Center News

Testing cancer's chemical connections

Nov. 6, 2003
Dr. Shannon Payne operating a microscope

Dr. Shannon Payne, postdoc in the Kemp lab, suspects that a strain of cancer-susceptible mice discovered at the center could help researchers understand how chemicals or dietary factors cause tumors. The mice harbor a mutation in a cancer gene called p27.

Photo by Todd mcNaught

Cigarette smoke, asbestos, PCB's and hundreds of other chemicals are known or suspected cancer-promoting agents. Yet after more than four decades of study, the toxicity of many of these agents-or how they cause disease-remains uncertain.

Now, center researchers seek insight into this problem through an experimental system developed at Fred Hutchinson for studying the connection between cell division and cancer.

In a paper published in the July/August issue of Toxicologic Pathology, postdoctoral fellow Dr. Shannon Payne and Dr. Chris Kemp of the Human Biology and Public Health Sciences divisions propose that a strain of cancer-susceptible mice could reveal much about how environmental exposures such as chemical or dietary factors contribute to tumor development. The knowledge could have an impact on public health, since it has been difficult to prove directly the development of cancers through epidemiological studies of populations exposed to suspected carcinogens.

"It's been a challenge for epidemiological studies to detect direct causal connections between exposure to chemical A and development of cancer B," said Kemp, who co-directs the University of Washington Comparative Mouse Genomics Center, a consortium that focuses on how cancer develops from interactions between genes and the environment. "What we'd like is to develop a test system that can realistically assess these types of suspected connections."

First discovered in 1996 by Drs. Matt Fero and Jim Roberts in the Basic Sciences Division, the cancer-susceptible mice have led to numerous advances in cancer biology (Fero is now an investigator in the Clinical Research Division). The strain is much more prone than normal mice to develop tumors when exposed to environmental insults such as radiation or certain chemicals due to a defect in a protein-known as p27-that is critical for control of cell division.

The mice are susceptible to a striking array of tumors, Kemp said. "The p27 protein acts sort of like a rheostat for cancer development," he said. "Reducing the amount of p27 seems to accelerate whatever process has initiated the cancer.

"Just about every organ can be affected, depending on the nature of the environmental exposure. I have an interest in toxicology, so this observation was extremely intriguing and may have practical benefit of identifying potential cancer causing compounds prior to human exposure."

The p27 protein is one of several proteins known as tumor suppressors that normally act to prevent cells from dividing. Reduction in the amount of p27 protein in cells has been found in all types of human cancers and is associated with more aggressive types of breast, colorectal, gastric and other tumors. Those observations form the basis for a prognostic test-developed by Roberts in collaboration with Dr. Peggy Porter of the Human Biology and PHS divisions-that predicts the aggressiveness of breast tumors.

Kemp and Payne propose that the p27-deficient mice could help to improve current methods used to determine whether or not a chemical agent is associated with an increased risk of cancer, which typically has been analyzed in two-year studies in normal rodents.

"These are difficult questions to address in humans," Payne said. "It's hard to obtain adequate sample sizes of exposed individuals and there is also the complication that human populations are outbred-that is, they have many genetic differences that make it hard to draw generalized conclusions about the effects of a particular exposure on disease risk."

Many researchers believe that the number of animals studied-and the time needed to conduct such research-could be significantly reduced by studying cancer-susceptible strains that have mutations analogous to those found to occur in humans with cancer.

Cells typically must accumulate several mutations in critical genes before cancer develops. Payne said that mice-or humans-that are born with mutations in tumor suppressors like p27 have already crossed the threshold of the cancer pathway.

"Such animals already have sustained 'one hit' in the multi-step process, so the time for a tumor to develop is much shorter," she said.

Mice with defects in other tumor-suppressors, such as the well-known p53 gene, have been used as models for studying how and whether chemicals cause tumors. But Payne and Kemp argue that the p27 mice could be better models for studying the process because the p53-deficient mice develop cancers that are not analogous to the majority of cancers that affect humans.

"The p27 model has the potential to help us learn more about exposures for which data are equivocal," Kemp said. "Ultimately, it could have important implications for federal regulatory agencies that deal with protecting the public from carcinogens-and for public health."

 


p27: The history of a protective protein

The p27 protein, one of several known tumor suppressors, is required to keep cells from dividing uncontrollably. Mice that lack one or both copies of the p27 gene are predisposed to tumors in multiple tissues when they are exposed to radiation or chemical carcinogens.

Center discoveries related to p27 include:

  • In 1992, Dr. Jim Roberts laboratory discovered the p27 protein, which acts in concert with a partner protein, cyclin E, to control cell division.
  • In 1996, Dr. Matt Fero and Roberts developed a strain of mice lacking normal levels of p27, which has served as an important model system for studying cancer development.
  • In 1997, Roberts, in collaboration with Dr. Peggy Porter, demonstrated that aggressiveness of breast tumors correlates with a tumor's levels of p27 and cyclin E. The findings form the basis for a prognostic test.
  • In 1998, Kemp and Roberts showed that mice lacking normal levels of p27 are susceptible to tumors when exposed to carcinogens.
  • In 2002, Dr. Bruce Clurman's lab identified a set of potential cancer genes that act in concert with p27 during tumor formation.
Fred Hutchinson Cancer Research Center is a world leader in research to prevent, detect and treat cancer and other life-threatening diseases.