagonism but also include such activities as inhibition of enzymes of hormone synthesis (e.g., via alterations in 5-{alpha}-reductase or aromatase). The recommended screening program is ambitious, entailing at least eight different assays that span the gamut from receptor binding assays to more complex whole animal studies in rats, frogs, and fish.

A recent review on the three EDSTAC options discusses the pros/cons of the individual assays based on the data generated since the EDSTAC report was issued (O’Connor et al., 2002Go). The EDSTAC screening program is time- and resource-consuming, with estimates of the cost of running the battery for a single compound in the range of $200–$300,000 (EDSTAC, 1998Go). Critical decisions remain as to which of the numerous assays will actually be incorporated into the screening battery after factoring in considerations such as the extent of redundancy needed, the efficiency in the use of animals, the sensitivity and specificity of the assays, and need to cover multiple taxa. The generation of screening data for large numbers of compounds will ultimately lead to establishment of triggers for Tier 2 testing, using multigenerational protocols.

A concern with the EDSTAC screening proposal is the fact that prenatal development is not represented, despite the fact that it is generally believed that the earliest life stages are the most sensitive to endocrine disruption. Therefore, it is possible that the screening program could generate a rate of false negatives that would be unacceptable. The main question is whether, indeed, these life stages represent critically unique periods of sensitivity or whether they merely represent quantitatively more sensitive stages. The problem with including prenatal exposure in the screen is that the developmental effects of endocrine disrupters tend to be latent; traditional end points of toxicity (i.e., altered structure or function) may not be detectable until sexual maturity. Because sexual maturity is not reached in lab rodent species until 8–10 weeks after birth, it has been considered impractical to evaluate the consequences of prenatal treatment in a screening-level assay. Pubertal male and female assays may address this point in part, but most development of the reproductive system has already occurred by the life stage at which dosing is commenced in these protocols. Although traditional end points of toxicity (e.g., structural malformations, diminished reproductive function) may be latent, it is highly likely that they are preceded by immediate and persistent changes in gene expression. It is now well established that the signal transduction pathway for steroid hormones involves changes in gene expression as an integral step; therefore, it should be possible to use gene expression changes in the fetus as a means of detecting endocrine-disrupting potential in the most relevant life stage.

Genomic tools that have become available over the past few years make it feasible to survey tissues for changes in gene expression. Recent work has demonstrated the validity of this approach for endocrine-disrupter screening: Estrogens of varying potencies have been shown to produce a characteristic transcript profile in the fetal reproductive tract of rats transplacentally exposed to these compounds (Naciff et al., 2002Go). It may be possible to use these transcript profiles to develop an alternative screening assay that is more sensitive than and obviates the need for many of the assays now being considered as part of the EPA battery. Similar activities are also underway for evaluation of effects in wildlife (e.g., Larkin et al., 2002Go). Of course, such approaches will need to go through the same standardization and validation process now being followed for the existing proposed screens.

Another aspect of endocrine-disrupter screening that has yet to be addressed is the prioritization of compounds for screening and testing. EDSTAC recommended that EPA extend its screening program beyond the legally mandated testing of food-use pesticides and drinking water contaminants to all the chemicals under its regulatory purview. This includes the inventory of more than 70,000 chemicals regulated under the Toxic Substances Control Act—a daunting number of chemicals for which to contemplate any sort of testing. There are some exclusion criteria by which some chemicals will be exempted from testing (most prominently, polymers greater than 1000 m.w.); however, this still leaves many thousands of chemicals in the pipeline for screening. To make good decisions as to which chemicals should be evaluated, tools for prioritization will need to be developed. These will include quantitative structure-activity relationship (QSAR) programs that predict receptor binding and high-throughput in vitro assays that evaluate receptor binding. There have been efforts to develop predictive QSAR programs (e.g., Mekenya et al., 2002Go; Serafimova et al., 2002Go; Shi et al., 2001Go; Suzuki et al., 2001Go) but the validation and refining of these programs is rate-limited by the available amount of receptor binding data covering a breadth of chemical structures. Indeed, in the recent proposal for identifying the process by which the first 200 chemicals will enter into the EDSTAC screening battery, a number of measures of exposure potential for pesticides was chosen over QSAR because of the lack of validated methodologies (Federal Register, 2002Go). High-throughput assays to evaluate receptor-binding activity are technically feasible; in fact, they are now a routine aspect of drug discovery programs in pharmaceutical companies. The challenge for endocrine-disrupter screening will be to adapt these approaches, which are optimized for detecting compounds with high potency, so that they can reliably detect substances with substantially less activity than the prototypical ligands. Even rather simple technical issues, such as compound solubilization, become a substantial obstacle due to the diverse chemical classes that need to be evaluated and will require novel approaches.


    Question 3: What Are the Dose-Response Characteristics in the Low-Dose Region?
 TOP
 ABSTRACT
 Question 1: What Effects...
 Question 2: What Are...
 Question 3: What Are...
 Question 4: Do Our...
 Question 5: What Extrapolation...
 Question 6: What Are...
 Question 7: How and...
 Question 8: What Are...
 Question 9: How Can...
 Conclusion
 REFERENCES
 
Most of the evidence from population and even toxicological studies have identified effects of endocrine-disrupting chemicals at levels in excess of those encountered in the general environment, and even here our knowledge base is fairly limited. Therefore, an important unanswered question is that of the nature of the dose-response curve at low-level exposures. There have been some reports in the literature indicating that developmental exposure to very low levels of some estrogens produces a nonmonotonic, inverted U-shaped dose-response curve; this has been reported for prostate weight in mice prenatally exposed to BPA, DES, or estradiol (Nagel et al., 1997Go; vom Saal et al., 1997Go). Prostate weight was increased at very low doses, then decreased with increasing dose. Others have tried and failed to repeat these experiments, using studies that duplicate the original experiments as much as possible but with more animals to achieve greater statistical resolving power (Ashby et al., 1999Go; Cagen et al., 1999aGo). Still others have examined similar end points in other laboratory species without evidence of nonmonotonicity (e.g., Cagen et al., 1999bGo; Kwon et al., 2000Go; Tyl, 2002). Another weakly estrogenic compound, nonylphenol, also failed to produce nonmonotonic dose responses when assessed in a robust study design (Chapin et al., 1999Go). An independent panel convened by the U.S. National Institute of Environmental Health Sciences (NIEHS) and the EPA concluded that "there is credible evidence that low doses of BPA can cause effects on specific end points. However, due to the inability of other credible studies in several different laboratories to observe low-dose effects of BPA and the consistency of these negative studies, the subpanel is not persuaded that a low-dose effect of BPA has been conclusively established as a general or reproducible finding." Furthermore, they stated, "Data are insufficient to establish the shape of the dose-response curve for bisphenol A in the low-dose region, and the mechanism and biologic relevance of reported low-dose effects are unclear" (Melnick et al., 2002Go). The panel remarked that any of a number of small discrepancies between the study designs may have been responsible for the difference; therefore, there was no basis for categorically rejecting one result or the other.

Additional research will be necessary to resolve this issue of U-shaped dose response curves. Because we don’t know which factors are the most important determinants of the development of the morphological end points measured in these studies, it is unlikely that simply conducting more studies using the same end points will do anything to clarify the muddle. Instead, future work will need to concentrate on characterizing the mechanism(s) by which an estrogen (or anti-androgen, for that matter) could produce one developmental effect at a low dose and something entirely different at a high dose. Such work might include an evaluation of whether the underlying gene expression induced by the estrogens is qualitatively different at high versus low doses.

If these nonmonotonic dose-response curves can be verified, it will also be important to determine whether the effects at low doses have any long-term adverse health consequences. Putz and co-workers (2001a,b) reported a nonmonotonic dose-response curve for weights of prostate, testis, and epididymidis in rats neonatally exposed to estradiol, but the low-dose effect was transitory. They surmised that the effect was attributable to acceleration of puberty; organ weights were comparable to controls on maturation. Accelerated puberty would be an important effect; whether this is related to a lasting health consequence (e.g., predisposition to prostatic hyperplasia) is an equally important question. One of the major limitations in assessing a role in prostate cancer is that animal models that are predictive for human prostate cancer are not available.


    Question 4: Do Our Testing Guidelines Adequately Evaluate Potential Endocrine-Mediated Effects?
 TOP
 ABSTRACT
 Question 1: What Effects...
 Question 2: What Are...
 Question 3: What Are...
 Question 4: Do Our...
 Question 5: What Extrapolation...
 Question 6: What Are...
 Question 7: How and...
 Question 8: What Are...
 Question 9: How Can...
 Conclusion
 REFERENCES
 
The types of end points included in newer testing guidelines for reproductive toxicity (EPA, 1998bGo) have clearly provided better coverage of endocrine-sensitive systems, particularly for agents that directly or indirectly alter steroid hormone responsivity. For example, recent experience with some phthalates have demonstrated that these modifications are capable of detecting effects at lower levels of exposure than had been the case with previous protocols (Mylchreest et al., 2000Go). There is still concern that other hormonal systems (e.g., the thyroid) are not adequately represented. Additionally, there is a general absence of comparable multigenerational protocols for avian, fish, and amphibian species. Both of these gaps are being addressed in work underway in support of the EDMVS process (see EDMVS Web site previously mentioned).

Of perhaps greater consequence for the testing guidelines are the implications of nonmonotonic dose responses, as mentioned earlier, to the practice of risk assessment. Current risk assessment practices involve the application of uncertainty factors to a NOAEL (No Observed Adverse Effect Level) to arrive at an exposure level with minimal risk. However, the NOAEL is derived from study designs that assume that the dose response is monotonic. If there are effects below the presumed NOAEL, these should be considered in setting acceptable exposure levels and would necessitate the redesign of safety studies to include a much wider range of dose levels. This would require increasing the magnitude of toxicity studies, particularly the number of animals used, as well as different statistical approaches to analyze data. At present, the EPA has taken the interim policy position that, pending further research on the relevant mechanisms of toxicity, it would be premature to require testing of low-dose effects in the Endocrine Disruptor Screening Program (see http://www.epa.gov/scipoly/oscpendo/low_dose_statement.pdf).


    Question 5: What Extrapolation Tools Are Needed?
 TOP
 ABSTRACT
 Question 1: What Effects...
 Question 2: What Are...
 Question 3: What Are...
 Question 4: Do Our...
 Question 5: What Extrapolation...
 Question 6: What Are...
 Question 7: How and...
 Question 8: What Are...
 Question 9: How Can...
 Conclusion
 REFERENCES
 
There are a number of aspects pertinent to extrapolation of EDC data that are similar to those encountered with other chemicals. The first pertains to the extrapolation of effects seen at high exposure levels to those of more environmental concern and is directly related to the issue of nonmonotonic dose responses. To date, much of the work on this issue has been empirically based and has generated considerable controversy, as noted in the discussion of Question 3. Additional research on the biological steps intervening between administered dose and tissue response are needed to provide a stronger mechanistic basis for findings in some test systems: These might include a better understanding of target tissue dosimetry and the quantitative relationships between changes at the molecular level (e.g., gene expression) and higher order responses. A second aspect of extrapolation where work is needed relates to the predictive nature of biomarkers—do measures such as vitellogenin represent only biomarkers of exposure or are they also biomarkers of effect that indicate an adverse response? Although some progress has been made in understanding the predictiveness of vitellogenin induction across levels of biological organization (e.g., Hutchinson and Pickford, 2002Go; Matthiessen et al., 2002Go; Mills et al., 2003Go; Robinson et al., 2003Go), more work is clearly needed. A third aspect of extrapolation needs is the result of the rather unusual blending of concern for wildlife and humans relevant to exposures to EDCs. Here there is a need to understand the relevance of effects observed in wildlife for humans, and vice versa. Given the conserved nature of steroid hormone systems in vertebrates, there has been a certain expectancy that effects could extrapolate fairly easily across species, and that has been borne out on a qualitative level for a number of estrogenic chemicals. However, molecular evidence suggests that, at least in terms of ligand-receptor interactions, we should expect quantitative differences in binding and perhaps even species-specific steroid receptor ligands to emerge, due to the differences in primary amino acid sequences (Harris et al., 2002Go; Matthews and Zacharewski, 2000Go; Matthews et al., 2000Go, 2002Go). Along these same lines are the difficulties in extrapolating data obtained from in vitro assays to effects observed in vivo and for the extrapolation of evidence of endocrine activity in screening assays to the ability to induce adverse effects in more traditional testing protocols. The evidence generated to date suggests that in vitro assays can generate both false positive and false negative results, but our experience has not been sufficient to begin to calculate the predictiveness of any of the assays.


    Question 6: What Are the Effects of Exposure to Multiple EDCs, and Will a Toxic Equivalency Factor (TEF) Approach Be Feasible?
 TOP
 ABSTRACT
 Question 1: What Effects...
 Question 2: What Are...
 Question 3: What Are...
 Question 4: Do Our...
 Question 5: What Extrapolation...
 Question 6: What Are...
 Question 7: How and...
 Question 8: What Are...
 Question 9: How Can...
 Conclusion
 REFERENCES
 
Because exposure to EDCs is rarely, if ever, to a single chemical, concern has arisen about the effects of exposure to multiple chemicals working through a common mechanism. For another large class of chemicals, those that bind to the Ah receptor and exert their toxicity through downstream events, the TEF has been developed to evaluate the potency of mixtures of known AhR ligands (van den Berg et al., 2000Go). A logical question would be whether this approach could be applied to estrogens, anti-androgens, or other EDC-mediated modes of action. Although recently there has been some rigorous efforts to explore the underlying hypothesis of additivity for estrogenic EDCs (e.g., Charles et al., 2002aGo,bGo), there has yet to be a concerted effort to develop a database from which relative potencies could be calculated. Indeed, given the complexity of the endocrine signaling systems, the multiple health outcomes possible, the differences in response across species, and the need to demonstrate parallel dose-response relationships across end points, it is unlikely that such a goal could be reached. Nevertheless, the issue of cumulative risk for EDCs will need to be addressed, at least for food-use pesticides, as mandated by the FQPA.


    Question 7: How and to What Degree Are Human and Wildlife Populations Exposed to EDCs?
 TOP
 ABSTRACT
 Question 1: What Effects...
 Question 2: What Are...
 Question 3: What Are...
 Question 4: Do Our...
 Question 5: What Extrapolation...
 Question 6: What Are...
 Question 7: How and...
 Question 8: What Are...
 Question 9: How Can...
 Conclusion
 REFERENCES
 
Exposure assessment has been the Achilles heel of the endocrine disruptor issue. Although there have been considerable efforts directed at measuring the levels of persistent organochlorine pollutants in a wide variety of species and geographical areas, much less attention has been paid to the nonpersistent pollutants. This task is made all the more difficult because the exposure assessment has to be done in the context of the sensitive life stage that could be impacted. The addition of potential endocrine disrupters to the list of agents being monitored for the National Human Exposure Report is one attempt to close this gap, and it has been successful at pointing to higher levels of phthalate exposure in women of reproductive age than had previously been suspected (see http://www.cdc.gov/nceh/dls/report/PDF/CompleteReport.pdf). However, these data can be used only to infer exposure at the critical periods of development that are susceptible to endocrine-active agents, not to measure it directly.

The presence and potential hazards of pharmaceuticals in the aquatic environment is a relatively new issue and has recently received increased attention from regulatory and scientific communities, due to a growing number of peer-reviewed papers reporting measurable levels of pharmaceuticals in water (e.g., municipal effluent and surface, ground, and drinking water). For instance, the U. S. Geological Survey detected 82 pharmaceutical and personal care products in surface waters of the U. S. (Kolpin et al., 2002Go). Even though a drug’s affinity to its target molecule is intentionally very high, the measurable levels of pharmaceuticals are generally quite low; therefore, exposure to pharmaceuticals in the water is unlikely to produce a significant acute toxicity hazard in nonmammalian species (Webb, 2001Go). Rather, the concern has focused on the potential for chronic, developmental, or reproductive effects to occur in aquatic organisms due to the biological activity of pharmaceuticals. The concern has been raised that these risks are currently not adequately assessed by current test methodologies and data evaluation strategies used for pharmaceuticals. To address this concern, it has been suggested that mammalian pharmacology/toxicology data, clinical efficacious levels in humans, and environmental concentrations can be used to prioritize those pharmaceuticals that may require additional aquatic toxicity evaluation (Huggett et al., 2003Go). The premise of this model is that many of the enzyme and receptor targets in mammalian species have significant sequence homology in nonmammalian species (e.g., Andersen et al., 2000Go; Menuet et al., 2002Go; Todo et al., 2000Go). The significance of these findings will continue to be debated until sufficient research is conducted to address the magnitude of this issue. The issue of pharmaceuticals in the environmentillustrates a key need in assessing the magnitude of the endocrine disrupter problem, namely, the prospective monitoring of aquatic species within our waterways to detect population trends has been limited. Expanded routine monitoring is a research need. Perhaps the most useful approach would be to develop a safety assessment study, possibly using fish, that could identify target organ toxicity by utilizing a histopathological examination similar to the screening approaches used in mammalian systems.


    Question 8: What Are the Major Sources and Environmental Fates of EDCs?
 TOP
 ABSTRACT
 Question 1: What Effects...
 Question 2: What Are...
 Question 3: What Are...
 Question 4: Do Our...
 Question 5: What Extrapolation...
 Question 6: What Are...
 Question 7: How and...
 Question 8: What Are...
 Question 9: How Can...
 Conclusion
 REFERENCES
 
The answer to this question is, of course, contingent on identifying the EDCs of concern and remains primarily an unanswered question. Work in the last decade has clearly implicated certain sewage treatment processes (Kirk et al., 2002Go) and pulp mill effluents (Durhan et al., 2002Go) as sources of EDCs in the aquatic environment and the use of TBT in marine paints as the cause of molluscan imposex. The use of some fungicides (e.g., vinclozolin and linuron) and herbicides (chlorotriazines) also represent avenues of environmental exposure. The human diet is also a potential source of EDCs, especially for those who consume products rich in phytoestrogens. Some biomonitoring has pointed to potentially higher exposures to some phthalates than had been presumed previously, but the source of those exposures, other than generally from their use in plastics or as a constituent of some solvents, remains unidentified (Blount et al., 2000Go). There are also emerging studies on the possibility that the use of pharmaceutical agents by the livestock industry may be contributing to the environmental loads of EDCs (e.g., Wilson et al., 2002Go).


    Question 9: How Can Unreasonable Risks Be Managed?
 TOP
 ABSTRACT
 Question 1: What Effects...
 Question 2: What Are...
 Question 3: What Are...
 Question 4: Do Our...
 Question 5: What Extrapolation...
 Question 6: What Are...
 Question 7: How and...
 Question 8: What Are...
 Question 9: How Can...
 Conclusion
 REFERENCES
 
Similar to the previous question, this is dependent on recognizing which of the potential EDCs are present at unacceptable levels. We must first know potency, target species sensitivity, and ranges of exposure, then take appropriate action to reduce environmental releases or otherwise mitigate environmental levels.

Selected High-Priority Research Needs
Our ability to answer each of the nine questions posed above is limited by the amount of information available. There are some clear research needs that, if addressed, would go a long way toward answering these questions. Here are several that we view as important:

  1. Resolve the Low-Dose Controversy. Answering this question will allow us to determine whether changes are needed in our toxicity testing and risk assessment approaches. Previous efforts to resolve the controversy have had limited success because the end points being measured are susceptible to influence by a number of experimental factors that are not easy to identify or control. A more fruitful course of action may be to determine whether there is mechanistic plausibility for qualitatively different effects of endocrine disrupters at extremely low dose levels.
  2. Develop a national exposure survey for exposure to prevalent endocrine-active agents, including life stages that may have the greatest susceptibility.
  3. Increase efforts for human health surveillance on a population level that focus on specific birth defects or other disease states that may have (at least in part) an endocrine etiology. This could be included in the planned National Children’s Study. Furthermore, creating a national birth defects registry would be a benefit not just to endocrine disrupter research but also for understanding trends for all birth defects.
  4. Expand ecological monitoring efforts to identify population-level effects, and coordinate this with exposure information. It may be possible to extend the EPA’s EMAP program to serve this function.
  5. Develop methods for cumulative risk assessment for endocrine disrupters.
  6. Continue the Endocrine Disrupter Screening Program validation work, but combine this with a continuing assessment of the quality and efficiency of the screening assays being validated, as well as continuing improvement of the screening battery by adoption of better, more predictive assessment technology.
  7. Improve the Endocrine Disrupter Screening Program’s prioritization process by developing a process that considers the potential to cause biological effects, rather than just exposure potential.


    Conclusion
 TOP
 ABSTRACT
 Question 1: What Effects...
 Question 2: What Are...
 Question 3: What Are...
 Question 4: Do Our...
 Question 5: What Extrapolation...
 Question 6: What Are...
 Question 7: How and...
 Question 8: What Are...
 Question 9: How Can...
 Conclusion
 REFERENCES
 
Although a good deal of information on the nature and extent of the endocrine disruptor hypothesis has been generated over the past decade, more focused research is still needed to bring clarity to the issue. The answers to the questions above are certainly influenced by the bias of the authors. However, we feel that they fairly represent the current state of the science on the key questions. As additional research is delivered on the questions, it will help determine the relative importance of endocrine disrupters as a public health concern, a question whose answer will determine how much of our toxicological testing and risk assessment resources need to be applied to this problem and the necessity to shift from others. The criteria for establishment of causality for health outcomes related to endocrine disruption developed by the IPCS (2002)Go will be a useful tool in this regard. In the end, the prioritization of problems and allocation of resources is guesswork; however, we will make better guesses if we have answered, even partially, the questions posed in this essay.


    NOTES
 
This document has been reviewed in accordance with the U.S. Environmental Protection Agency policy and approved for publication. Approval does not signify that the contents reflect the views of the Agency, nor does mention of trade names or commercial products constitute endorsement or recommendation for use.

1 To whom correspondence should be addressed. Back


    REFERENCES
 TOP
 ABSTRACT
 Question 1: What Effects...
 Question 2: What Are...
 Question 3: What Are...
 Question 4: Do Our...
 Question 5: What Extrapolation...
 Question 6: What Are...
 Question 7: How and...
 Question 8: What Are...
 Question 9: How Can...
 Conclusion
 REFERENCES
 
Adami, H. O., Lipworth, L., Titus-Ernstoff, L., Hsieh, C. C., Hanberg, A., Ahlborg, U., Baron, J., and Trichopoulos, D. (1995). Organochlorine compounds and estrogen-related cancers in women. Cancer Causes Controls 6, 551–566.[ISI][Medline]

Andersen, O., Eijsink, V. G. H., and Thomassen, M. (2000). Multiple variants of the peroxisome proliferator-activated receptor (PPAR) are expressed in the liver of the Atlantic salmon (Salmor salar). Gene 255, 411–418.[CrossRef][ISI][Medline]

Ankley, G. T., Johnson, R. D., Toth, G., Folman, L. C., Detenbeck, N. E., and Bradbury, S. P. (1997). Development of a research strategy for assessing the ecological risk of endocrine disruptors [Review]. Toxicolology 1, 71–106.[CrossRef]

Ashby, J., Tinwell, H., and Haseman, J. (1999). Lack of effects for low dose levels of bisphenol A and diethylstilbestrol on the prostate gland of CF1 mice exposed in utero. Regul. Toxicol. Pharmacol. 30, 156–166.[CrossRef][ISI][Medline]

Birnbaum, L. S. and Fenton, S. E. (2003). Cancer and developmental exposure to endocrine disruptors. Environ. Health Perspect. 111, 389–394. DOI:10.1289/ehp.5686; available at http://dx.doi.org/.[ISI][Medline]

Blount, B. C., Silva, M. J., Caudill, S. P., Needham, L. L., Pirkle, J. L., Sampson, E. J., Lucier, G. W., Jackson, R. J., and Brock, J. W. (2000). Levels of seven urinary phthalate metabolites in a human reference population. Environ. Health Perspect. 108, 979–982.

Cagen, S. Z., Waechter, J. M., Dimond, S. S., Breslin, W. J., Butala, J. H., Jekat, F. W., Joiner, R. L., Shiotsuka, R. N., Veenstra, G. E., and Harris, L. R. (1999a). Normal reproductive organ development in Wistar rats exposed to bisphenol A in the drinking water. Regul. Toxicol. Pharmacol. 30, 130–139.[CrossRef][ISI][Medline]

Cagen, S. Z., Waechter, J. M., Dimond, S. S., Breslin, W. J., Butala, J. H., Jekat, F. W., Joiner, R. L., Shiotsuka, R. N., Veenstra, G. E., and Harris, L. R. (1999b). Normal reproductive organ development in CF-1 mice following prenatal exposure to bisphenol A. Toxicol. Sci. 50, 36–44.[Abstract]

Carson, R. (1962). Silent Spring. Houghton Mifflin, New York.

Chapin, R. E., Delaney, J., Wang, Y., Lanning, L., Davis, B. D., Collins, B. J., Mintz, N., and Wolfe, G. (1999). The effects of 4-nonylphenol in rats: A multigeneration reproduction study. Toxicol. Sci. 52, 80–91.[Abstract]

Charles, G. D., Gennings, C., Zacharewski, T. R., Gollapudi, B. B., and Carney, E. W. (2002a). An approach for assessing estrogen receptor-mediated interactions in mixtures of three chemicals: A pilot study. Toxicol. Sci. 68(2), 349–360.[Abstract/Free Full Text]

Charles, G. D., Gennings, C., Zacharewski, T. R., Gollapudi, B. B., and Carney, E. W. (2002b). Assessment of interactions of diverse ternary mixtures in an estrogen receptor-alpha reporter assay. Toxicol. Appl. Pharmacol. 180(1), 11–21.[CrossRef][ISI][Medline]

Colborn, T., and Clement, C. (Eds.). (1992). Chemically-Induced Alterations in Sexual and Functional Development: The Wildlife/Human Connection. Princeton Scientific Publishing, Princeton, NJ.

Colborn, T., Dumanoski, D., and Myers, J. P. (1997). Our Stolen Future: Are We Threatening our Fertility, Intelligence and Survival? A Scientific Detective Story. Plume Books/Penguin USA, New York.

Durhan, E. J., Lambright, C., Wilson, V., Butterworth, B. C., Kuehl O. W., Orlando E. F., Guillette, Jr., L. J., Gray L. E., and Ankley, G. T. (2002). Evaluation of androstenedione as an androgenic component of river water downstream of a pulp and paper mill effluent. Environ. Toxicol. Chem. 21(9), 1973–1976.[CrossRef][ISI][Medline]

Endocrine Disrupter Screening and Testing Advisory Committee (EDSTAC). (1998). Endocrine Disrupter Screening and Testing Advisory Committee Final Report. http://www.epa.gov/scipoly/oscpendo/history/finalrpt.htm.

EPA (1998a). Research Plan for Endocrine Disruptors. EPA/600/R-98/087. Office of Research and Development, February 1998, Washington, DC.

EPA (1998b). Health Effects Test Guidelines. OPPTS 870.3800 Reproductive and Fertility Effects. EPA 712-C-98–208, August 1998, Washington DC. Available at http://www.epa.gov/opptsfrs/OPPTS_Harmonized/870_Health_Effects_Test_Guidelines/Series/870–3800.pdf.

Federal Register (2002). Endocrine Disruptor Screening Program, Proposed Chemical Selection Approach for Initial Round of Screening; Request for Comment. Federal Register, December 30, 2002 67:250 [Notices]: 79611–79629.

Fisch, H., Goluboff, E. T., Olson, J. H., Felshuh, J., Broder, S. J., and Barad, D. H. (1996). Semen analyses in 1,283 men from the United States over a 25-year period: No decline in quality. Fertil. Steril. 65, 1009–1014.[ISI][Medline]

Goldman, J. M., Laws, S. C., Balchak, S. K., Cooper, R. L., and Kavlock, R. J. (2000). Endocrine-disrupting chemicals: Prepubertal exposures and effects on sexual maturation and thyroid activity in the female rat. A focus on the EDSTAC recommendations. Crit. Rev. Toxicol. 30(2), 135–196.[ISI][Medline]

Gray, L. E., Ostby, J., Wilson, V., Lambright, C., Bobseine, K., Hartig, P., Hotchkiss, A., Wolf, C., Furr, J., Price, M., et al. (2002). Xenoendocrine disrupters-tiered screening and testing. Filling key data gaps. Toxicology 27, 181–182, 371–382.[CrossRef]

Harris, H. A., Bapat, A. R, Gonder, D. S., and Frail, D. E. (2002). The ligand binding profiles of estrogen receptors alpha and beta are species dependent. Steroids 67(5), 379–384.[CrossRef][ISI][Medline]

Hatch, E. E., Palmer, J. R., Titus-Ernstoff, L., Noller, K. L., Kaufman, R. H., Mittendorf, R., Robboy, S. J., Hyer, M., Cowan, C. M., Adam, E., et al. (1998). Cancer risk in women exposed to diethylstilbestrol in utero. JAMA 280, 630–634.[Abstract/Free Full Text]

Huggett, D. B., Cook, J. C., Ericson, J. F., and Williams, R. T. (2003). A theoretical model for utilizing mammalian pharmacology and safety data to prioritize potential impacts of human pharmaceuticals to fish. J. Human and Ecological Risk Assessment.

Hutchinson, T. H., and Pickford, D. B. (2002). Ecological risk assessment and testing for endocrine disruption in the aquatic environment. Toxicology 181182, 383–387.[CrossRef][ISI]

International Programme on Chemical Safety (IPCS). 2002. Global Assessment of the State-of-the-Science of Endocrine Disruptors. T. Damstra, S. Barlow, A. Bergman, R. Kavlock, and G. van der Kraak, Eds. World Health Organization, Geneva.

Jorgensen, N., Carlsen, E., Nermoen, I., Punab, M., Suominen, J., Andersen, A. G., Andersson, A. M., Haugen, T. B., Horte, A., Jensen, T. K., et al. (2002). East-West gradient in semen quality in the Nordic-Baltic area: A study of men from the general population in Denmark, Norway, Estonia and Finland. Human Reprod. 17, 2199–2208.[Abstract/Free Full Text]

Kanno, J., Onyon, L., Haseman, J., Fenner-Crisp, P., Ashby, J., and Owens, W. (2001). The OECD program to validate the rat uterotrophic bioassay to screen compounds for in vivo estrogenic responses: Phase 1. Environ. Health Perspect. 109, 785–794.[ISI][Medline]

Kavlock, R. J., Daston, G. P., DeRosa, C., Fenner-Crisp, P., Gray, L. E., Kaattari, S., Lucier, G., Luster, M., Mac, M. J., Maczka, C., et al. (1996). Research needs for the risk assessment of health and environmental effects of endocrine disruptors: A report of the U.S. EPA-sponsored workshop. Environ. Health Perspect. 104(Suppl. 4), 715–740.[ISI][Medline]

Kirk, L. A., Tyler, C. R., Lye, C. M., and Sumpter, J. P. (2002). Changes in estrogenic and androgenic activities at different stages of treatment in wastewater treatment works. Environ. Toxicol. Chem. 21(5), 972–979.[CrossRef][ISI][Medline]

Kolpin, D. W., Furlong, E. T., Thurman, E. M., Zaugg, S. D., Barber, L. B., and Buxton, H. T. (2002). Pharmaceuticals, hormones and other organic wastewater contaminants in U.S. streams, 1999–2000. A national reconnaissance. Environ. Sci. Tech. 36, 1202–1211.[CrossRef][ISI][Medline]

Krieger, N., Wolff, M.S., Hiatt, R.A., Rivera, M., Vogelman, J., and Orentreich, N. 1994. Breast cancer and serum organochlorines: A prospective study among white, black, and Asian women. J. Natl. Cancer Inst. 86, 589–599.[Abstract]

Kwon, S., Stedman, D. B., Elswick, B. A., Cattley, R. C., and Welsch, F. (2000). Pubertal development and reproductive functions of Crl:CD BR Sprague-Dawley rats exposed to bisphenol A during prenatal and postnatal development. Toxicol. Sci. 55(2), 399–406.[Abstract/Free Full Text]

Larkin, P., Folmar, L. C., Hemmer, M. J., Poston, A. J., Lee, H. S., and Denslow, N. D. (2002). Array technology as a tool to monitor exposure of fish to xenoestrogens. Mar. Environ. Res. 54(3–5), 395–399.[CrossRef][ISI][Medline]

Matthews, J., Celius, T., Halgren, R., and Zacharewski, T. (2000). Differential estrogen receptor binding of estrogenic substances: A species comparison. J. Steroid Biochem. Mol. Biol. 74(4), 223–234.[CrossRef][ISI][Medline]

Matthews, J., and Zacharewski, T. (2000). Differential binding affinities of PCBs, HO-PCBs, and aroclors with recombinant human, rainbow trout (Onchorhynkiss mykiss), and green anole (Anolis carolinensis) estrogen receptors, using a semi-high throughput competitive binding assay. Toxicol. Sci. 53(2), 326–339.[Abstract/Free Full Text]

Matthews, J. B., Fertuck, K. C., Celius, T., Huang, Y. W., Fong, C. J., and Zacharewski, T. R. (2002). Ability of structurally diverse natural products and synthetic chemicals to induce gene expression mediated by estrogen receptors from various species. J. Steroid Biochem. Mol. Biol. 82(2–3), 181–194.[CrossRef][ISI][Medline]

Matthiessen, P., Allen, Y., Bamber, S., Craft, J., Hurst, M., Hutchinson, T., Feist, S., Katsiadaki, I., Kirby, M., Robinson, C., et al. (2002). The impact of oestrogenic and androgenic contamination on marine organisms in the United Kingdom—summary of the EDMAR programme. Endocrine Disruption in the Marine Environment. Mar. Environ. Res. 54(3–5), 645–649.[CrossRef][ISI][Medline]

Mekenya, O., Kamenska, V., Serafimova, R., Poellinger, L., Brouwer, A., and Walker, J. (2002). Development and validation of an average mammalian estrogen receptor-based QSAR model. SAR QSAR Environ. Res. 13(6), 579–595.[CrossRef][ISI][Medline]

Melnick, R., Lucier, G., Wolfe, M., Hall, R., Stancel, G., Prins, G., Gallo, M., Reuhl, K., Ho, S. M., Brown, T., et al. (2002). Summary of the National Toxicology Program’s report of the endocrine disruptors low-dose peer review. Environ. Health Perspect. 110, 427–431.[Medline]

Menuet, A., Pellegrin, E., Anglade, I., Blaise, O., Laudet, V., Kah, O., and Pakdel, F. (2002). Molecular characterization of three estrogen receptor forms in zebrafish: Binding characteristics, transactivation properties, and tissue distributions. Biol. Reprod. 66, 1881–1892.[Abstract/Free Full Text]

Mills, L. J., Gutjahr-Gobell, R. E., Horowitz, D. B., Denslow, N. D., Chow, M. C., and Zaroogian, G. E. (2003). Relationship between reproductive success and male plasma vitellogenin concentrations in the Cunner, Tautogolabrus adsperus. Environ. Health Perspect. 111, 93–99.[ISI][Medline]

Mittendorf, R. 1995. Teratogen update: Carcinogenesis and teratogenesis associated with exposure to diethylstilbestrol (DES) in utero. Teratology 51, 435–445.[ISI][Medline]

Mylchreest E., Wallace, D. G., Cattley, R. C., and Foster, P. M. (2000). Dose-dependent alterations in androgen-regulated male reproductive development in rats exposed to Di(n-butyl) phthalate during late gestation. Toxicol. Sci. 55(1), 143–151.[Abstract/Free Full Text]

Naciff, J. M., Jump, M. L., Torontali, S. M., Carr, G. J., Tiesman, J. P., Overmann, G. J., and Daston, G. P. (2002). Gene expression profile induced by 17a-ethinyl estradiol, bisphenol A, and genistein in the developing female reproductive system of the rat. Toxicol. Sci. 68, 184–199.[Abstract/Free Full Text]

Nagel, S. C., vom Saal, F. S., Thayer, K. A., Dhar, M. G., Boechler, M., and Welshons, W. V. (1997). Relative binding affinity-serum modified access (RBA-SMA) assay predicts the relative in vivo bioactivity of the xenoestrogens bisphenol A and octylphenol. Environ. Health Perspect. 105, 70–76.[ISI][Medline]

National Research Council. (1999). Hormonally Active Agents in the Environment. National Academy Press, Washington, DC.

O’Connor, J. C., Cook, J. C., Marty, M. S., Davis, L. G., Kaplan, A. M., and Carney, E. W. (2002). Evaluation of tier I screening approaches for detecting endocrine-active compounds (EACs). Crit. Rev. Toxicol. 32, 521–549.[ISI][Medline]

Paulozzi, L. J., Erickson, J. D., and Jackson, R. J. (1997). Hypospadias trends in two U.S. surveillance systems. Pediatrics 100, 831–834.[Abstract/Free Full Text]

Putz, O., Schwartz, C. B., Kim, S., LeBlanc, G. A., Cooper, R. L., and Prins, G. S. (2001a). Neonatal low-and high-dose exposure to estradiol benzoate in the male rat. I. Effects on the prostate gland. Biol. Reprod. 65, 1496–1505.[Abstract/Free Full Text]

Putz, O., Schwartz, C. B., LeBlanc, G. A., Cooper, R. L., and Prins, G. S. (2001b). Neonatal low- and high-dose exposure to estradiol benzoate in the male rat. II. Effects on male puberty and the reproductive tract. Biol. Reprod 65, 1506–1517.[Abstract/Free Full Text]

Reiter, L. W., DeRosa, C., Kavlock, R. J., Lucier, G., Mac, M. J., Melillo, J., Melnick, R. L., Sinks, T., and Walton, B. T. (1998). The U.S. federal framework for research on endocrine disruptors and an analysis of research programs supported during fiscal year 1996. Environ. Health. Perspect. 106(3), 105–113.[ISI][Medline]

Robinson, C. D., Brown, E., Craft, J. A., Davies, I. M., Moffat, C. F., Pirie, D., Robertson, F., Stagg, R. M., and Struthers, S. (2003). Effects of sewage effluent and ethynyl oestradiol upon molecular markers of oestrogenic exposure, maturation and reproductive success in the sand goby (Pomatoschistus minutus, Pallas). Aquat. Toxicol. 62(2), 119–134.[CrossRef][ISI][Medline]

Serafimova, R., Walker, J., and Mekenyan, O. (2002). Androgen receptor binding affinity of pesticide "active" formulation ingredients: QSAR evaluation by COREPA method. SAR QSAR Environ. Res. 13, 127–134.[CrossRef][ISI][Medline]

Shi, L. M., Fang, H., Tong, W., Wu, J., Perkins, R., Blair, R. M., Branham, W. S., Dial, S. L., Moland, C. L., and Sheehan, D. M. (2001). QSAR models using a large diverse set of estrogens. J. Chem. Inf. Comput. Sci. 41, 186–195.[CrossRef][ISI][Medline]

Silver, R. I. 2000. What is the etiology of hypospadias? A review of recent research. Del. Med. J. 72, 343–347.[Medline]

Stoker, T. E., Parks, L. G., Gray, L. E., and Cooper, R. L. (2000). Endocrine-disrupting chemicals: Prepubertal exposures and effects on sexual maturation and thyroid function in the male rat. A focus on the EDSTAC recommendations. Endocrine Disrupter Screening and Testing Advisory Committee. Crit. Rev. Toxicol. 30(2), 197–252.[ISI][Medline]

Suzuki, T., Ide, K., Ishida, M., and Shapiro, S. (2001). Classification of environmental estrogens by physicochemical properties using principal component analysis and hierarchical cluster analysis. J. Chem. Inf. Comput. Sci. 41(3), 718–726[CrossRef][ISI][Medline]

Swan, S. H., Elkin, E. P., and Fenster, L. (1997). Have sperm densities declined? A reanalysis of global trend data. Environ. Health Perspect. 105, 1228–1232.[ISI][Medline]

Todo, T., Ikeuchi, T., Kobayashi, T., Kajiura-Kobayashi, H., Suzuki, K., Yoshikuni, M., Yamauchi, K., and Nagahama, Y. (2000). Characterization of a testicular 17a, 20b-dihydroxy-4-pregnen-3-one (a spermiation inducing steroid in fish) receptor from teleost, Japanese eel (Anguilla japonica). FEBS Lett. 465, 12–17.[CrossRef][ISI][Medline]

Tyl, R. W., Myers, C. B., Marr, M. C., Thomas, B. F., Keimowitz, A. R., Brine, D. R., Veselica, M. M., Fail, P. A., Chang, T. Y., Seely, J. C., et al. (2002). Three-generation reproductive toxicity study of dietary bisphenol A in CD Sprague-Dawley rats. Toxicol. Sci. 68(1),121–146.[Abstract/Free Full Text]

van den Berg M., Peterson R. E., and Schrenk D. (2000). Human risk assessment and TEFs. Food Addit. Contam. 17(4), 347–358.[CrossRef][ISI][Medline]

vom Saal, F. S., Timms, B. G., Montano, M. M., Palanza, P., Thayer, K. A., Nagel, S. C., Dhar, M. D., Ganjam, V. K., Parmigiani, S., and Welshons, W. V. (1997). Prostate enlargement in mice due to fetal exposure to low doses of estradiol and diethylstilbestrol and opposite effects at high doses. Proc. Natl. Acad. Sci. U.S.A. 94, 2056–2061.[Abstract/Free Full Text]

Webb, S. F. (2001). A data-based perspective on the environmental risk assessment of human pharmaceuticals I. Collation of available ecotoxicity data. In: Pharmaceuticals in the Environment (K. Kummerer, Ed.), pp. 175–201. Springer, New York.

Wilson, V. S., Lambright, C., Ostby, J., and Gray, L .E., Jr. (2002). In vitro and in vivo effects of 17beta-trenbolone: A feedlot effluent contaminant. Toxicol. Sci. 70(2), 202–211.[Abstract/Free Full Text]