Government scientists are launching an ambitious collaboration to shift the testing of potentially toxic chemicals away from animals to methods that use high-speed automated robots. The robots would test chemicals on human cells at various concentrations, generating data relevant to humans faster and more cheaply than current methods, researchers from the National Institutes of Health (NIH) and the Environmental Protection Agency (EPA) announced in Boston last week during a teleconference at the annual meeting of the American Association for the Advancement of Science. The project is also described in a paper in the Feb. 15 issue of Science.
In the past 30 years the National Toxicology Program (NTP) of the NIH has tested about 2,500 chemicals in great detail, says NTP Associate Director John R. Bucher. Now, with advances in molecular and computational biology, the same number could easily be done in an afternoon. Human cells of interest, such as liver or skin cells, can be placed in each of 1,536 tiny wells in a single dish, says NIH Chemical Genomics Center Director Christopher Austin. Concentrations that vary more than a thousandfold can be applied to each well, or 1,536 different chemicals could be applied. Scientists can then analyze the cells to see which survive, grow, die, or stop dividing. They may also determine whether a particular cellular biochemical pathway is affected, says Austin.
The agencies will initially focus on the roughly 2,500 chemicals for which solid data have been collected in previous studies, the researchers say.
Historically, much testing involved injecting a chemical into an animal, watching to see whether it got sick and then investigating its tissue, says Francis Collins, director of the NIH’s National Human Genome Research Institute (NHGRI). “It’s slow, expensive, and has limited predictive power,” says Collins. “We are not rats.”
But animal testing can’t be abandoned overnight, says Elias Zerhouni, director of the NIH. Animal and human studies will remain intertwined until the effects that have been observed in previous studies are validated, he says.
The memorandum of understanding for the project outlines a 5-year effort combining the forces of two NIH agencies—NTP and NHGRI—and the EPA’s recently formed National Center for Computational Toxicology. The high-speed robots to be used in the project are from NIH’s Chemical Genomics Center. The EPA will pull together the data, comparing the animal and human studies.
Previously, there has been cross-agency pollination on some projects, but each group was guided by its respective priorities and budgets, says Anne Miracle, a senior research scientist with the Environmental Sustainability Division of the Pacific Northwest National Laboratory in Richland, Wash. Now, she says, “It’s very heartening to see that these agencies are going to pool their efforts.”
The robots will study human cells and test chemicals on cells from model organisms such as zebra fish and roundworms, which are studied extensively for insight into biochemical reactions.
