Screening and testing for endocrine disrupters
Science lags behind U.S. legislative demands.
 |
| © DAVID JULIAN/SWEETLIGHT |
In the United States, the lobbying of women's rights activists demanding more breast cancer research and environmentalists pushing to protect wildlife was enough to force the U.S. Environmental Protection Agency (EPA) to take a closer look at endocrine disrupters. Before the agency had even gotten its feet wet, two bills were passed by Congress in August 1996--the Food Quality Protection Act (FQPA) and amendments to the Safe Drinking Water Act (SDWA) (Anal. Chem. 1996, 68, 598 A)--mandating that EPA develop a screening and testing program for endocrine disrupters in food and drinking water within two years, implement the program by August 1999, and report back to Congress by August 2000 (Anal. Chem. 1997, 69, 15 A). The first two years are up, and many fear that policy is moving too fast for science to keep up.
The definition of an endocrine disrupter has been at the center of the debate from the beginning. Today, experts still cannot agree on what is meant by the term. Nonetheless, the FQPA specifically requires screening and testing of pesticide active ingredients and other chemicals that show estrogenic effects in humans. Any chemical shown to have androgenic and thyroid-related (as well as their anti-) effects or hormonal effects in wildlife, along with certain drinking-water contaminants, pesticide inerts, and anything on the Toxic Substances Control Act inventory may also be included. In other words, EPA has the authority to include in its final regulatory implementation any chemical that it deems appropriate. Many believe that few classes of chemicals will be exempt from this ruling.
Developing a screening and testing program that encompasses tens of thousands of chemicals centered around something as complicated as the endocrine system is not an easy task. Trying to develop such a program in two years makes it even more challenging. To help EPA get the process moving, the Endocrine Disrupter Screening and Testing Advisory Committee (EDSTAC), representing a wide range of constituencies, was chartered in October 1996. Its mission was to advise EPA on "a strategy for selecting and prioritizing chemicals for screening and testing, identifying new and existing screening tests and mechanisms for validation, and determining what tests to conduct and when to test beyond screening."
|
| Anthony Maciorowski |
EDSTAC recommends. . .
EDSTAC recommends that a two-tier system, which encompasses a battery of in vitro and in vivo biological assays, be used for screening and testing. A screening tier would be used to "identify substances with endocrine-disruption potential for further testing", and a testing tier would "identify adverse effects and establish dose-response relationships for hazard assessment." The proposed tier-one screening (T1) and tier-two testing (T2) batteries are given in Tables 1 and 2, respectively. None of the T1 assays are fully validated, and some are more developed than others. EDSTAC recommends that stakeholders be involved in the validation and standardization process, which is expected to take two to three years. In addition to these biological assays, the committee recommends that high-throughput pre-screening (HTPS) assist in priority setting. The overall screening and testing framework recommended by EDSTAC is depicted in Figure 1.
Figure 1. Screening and testing framework recommended by EDSTAC.
"In terms of initial sorting of chemicals, there are four major categories," says Maciorowski. The first category includes chemicals for which there is already sufficient information to determine that they are not a problem. According to Maciorowski about 30,000 polymers will likely go into this "hold box" and not be tested. However, monomers (that comprise these polymers), plasticizers, additives, and degradation products are not included in this category. On the other side of the coin are chemicals for which there is sufficient data to know that they are going to cause problems. There are fewer than 50 chemicals that fall into this category, says Maciorowski, for which there is enough information to bypass T1 and T2 and go directly to hazard assessment.
Falling somewhere in the middle of these two "book ends" are the remaining categories. One includes chemicals for which there is a small amount of screening data, enough to bypass T1 and go directly to T2. Maciorowski believes that there are about 600 chemicals that would fall into this category. The remaining category includes chemicals for which data are insufficient to know anything. This last category--estimated to contain about 15,000 commodity chemicals and pesticide active ingredients--is what really needs to be prioritized for screening, says Maciorowski.
It is for this last category of chemicals that EDSTAC recommends using HTPS methods. "HTPS consists of automated reporter gene assays that will allow you to determine whether a chemical will bind to the estrogen, androgen, or thyroid receptor," says Maciorowski. "Positive results in HTPS do not necessarily make a chemical an endocrine disrupter, but it gives you some sense of priority that these ought to be looked at by additional in vitro and in vivo screening."
In addition, quantitative structure-activity relationships (QSARs) can be used to predict whether a compound will bind to a given receptor. "HTPS could be used to validate the QSARs," says Maciorowski. The more information that is known about a chemical compound, the easier it is to predict how it will behave. "You set priorities based on all available information. Once you complete HTPS you are moving out of sorting and into priority setting. You then start screening those with the highest priority."
EDSTAC recommends that a "weight-of-evidence" approach be used when interpreting results from a battery of assays. "No single test endpoint has the whole answer. You have to look at the overall weight of all of the data," says Maciorowski. Certain tests will have more meaning than others--you try to arrange your data in a hierarchy, he says. "Generally, a battery of screens with their sum total of results will give you a better picture than a single-receptor binding assay."
The process of collecting these results is expected to be expensive and time-consuming. In a survey coordinated by Chris Borgert of Applied Pharmacology and Toxicology, 14 labs performed the procedures proposed by EDSTAC and estimated the total costs per compound to be about $200,000 and $1 million for screening and testing, respectively. However, these cost figures could change as new assays evolve and the proposed assays become validated.
Parallel efforts
The United States is not alone in its concern over endocrine disrupters. The Organisation for Economic Cooperation and Development (OECD), headquartered in Paris, has established a working group to internationalize the development and validation of methods for "testing and assessing the endocrine-disrupting potential of chemicals." At a meeting in March, the working group agreed on a testing strategy that is similar to what EDSTAC proposed.
According to a press release, the conceptual framework includes screening to "set priorities and study modes of action" and two-level in vivo testing to "identify and characterize potential endocrine-disrupting effects of chemicals and provide a definitive answer." The working group also agreed to launch a validation project in which member countries and industry will participate. The project will consist of two teams, one for validation of mammalian tests and one for fish tests. The next meeting is scheduled for this fall.
Other efforts are also gaining an international flavor. About one year before EDSTAC was established, the White House formed a working group on endocrine disrupters under its Committee on Environment and Natural Resources (CENR) with members from 14 federal agencies.
The CENR endocrine-disrupter working group has established an online federal research inventory, which is currently available on EPA's Website (http://www.epa.gov/endocrine), but, according to Robert Kavlock, director of reproductive toxicology at EPA's National Health and Environmental Effects Research Laboratory, the federal inventory will soon be merged into what is being called "a global inventory"--an online database of endocrine-disrupter research, thus far including U.S., Canadian, European, and Japanese inventories. Industry research is not included in the U.S. and Canadian inventories. "We've had many discussions with CMA [Chemical Manufacturers Association] on adding industry projects--there are legal roadblocks," says Kavlock.
The International Programme on Chemical Safety (IPCS), which is part of the World Health Organization (WHO), the United Nations Environment Program, and the International Labor Organization will take the lead in overseeing the global inventory, says Terri Damstra of IPCS, who is coordinating the project. Long-term maintenance of the inventory will be handled by the Joint Research Center in Ispra, Italy. In addition to the global inventory, the IPCS is also leading work on the international assessment of the state of the science--a report that will summarize what is and is not known worldwide regarding endocrine disrupters. The report is expected to be complete in two years.
Are we being exposed?
Researchers around the world are beginning to assess human exposure to various potential endocrine-disrupting chemicals. For example, the Centers for Disease Control and Prevention (CDC), in collaboration with the National Cancer Institute, the National Institute of Environmental Health Sciences (NIEHS), the National Institute of Occupational Safety and Health, researchers in northern Europe, and others have been investigating various chemicals that could potentially induce breast cancer, endometrial cancer, and non-Hodgkin's lymphoma. One study involves tracking dioxins, furans, PCBs, and persistent pesticides in serum samples from Norway. "This is a case-control study to look at about 300 serum samples collected as far back as 1973 from women who later developed breast cancer," explains Donald Patterson of the National Center for Environmental Health at CDC.
Nonpersistent compounds such as phthalates, which are not believed to accumulate in the body, are much more difficult to monitor. Rather than look for the parent phthalates, researchers at CDC are working on a method to detect phthalate monoesters in urine, says John Brock, one of the founding members of CDC's endocrine-disrupter working group. "The parent phthalates are ubiquitous, therefore our analytical team is focusing on the human metabolite--the monoester." Phthalates are commonly used as plasticizers and are found in many plastics, including children's toys.
The CDC endocrine-disrupter working group is focusing its efforts on fetal-exposure effects. "We are looking at neurodevelopment, hypospadias [a condition believed to be increasing in baby boys in which the opening of the urethra is on the underside of the phallus rather than on the tip], cryptorchism [failure of the testes to descend into the scrotum], and hearing effects," says Brock. "The development of the middle ear in the fetus is very dependent on normal thyroid function in the mother," he says. "We are also very concerned about PCBs because they metabolize to hydroxy PCBs, which are structurally similar to thyroxin and have been shown to bind to thyroxin receptors."
Historically, CDC has been the primary laboratory working on the National Health and Nutrition Examination Survey, which involves collecting national averages for substances such as lead, nonpersistent pesticide metabolites, and volatile organic chemicals, says Larry Needham, chief of CDC's Toxicology Branch. "We are also working with NIEHS on obtaining reference values for phytoestrogens [naturally occurring estrogen-mimicking compounds found in foods such as soybeans and alfalfa sprouts]," adds Needham. The studies are designed to show the normal background levels of these chemicals in humans.
One way to monitor exposure of estrogen-mimicking compounds to fish is to look at the biomarker vitellogenin, a complicated proteinaceous substance found in the yolks of eggs. Only mature female fish are expected to have vitellogenin in their blood, says Bente Nilsen of Biosense Laboratories (Norway); however, male and juvenile fish that have been exposed to estrogenlike compounds will also have significant amounts. "We have developed antibodies against vitellogenin. By using these antibodies in immunological methods such as ELISA and Western blot, you can detect vitellogenin inductions," explains Nilsen. It is difficult to find an antibody that will recognize vitellogenin in all fish species because vitellogenin is similar but not identical in different kinds of fish, says Nilsen. However, Biosense has found antibodies that will recognize vitellogenin from many different fish species. Another biomarker, which some believe is more sensitive because it is induced at low concentrations of endocrine disrupters, is zona radiata protein (Zrp). Biosense also makes antibodies against Zrp.
Beyond bioassays
Just because a compound binds to a hormonal receptor does not mean that it will cause disease. Jeff Carbeck, a professor of chemical engineering at Princeton University, has combined the separation power of CE with the selectivity of immunoassays (affinity CE) to help identify which compounds will likely cause adverse effects. "With affinity CE you can screen compounds directly for binding to a hormone receptor, plus you can determine the ability of the activated hormone to bind to the DNA hormone response element," says Carbeck. In principle, the compound's identity can then be confirmed if you couple CE to MS. Carbeck believes that a big advantage of CE is that it can be miniaturized on microchips, making the entire assay portable. The big problem, he says, is sensitivity. "You want to make sure you have enough of the compound to get significant binding." Often, aqueous samples will contain low concentrations of these chemicals, particularly the hydrophobic estrogenic compounds.
Aaike Oosterkamp of the Department of
Medical Bioanalysis at IIBB/CSIC (Spain)
and co-workers at the University of Leiden
(The Netherlands) have developed an LC
system with on-line receptor affinity detection that can detect levels of 17
-estradiol
and diethylstilbestrol down to 5 nM based
on a competitive fluororeceptor assay. According to Oosterkamp, a solution containing the human estrogen receptor is added to
the LC effluent and allowed to react in a first
reaction coil. "In a second step, the fluorescent estrogenic compound coumestrol is
added to saturate the remaining free binding
sites," he says.
Prior to fluorescence detection, the bound and free coumestrol are separated in a restricted-access column. To detect lower concentrations of estrogens, a preconcentrating procedure involving solid-phase extraction can be used. The method has also been used for the analysis of nonylphenol, a degradation product of nonionic surfactants, which is believed to have endocrine-disrupting abilities.
EDSTAC's proposed screening and testing battery focuses primarily on identifying what chemicals have the ability to disrupt the endocrine system. What's missing is the environmental information on the concentrations and occurrence for many of these compounds, says Larry Barber of the U.S. Geological Survey. "We are trying to come up with a relatively routine method that would be suitable for a production laboratory for analyzing alkylphenolic-type compounds in sewage effluents," says Barber.
One of the problems with the alkylphenol compounds is that they can occur as a series of about 25 isomers. "Using continuous liquid-liquid extraction followed by GC/MS, we can routinely resolve about 15 isomers," says Barber. The alkylphenol that has received the most attention is nonylphenol (nine carbons), but there are also a number of other structurally similar chemicals that have eight or ten carbons. During wastewater treatment, surfactants are degraded primarily along the ethylene oxide chain. What results is a complicated mixture of ethoxylate oligomers and a carboxylic acid derivative of each one of those oligomers, in addition to the nonylphenol isomers, explains Barber.
The method has also been used to look at bisphenol A (found in polycarbonate) and biogenic steroids, such as ethinylestradiol, testosterone, and estrone. "The problem with this method is that the detection limit is too high for steroids. Sewage effluents typically contain less than 10 pptr." For almost all of the components, detections limits are in the 0.05-0.5 ppb range.
One way to eliminate the chromatography step and go directly to mass identification is by Fourier transform ion cyclotron resonance (FT-ICR) MS. The instrument is not cheap but is unique in its high mass-resolving power and mass accuracy. When used in conjunction with electrospray ionization, FT-ICR MS can be used to characterize the binding between small molecules and large biomolecular targets, including hormone receptors, enzymes, and nucleic acids. The high mass-resolving power of FT-ICR allows researchers to perform mass spectrometric binding assays on thousands of compounds in a single 15-minute experiment, says Steven Hofstadler of the Ibis Therapeutics Division at Isis Pharmaceuticals. Specific compounds that bind to the target can be isolated in the gas phase and further characterized using tandem MS. Hofstadler calls the technique "mass chromatography" and uses it to test thousands of compounds in combinatorial libraries for their ability to bind to synthetic RNA drug targets. "Not only can we determine which compounds from the library bind to the target, we can determine the compound's binding affinity and binding location," he says. This method shows great promise for HTPS of potential endocrine-disrupting chemicals and represents a new paradigm in the drug discovery process.
Although biological assays are sensitive, they cannot confirm the identity of the chemical that causes a positive response, says Andrew Grange, senior research associate at EPA's Environmental Sciences Branch in Las Vegas. He and his co-workers are also turning to high-resolution MS for confirmation. They have developed software--Mass Peak Profiling from Selected Ion Recording Data and a Profile Generation Model--for the determination of elemental compositions of molecular and fragment ions in complex mixtures that enter a double-focusing mass spectrometer as chromatographic peaks. The method provides a 100-fold increase in sensitivity and a 6-fold increase in scan speed relative to full scan techniques, says Grange. Elemental compositions are determined for ions with masses up to 600 Da that contain carbon, hydrogen, oxygen, nitrogen, or sulfur.
Methods for identifying chemicals with endocrine-disrupting potential are evolving rapidly. Within one year, EPA will be required to implement a screening and testing program for endocrine disrupters based on standardized and validated tests. Several assays proposed by EDSTAC still need a fair amount of work. Many fear that by the time validation is complete, new assays that are more cost-effective will have been developed. Will EPA allow flexibility in what assays and tests are allowed in their final implementation? Based on the EPA's recent move toward performance-based methods within the agency, alternative methods will likely be considered. In addition to new biological-based assays, perhaps new analytical chemistry methods will also help to fill in some of the missing pieces.
