Toxicogenomics research consortium sails into uncharted waters. (NIEHS News).

In the current state of gene expression technology, there are various methodologies for assessing gene expression, making it difficult to compare and compile data across laboratories and investigators. To stretch research dollars and coordinate research efforts among scientists who are using genomics to better understand environmental effects on human health, the NIEHS Division of Extramural Research and Training last year funded a $37 million partnership called the Toxicogenomics Research Consortium (TRC). The idea is simple, yet powerful: the synchronization of research efforts of many scientists in various locations promises to produce high-quality results faster and more reliably than isolated research teams working alone.

This consortium merges the efforts of investigators with similar environmental health research goals from the NIEHS Microarray Center and five of the country's leading academic research institutions: the University of North Carolina (UNC) at Chapel Hill, Duke University in Durham, North Carolina, the Fred Hutchinson Cancer Research Center/University of Washington (UW) in Seattle; the Massachusetts Institute of Technology (MIT) in Cambridge, and Oregon Health & Science University (OHSU) in Portland. Each of the five academic institutions will receive $7.5 million over five years.

The consortium is charged with defining sites of genetic variability, developing standards for gene expression experiments, and applying gene expression technology to study environmental stress responses in biological systems. These efforts are critical for defining how environmental agents cause disease, identifying biomarkers of disease, predicting chemical toxicity, and understanding why individuals show such a wide variability in their sensitivity to environmental toxicants.

Consortium research projects fall within the realm of environmental toxicogenomics. This emerging discipline of toxicology enables scientists to identify and characterize the genomic signatures of environmental toxicants, as well as use gene and protein expression profiles to study the relationship between exposure and disease and understand gene-environment interactions and their impact on human health.

Research among the Academic Partners

At Fred Hutchinson/UW, principal investigator Helmut Zarbl is leading efforts using DNA microarrays to determine whether particular genes are sensitive to the actions of chemical toxicants and the role of these genes in cancer development. Microarray technology enables researchers to measure the transcription of multitudes of genes simultaneously on tiny microchips containing thousands of targets of complementary DNA (cDNA) or synthetic DNA segments (oligonucleotides) immobilized in a preset arrangement.

Zarbl's group is studying rat strains with different susceptibilities to mammary carcinogenesis to help identify which differentially expressed genes are related to the mechanims of sensitivity. "The ultimate goal is to predict an individual's risk of cancer based on [his or her] genetic profile and environmental exposures," says Zarbl, a member of the center's Human Biology and Public Health Sciences Divisions and an affiliate associate professor of pathology, environmental health, and toxicology at UW.

In other studies at Fred Hutchinson/UW, researchers hope to better understand how environmental factors--metal exposure, nutritional changes, and physical factors such as hyperthermia due to maternal fever--can damage the developing nervous system, and how exposures to certain organophosphates used in pesticides can affect child behavior. Other projects seek to develop tests to measure toxic exposure and stress responses using lab-cultured liver cells.

Researchers at OHSU are collaborating with the Boston Biomedical Research Institute to compare gene and protein expression patterns of nervous system cells in both normal and toxicant-exposed states. Under the direction of OHSU principal investigator Peter Spencer, they hope to increase understanding of the mechanisms underlying toxic diseases of nerve cells and their axons, as well as clarify ways to screen for environmental agents and hazards that have the potential to cause neurodegeneration.

"Toxicogenomics should allow us to rapidly identify changes in the genome and proteome Jail the proteins expressed by a genome], giving us a `fingerprint' of the effect of a particular chemical," explains Spencer, a professor of neurology in OHSU's School of Medicine and director and senior scientist at the university's Center for Research on Occupational and Environmental Toxicology (CROET). That fingerprint, or gene expression profile, he says, allows the team to see if classes of chemicals are working to produce a particular toxicological effect.

CROET researchers have exposed mice to two isomers of diacetylbenzene, one neurotoxic and the other not, hoping to assess gene modulation and isolate the neurotoxic component. By looking at the common gene expression signatures expressed by both agents, they can subtract a portion of each signature, leaving only the shared portion, which reflects changes caused by the neurotoxic element. This element will be compared with the neurotoxic signature of homologous neurotoxic chemicals to determine whether commonalities can be identified.

A second OHSU project will study the effects of a potent genotoxin present in the seed of the cycad, a tropical palm-like plant. Cycad seeds have been used for food and medicines on three Pacific islands--Guam, Papua, and the Kii peninsula of Honshu Island, Japan--where disproportionately high numbers of people have suffered from lytigo-bodig, a rare condition that shares symptoms with amyotrophic lateral sclerosis (Lou Gehrig disease) and Parkinson and Alzheimer diseases. "Understanding the cause and pathogenesis of this prototypical neurodegenerative disease may shed light on look-alike disorders worldwide," says Spencer.

On the East Coast, Leona Samson heads up a team of scientists at MIT who are studying the effects of aflatoxin [B.sub.1] (which is secreted by mold that grows on grains and is associated with liver cancer) as well as environmental alkylating agents in various cell types, including liver cells. "We're using model systems--animals or cells we know have [well-defined] biological outcomes when exposed to environmental agents--to determine what we can say about the toxicogenomic response," Samson says. "These agents represent toxic agents found in our external environment, our food supply, the endogeneous cellular environment, and the cancer chemotherapy clinic."

Another MIT study involves engineering liver cells to create liver "bioreactors" that process toxic compounds. These three-dimensional structures secrete the protein albumin and even reproduce the force and sheer caused by blood flow, allowing them to mimic how the liver actually functions in a living body. Using the engineered version, scientists will be able to explore how the human liver might respond to toxic insult.

At UNC, researchers from the Lineberger Comprehensive Cancer Center and the Center for Environmental Health and Susceptibility will study profiles of genetic susceptibility to toxicant stress. Led by William K. Kaufmann, a professor of pathology and laboratory medicine in the School of Medicine, UNC researchers are using microarrays to study known environmental and clinical carcinogens as well as nongenotoxic carcinogens, which appear to induce cancer through mechanisms other than DNA damage. "A protein inside the cells interacts with the compound through specific receptors, which accounts for some or all of its toxicity," explains Kaufmann.

In other UNC studies, researchers will compare gene expression changes in mammary epithelial cells after treatment with various breast cancer therapies, focusing on what appear to be overlapping responses. They will evaluate how genetically pure mouse strains differ in their response to alkylating agents, watching for clues about the genetic predisposition within families to developing a particular cancer.

Researchers from Duke University's School of Medicine and the Nicholas School of the Environment and Earth Sciences, led by David A. Schwartz, a professor of pulmonary and critical care medicine, will watch how novel genes influence an organism's ability to defend itself against endotoxins, cell wall components of gram-negative bacteria. "So far, we've found that there are several important genes expressed that appear to genetically regulate the response to endotoxin," says Schwartz.

A second group of researchers at Duke is studying gene expression in developing zebrafish to see how certain components of retinoic acid synthesis and breakdown pathways are expressed in the neural tube. In a human component of this research, investigators have been studying how variants of the associated genes relate to neural tube defects in children, with results so far ruling out the retinoic acid pathway.

A third project, still in development, will bring together researchers from Duke, Fred Hutchinson, and MIT to study gene response to toxic metals in four different model systems: human cells, mice, zebrafish, and Caenorhabditis elegans, a unique roundworm whose genome is fully mapped. "The goal is to identify genes relevant across model systems that are common in all model systems," says Schwartz. "If exposure to toxic agents moderates a gene response in an evolutionarily conserved set of genes or stimulates common gene pathways in multiple species, then [that] narrows the biology of the response," he hypothesizes. "I've been encouraged by the interest among [TRC members] in working together to achieve common goals," he adds. "The cross-species comparisons are only possible with the cooperation of multiple scientists at different locations."

No matter what they are studying, all of the scientists participating in the consortium must wrestle with the "huge question of reproducibility," Samson cautions. "We must compare our results and see where the variabilities come from," she says. "Do they come from certain genes or certain array platforms?" She adds that member scientists should develop initial standards of procedure from the outset "to make sure we're all producing high-quality data."

Research at the NIEHS Microarray Center

Besides the university constituents, the NIEHS Microarray Center participates in the TRC as well. Center director Richard Paules says the NIEHS group is studying the mechanisms involved in nongenotoxic carcinogens, the role that estrogenic compounds play in genetic damage, and human susceptibility syndromes and how oxidative damage contributes to cancer development. In the latter study, Paules says, "we'll focus less on tissue and more on mechanistic pathways to find out the role of oxidative stress in the cancer process. People with inherited susceptibility to cancer lack the p53 gene that generates cell-cycle checkpoint arrest in DNA. These gene mutations are involved in the DNA damage response." He hopes the group's findings will lead to better diagnoses and treatment therapies for cancer patients.

Paules is currently leading a study of how rodent toxicants affect the gene response in rat livers, kidneys, and peripheral lymphocytes in the blood. "We want to know which exposures are related to disease process effects, not just pharmacological effects," he explains. His group will also search for phenotypic anchors of gene expression changes: "We want to make sure we have a very dear definition of the biology of what's going on in a particular organism, by defining traditional toxicological parameters and making sure gene expression changes are linked to phenotypic changes in the liver models."

Paules looks forward to joining forces with like-minded counterparts at the other TRC member institutions. "Our goal is to be very interactive and collaborate not only with academic partners, but industry as well," he says. He believes this goal will be facilitated by the Chemical Effects in Biological Systems "knowledge base" being developed by the NGT together with the consortium [see "Toxicogenomics: An Emerging Discipline," p. A750 this issue]. "It will contain as much biological, chemical, and toxicological information as possible related to gene expression changes," he explains. "Our goal is to build a publicly accessible database so we can do predictive toxicology."