Toxic Influence

EPA develops neurotoxicants list, new testing

Wednesday, December 22nd, 2010 

This story is being co-published with Politics Daily.

RESEARCH TRIANGLE PARK, N.C. — As the rates of learning disabilities, autism and related conditions rise, the Environmental Protection Agency is preparing to release a roster of the pollutants likely to contribute to these or other disorders, American University’s Investigative Reporting Workshop has learned.

In an ongoing, three-year effort, an EPA team has determined which developmental neurotoxicants — chemicals that damage the fetal and infant brain — may pose the biggest risk to the American public.

Some of the compounds on the EPA’s list are ubiquitous in household products, drinking water, medicine and the environment, and range from cadmium, which is used to etch colorful cartoons onto children’s glasses, to flame retardants that are used to fireproof upholstered furniture.

The Investigative Reporting Workshop visited the scientists here in North Carolina and obtained a draft of the first chemicals expected to be on the list. Its release, expected early next year, is awaited eagerly by public health activists, who lack government confirmation that many chemicals they suspect are dangerous, and more warily by manufacturers.

“It legitimizes the public health concerns,’’ said Elise Miller, director of the Collaborative on Health and the Environment and a member of the EPA’s advisory committee on children’s health protection. “So many consumers hear contradictory reports from different scientific studies highlighted in the media, so having an EPA list more or less ‘settles’ the question of whether or not these are chemicals of concern. ’’

In addition to naming specific chemicals, solvents and other pollutants that turn up in everything from light bulbs to flea-and-tick products for pets, the compendium will be used to develop new, automated testing of plates of cells, to supplement and in some cases replace testing on animals, said EPA research toxicologist Kevin Crofton.


EPA photo

Kevin Crofton

This in vitro testing, as it is known, would speed up the screening of EPA’s backlog of more than 10,000 chemicals that could potentially damage the brain.

“There are too many chemicals and not enough data,” Crofton said. “There are literally thousands and thousands of chemicals in use for which we really don’t have much information at all.”

Many experts praised Crofton and William Mundy, part of a 12-member research group, for their enterprise but said the EPA should have compiled a list of chemical pollutants years ago  — and warned that the in vitro screen has some sizable holes.

“We need to do this, and we need to do it fast,” said R. Thomas Zoeller, an endocrinologist and professor of biology at the University of Massachusetts, who has served on numerous federal advisory committees. “Right now, one in six children in this country is diagnosed with a neurobehavioral disorder. The impact on our society and on our economy is really dependent on Kevin and Bill to come up with something that is workable, and it’s not going to be simple, and it’s not going to be fast.”

Experts spar over the precise numbers, but most agree that the data on learning and developmental disabilities indicate a rise — and that those numbers go well beyond the amount attributable to increased diagnoses.

The Collaborative on Health and the Environment, a partnership of public health researchers and environmental advocates, puts the rate of learning and developmental disabilities at 5 to 15 percent — or more than 12 million children. The National Parkinson Foundation estimates that 1 million Americans suffer from Parkinson’s disease, with 50,000 to 60,000 new cases diagnosed each year. That number is expected to skyrocket as the population ages.

Many scientists say that a significant number of these cases are related to exposure to toxic chemicals. The National Research Council, a division of the National Academy of Sciences, has reported that while only 3 percent of developmental disabilities can be traced to environmental pollutants, nearly an additional 25 percent are caused by interactions between individuals’ genes and what’s in the environment.


EPA photo

William Mundy

“The dose makes the poison” was a longstanding principle of risk assessment. In the past few years, that maxim has been replaced by “the timing makes the poison,” reflecting the evidence that exposure to pollutants during critical windows of development can wreak havoc on the growing nervous system.

Theodore Slotkin, a pharmacology professor at Duke University Medical Center, likened the developing brain, with its sequences of synchronized growth, to a pianist playing a sonata.

“Imagine that the assembly of the brain is you sitting at the piano, playing a Beethoven sonata,” Slotkin said. “… Certain keys have to be struck at the right intensity at the right pattern. Now, imagine that someone comes along with a chunk of 2-by-4 and presses down a dozen keys.” Different chemicals, Slotkin said, whack the brain the same way the 2-by-4 whacks your piano — with the result being a far different tune than what was intended. “The worst outcome of all,” he said, “is when you press down the keys and you can’t tell what piece is being played.”

From the standpoint of science protecting the public health, Slotkin said, “What we’ve been doing the past 30 years is a failure. They [industry] can produce these chemicals faster than we can test them.”

A group of prominent scientists said in a consensus statement two years ago that the evidence was convincing that 200 chemicals, manufactured in large quantities each year, can cause developmental disorders, defined as deficits in learning and memory, reduced IQ, attention-deficit disorder, autism spectrum disorders, behavioral problems and developmental delays.

But under the Toxic Substances Control Act, the EPA permits companies to put many chemicals on the market without first proving they are safe. Then, if evidence mounts that a particular chemical, such as Bisphenol A, or metal, such as lead, could, in fact harm people who are exposed to it in varying situations,  the agency has to play catch-up. “EPA can not even study the effects of a chemical, unless some independent research shows potential harmful effects,” said Dana Barr, an Emory University professor who ran the Centers for Disease Control and Prevention’s pesticide laboratory for 15 years. “There has to be a reason for EPA to study a chemical. They can’t say, ‘It is being used. Let’s study it.’ ”

EPA Administrator Lisa Jackson, other federal officials and a group of Democratic lawmakers tried this year to revamp the 1976 law, switching the burden of proof of chemicals’ safety to the manufacturers, but were beaten back by chemical industry lobbyists.

How work on the list unfolded

Crofton and Mundy’s group started with more than 400 suspect chemicals, taken from various sources, including a landmark article in the medical journal The Lancet in 2006.

In whittling down the list, their working team culled through all available animal and human studies in scientific literature and divided the studies into three groups based on their extent of proof: no evidence; minimal evidence; or substantial evidence. For substantial evidence, confirmation had to come from more than one research laboratory. “When you get into the literature, there is no real agreement on which are the neurotoxicants, and not much have been tested,” Crofton said.

He and Mundy hope to change that. They plan to use their high-evidence chemicals to test screening methods that they believe will expedite the testing of the thousands of chemicals currently in commerce. These automated laboratory experiments, called high-throughput in vitro tests, have generated tremendous enthusiasm among animal-rights groups and the research community, including the National Academy of Sciences, and are being used or put into place throughout federal agencies.

In these studies, cells are grown in a dish and then are exposed to one or more chemicals at a time. A scientist or computer tracks whether a given chemical causes an adverse reaction in the cells, or whether the chemical kills any cells. Crofton and Mundy plan to try out their high-throughput tests and compare the results with the hundreds of animal and human tests they’ve already reviewed. They envision a series of studies examining the effect of chemicals, alone or in combinations, on a cascade of events necessary for brain development, including the proliferation of cells, migration, division and formation of synapses.

All this action and reaction can be observed in a 96-well tray and analyzed in as little as 20 minutes.

Their goal is to create a way for the EPA to prioritize chemicals for further testing in animals and alternative species, such as C. elegans, a tiny roundworm, and zebra fish.

Crofton said animal testing is too slow. “If you give me a chemical and we go off for one or two years, we can give you answers,” he said. But with so many chemicals to be studied, “We can’t do it. There aren’t enough neurotoxicologists, enough rats or mice.” Or, Mundy added, “enough money.”

Still, several critics warn that the high-throughput in vitro tests will fail to detect many types of brain damage that can forever alter a person’s potential — tests in which behavioral changes are more subtle or observable only in a living, breathing creature— and therefore should not replace animal tests, even as a screening tool. They also question why it has taken the EPA so many years to finish the list, given the explosion of peer-reviewed studies showing the harmful effects of dozens of pollutants on the developing brain.

“My initial reaction is I think we should be further along with this,” said Steven G. Gilbert, director of the Institute of Neurotoxicology and Neurological Disorders, a nonprofit research center in Seattle. “I think there’s no reason [to take this long.] We’ve got data on a lot of neurotoxicants that should be ranked. That should have been done. We should have moved forward on the high-throughput screening.”

Although the project is still under way, the Investigative Reporting Workshop discovered an early list of the chemicals, presented on a poster at a toxicology symposium. The poster was then published on the EPA’s website — prematurely, according to Crofton, who said the list will be reviewed and changed in the coming months before being published.

Initially, the researchers designated the chemicals as, “Generally recognized as developmentally neurotoxic,” but then toned it down, and now call it the “level of evidence for developmental neurotoxicity” list.

In concert with Politics Daily, The Investigative Reporting Workshop website today is publishing the list of those chemicals with highest evidence of neurotoxicity based on the studies and which, therefore, are expected to be part of the EPA’s final list. Some of the compounds, such as lead and methylmercury, are already widely known for their ability to diminish intellectual potential and impair memory, among other hazards. Others, including chemicals used in medications, industrial processes and household products, are lesser known as developmental neurotoxicants.

Among those is benzene, one of the top 20 chemicals in production in the United States. Benzene is used to make compounds that are then incorporated into plastics, resins, nylon and synthetic fibers. Although benzene is also a byproduct of forest fires and volcanoes, people generally come in contact with it by using lubricants, rubbers, dyes, detergents, drugs and pesticides or in the workplace.

Another compound, acrylamide, can form in some foods, especially potatoes during frying, roasting or baking, and is also found in cigarette smoke. Acrylamide, too, is used in industrial processes to make plastics and cosmetics.

Other chemicals on the draft roster include:
    •   Aspartame, an artificial sweetener found in sodas, and in other food and drink

    •   Bisphenol A, a chemical widely used in consumer goods, including the resin lining of most food and beverage cans; products made from polycarbonate plastics, which include water bottles and baby bottles; and some dental sealants

   •    Cadmium, a heavy metal that is used in batteries, coatings and pigments and is plentiful in tobacco smoke

   •    PBDEs, or polybrominated diphenylethers, a class of chemical flame retardants that are on upholstery and in electronics,  among other things

   •   Pesticides and insect repellents, among them: aldicarb, DEET, lindane (used for lice and scabies), maneb and paraquat

   •   Trichloroethylene, formerly used in dry cleaning but still available as a cleaning and degreasing agent and a  contaminant in drinking water

Ultimately, the EPA group expects to name more than 100 chemicals, which will serve as a benchmark to determine whether their screens pick up the same problem chemicals discovered in animal studies.

Bernard Weiss, a professor of environmental medicine and pediatrics at the University of Rochester who is revered as one of the founders of the field of neurotoxicology, said the move to in vitro tests is a mistake. “It just seems to me that relying on in vitro screening for uncovering rather subtle effects, such as feminizing male play behavior, is putting a rather difficult question to a rather simple technique,” Weiss said. “My conclusion is that relying on massive batteries of in vitro screens to try to discover long-term subtle effects is an enterprise whose success lies very far in the future.”

Among other problems that won’t be picked up, Weiss said, are lowered testosterone levels, which are associated with an increased risk of Alzheimer’s disease; insulin resistance; and inherited changes in the expression of genes that may not show up for years.

Philippe Grandjean, co-author of the Lancet article, an adjunct professor at the Harvard School of Public Health and a professor at the University of Southern Denmark, called Crofton “brilliant” and said the project is a reasonable place to start. But, like Weiss, he is concerned that this new approach may miss dangerous pollutants.

One of Grandjean’s worries is that the chemical being tested might not be toxic but that the metabolite — the breakdown product that occurs when enzymes change the chemical structure — might be. “The metabolite may be the toxic agent, and if that does not occur in vitro, it may mean, like Bernie says, they will miss the toxicity. That is a major concern.” He said the in vitro cells’ lack of a blood-brain barrier, which protects the brain from many toxic chemicals, is also a major drawback to the new testing method.

Miller, of the Collaborative on Health and the Environment, said she was pleased to see the new test methods launched but wanted to ensure that studies on animals and people are not neglected. “We’re exposed to mixtures, not one at a time. That’s where the high-throughput testing will be helpful,” she said.

“It’s easy to get caught up in the new technology and get infatuated,” she said. “It needs to be complemented by animal and epidemiological studies, and an overhaul of the risk assessment. The high-throughput test isn’t going to tell you how a chemical is going to affect someone in West Harlem.” In addition, some chemicals, such as lead, polychlorinated biphenyls (PCBs), many pesticides, manganese, acrylamide and certain solvents, have already been studied quite enough, she said.

In several interviews with the Workshop, Crofton and Mundy stressed that they do not expect their high-throughput testing program to replace animal studies. They view it as a way to meet a critical need to test — in a cost-effective way — thousands of chemicals that are already ubiquitous in consumer goods, medicines, drinking water and the environment.

Since 1991, the EPA has tested only about 85 chemicals for neurotoxicity, according to Crofton. Some of the untested chemicals are used as inert, or inactive, ingredients in products. Noting that animal tests cost $700,000 to $1 million per chemical, he said, “If we had 25,000 Bernie Weisses and 60,000 labs and millions of rats, we could take a different approach.”

The researchers acknowledge the limitations of their methodology but contend that those limits aren’t flaws. “Science isn’t at this stage of the game, to say, ‘If it doesn’t produce an effect on the in vitro level, it’s safe,’ ” Crofton said. “The idea isn’t to use it as a substitute. The idea would be that instead of spending all the resources on more studies on lead and methylmercury … my vision is to look at new ones, rather than 45 more studies on lead. We know lead is bad. It’s my opinion that we spend way too much of our money on that kind of research as opposed to chemicals that we don’t know anything about at all.”

Mundy agreed. “The thing to understand is, if there is no effect in the in vitro [tests], it doesn’t mean the chemical is safe, it just has a lower priority” for further testing, he said. “Right now we don’t have all the tools, so our job here as scientists is to develop the tools and explain to risk assessors what kind of data you get out of them.’’

Next steps

It won’t be easy to parlay the test results into protective policies. The researchers included industry representatives in numerous meetings over the last five years to discuss their plans, but already there is pushback from industry.

Lorenz Rhomberg, a former EPA employee who now works for Gradient Corp., an environmental risk consulting firm that often works with businesses, said, “In industry, there are two strains of thought. One, this promises a chance to look at a lot of chemicals more thoroughly, much more cheaply and quickly, and could provide us with much better answers. And there are other people who say we are worried about how they might be used, because there is always this problem that chemicals might make physiologically detectable changes in the body that are not adverse. “

That’s precisely the argument that troubles Deborah Rice, an epidemiologist with the state of Maine and recent winner of the prestigious Heinz award for her neurotoxicology research. Industry is already arguing that adverse affects on a “pathway” in the dish don’t necessarily mean adverse effects in people, she said. If the EPA doesn’t accept results of the in vitro tests as evidence that a chemical may be dangerous, Rice said, “Then, that’s it, game over. … You are doing this for nothing.”

Public health experts also worry about which chemicals are being used to replace those found to be hazardous. Barr, of Emory, said she is especially concerned about Bisphenol A, which is used in the resin lining of cans and also in bottles and other products. Canada is moving to ban Bisphenol A, commonly known as BPA. “Now you have BPA-free containers, but nobody asks what the new containers are made of,” Barr said. “It’s a new polymer that has phthalate in it, plus two other chemicals. Two of them have no toxicity testing whatever. None has been done."

Barr also takes issue with the CDC’s biomonitoring program, which measures the types and amounts of toxic chemicals in people around the country and is used by the EPA to help set regulatory standards.

“One of the problems with the report is that CDC is just putting information out there and not offering any interpretation,” Barr said of the voluntary tests conducted periodically. “Lead is the one chemical that [CDC] has reported back to the participant, because there are interventions.” She would like to see the CDC contact participants who may have too much of other metals or chemicals in their bodies so they can follow up with their own doctors. “There are a lot of ethical issues,” she said.

In addition, she said, the CDC underestimates the levels of chemicals accumulating in people by choosing a time to take the blood samples based more on expediency and ease than seasons. “They sample the North in the summer and the South in the winter, so they are probably underestimating pesticide exposure,” she said as one example. “The whole value of it is you can extrapolate to the whole U.S. population, but they aren’t allowing for regional differences.”

Other scientists question how seriously the government takes the whole issue of environmental health hazards. Melanie Marty, branch chief of the air toxicology and epidemiology branch of California’s Office of Environmental Health Hazard ssessment, said, “People say, ‘What are a few IQ points?’ That is something we heard in the 1990s on lead,” Marty said. “The person will not know the difference. But as a society, do we want to be shifting our IQ lower? I don’t think so.”

“It’s very important to keep animal models,” added Edward Levin, professor of psychiatry and behavioral science at the Duke Institute for Brain Sciences. “It’s a lot of hubris to think we are anywhere near to understanding how the brain works, and all the points of vulnerability in terms of brain function and development.” The reality, said Levin, who conducts research on behavioral impairments caused by early exposure to pesticides and heavy metals, is that “everything is tested on animals. It’s a matter of who is first, the lab animal or the first person who uses a shampoo or takes a drug or is exposed to an environmental toxicant. We are animals, too.”

Levin, a past president of the Neurobehavioral Teratology Society, many of whose members study developmental neurotoxicity, said, “Behavioral toxicology is a field defined as a race to figure out what’s going wrong and how to deal with it before we’re too stupid to care.”

Additional research support provided by the Nation Institute’s Investigative Fund.


Toxic Influence

Stories in this investigation



See all documents

About the author

Sheila Kaplan