A scientist finds benefit in small doses of toxins

Gareth Cook, Globe Staff
12/12/2003

AMHERST -- Edward J. Calabrese, a gray-haired man who works in a rundown office surrounded by documents on highly toxic chemicals, has an explosive idea.

For more than a decade, Calabrese, a respected professor of toxicology at the University of Massachusetts, endured ridicule as he gathered evidence showing that small amounts of poisons, even cancer-causing chemicals like dioxin, can be good for you.

His research threatens to overturn a key principle of environmental regulation, which assumes that if a large quantity of a chemical causes cancer, then a small quantity is still dangerous, and that the ideal amount is zero. Calabrese's work suggests that for many chemicals, exposure to a low level may be healthier than no exposure at all.

Though long relegated to the scientific fringe, Calabrese's idea is suddenly being taken seriously. He has landed several papers in prestigious research journals. Other scientists are citing his work, the invitations to speak at universities and scientific meetings are flooding in, and the concept has been added to two leading toxicology textbooks.

All of this has put Calabrese at the center of a politically charged debate with broad implications for health. If the regulations that protect the nation's air, water, and soil are not stringent enough to keep toxins below hazardous levels, Americans will die. Yet if Calabrese is correct, and small quantities of many toxins can actually be beneficial, then it could bring innovative drug therapies, save billions by relaxing overly strict environmental standards, and fundamentally change the way scientists and the public think about poisons.

"I think he is shaking us all up in a way that is really useful," said George Gray, a toxicologist who is executive director of the Harvard Center for Risk Analysis.

The concept underlying Calabrese's work is called "hormesis." In the broad sense it is hardly controversial. Vitamins are healthy in the right dose; toxic in larger ones. A glass of red wine a day can be good for you; a gallon is not. But this is not how scientists have traditionally thought about the risks posed by environmental chemicals. One of toxicology's most important tools is to observe the effects of large doses of a chemical on laboratory animals, and then use that data to estimate the effects of much lower doses on humans over longer periods.

In the case of cancer-causing agents, toxicologists assume that the harmful effects decrease as the dosage goes down, but that they do not hit zero until the exposure is zero. For threats not involving cancer, the model is only slightly different; scientists also assume that smaller doses cause less harm, and the harmful effects hit zero as soon as the dose hits a certain low threshold.

These two ideas form the bedrock of modern toxicology, but Calabrese began to suspect that they were wrong when he discovered, as a college student, that spraying peppermint plants with very low doses of a growth retardant made the plants grow larger. So at low doses, the growth inhibitor didn't just stop working -- it had the opposite of its intended effect.

Other scientists have noticed unexpected effects like this, Calabrese said. At low doses, both dioxin and DDT have been shown to reduce some cancers in lab animals. Low doses of cadmium, which can be highly toxic, reduces liver cancer in rats.

In his research, Calabrese, 57, has shown that these effects may be very common.

Calabrese and a colleague searched through the toxicology literature, looking for all examples where scientists had measured the response to doses below the threshold at which the chemical is thought to have no effect. Their statistical analysis, published this year in the journal Toxicological Sciences, showed that, on average, these low doses had a measurable effect -- itself a surprise -- and that the effect was the opposite of the large-dose effect. Chemicals that had a bad effect at high doses tended to have a beneficial one at small doses.

Their analysis included a wide variety of life forms -- including plants, animals, and microbes -- and of effects -- such as growth, reproduction, and behavior.

"What I think is going on here is a revolution in thought to a bunch of people who are not used to a revolution," Calabrese said.

Before hormesis could be used to justify changes in regulations, scientists would need a better understanding for exactly how it works, said William H. Farland, acting deputy assistant administrator for science in the Office of Research and Development at the Environmental Protection Agency. A chemical that is beneficial in one way may cause problems in other areas, he said. Or levels of a chemical that may be healthy for some people, or even positive, may be harmful for children, pregnant women, or others.

And even when unexpected things happen at small doses, they are not always positive. Several researchers have found that certain chemicals that act like hormones can cause damage at much lower levels than anyone had expected.

As the science improves, Farland said, researchers are seeing more and more surprises at low levels of exposure.

This "most likely represents very complex biology," Farland said, "and what we have to do now is understand that biology."

One mechanism behind hormesis is that small amounts of chemicals can evoke a stress response from cells, causing them to devote resources to defending themselves.

For example, laboratory animals forced to fast periodically, and thereby put stress on their entire system, develop neurons that are more resistant to diseases like Parkinson's, according to research done by Mark P. Mattson, chief of the Laboratory of Neuroscience at the National Institute on Aging. Other experiments have shown that chemicals can evoke the same stress responses.

Mattson recently invited Calabrese to visit his lab to discuss the possibility of experiments that would test whether low doses of otherwise toxic chemicals strengthen the brain's defenses against diseases like Alzheimer's, Parkinson's, or Huntington's.

"The idea would be to identify a treatment that could be given long term and delay the onset of disease," Mattson said. "This is a long way from something that could be applied in humans, but it is worth pursuing.

Calabrese's work, said Farland and other scientists, is part of a dramatic rethinking of the biological effects of low-level exposures. New scientific tools and the explosion of detailed genetic information is allowing scientists to move from the standard animal toxicity tests, which use massive doses, to a more detailed looks at how individual molecules interact with living cells. This has led to a growing recognition that effects can differ in kind, not just degree, as the concentration changes.

Indeed, the National Toxicology Program, the government's clearinghouse for toxicology research, just began an expansive, year-long review to change its approach, given the developing scientific approaches.

"The idea is, what are we going to do to change this discipline," said Christopher J. Portier, the program's associate director. "I am sure that hormesis will be a part of the discussion as we move through this."