However, a new report from a team of Vanderbilt-Ingram Cancer Center scientists demonstrates that tumors can develop from completely normal epithelial cells solely because of changes in signals from nearby supporting cells.
Writing in the Feb. 6 issue of Science, the team reports the dramatic development of precancerous prostate lesions and invasive cancers in the forestomachs of mice bred to eliminate a particular cell signaling pathway only in fibroblasts. These fibroblasts are supporting cells adjacent to the epithelial cells.
"The key finding is that we were able to show that a signaling pathway in the fibroblasts is important and can have a dramatic effect on epithelial cells in these animals," said Dr. Harold L. Moses, Benjamin F. Byrd Professor of Oncology, director of the Vanderbilt-Ingram Cancer Center, and director of the Frances Williams Preston Laboratories of the T.J. Martell Foundation for Leukemia, Cancer and AIDS Research. The paper results from a collaboration between Moses' lab and the lab led by Dr. Eric G. Neilson, Hugh J. Morgan Professor and Chair of Medicine.
The Moses lab focuses on the role of transforming growth factor beta and its ability to either stimulate or inhibit cell growth depending upon the conditions. A standard approach to studying a particular protein or signaling pathway is to develop a genetically engineered "knock-out" mouse that does not express the particular protein in question. But TGF beta and its receptor are so critical to mammalian development, mice bred to eliminate this signaling pathway in all cells die as embryos.
To eliminate the TGF-beta pathway in only certain cells, the Moses lab in collaboration with Dr. Mark Magnuson's and Dr. Christopher V.E. Wright's laboratories developed a mouse in which the TGF-beta receptor included recognition points for a gene called Cre and its protein product. In cells where Cre is expressed, the TGF-beta receptor is selectively turned off.
The Neilson lab focuses on fibroblasts and had recently reported the potential of fibroblast-specific-protein 1 (FSP1) as a marker for cancer metastasis. This lab had developed a mouse that, under influence by FSP1, selectively expressed Cre in fibroblasts alone.
By crossing the two strains of mice, a new strain of animal was developed that expressed Cre only in the fibroblasts and so, in turn, eliminated TGF-beta expression only in the fibroblasts. The other cells in the mice were completely normal.
"We had developed the FSP1.cre mouse because we wanted to study selective gene interruptions in fibroblasts," Neilson said. "About the time our lab finished breeding our mouse, Hal and I realized that we would have an interesting experiment by crossing the two. At the time we decided to do the experiment, we weren't thinking the mice would develop cancer."
But that is exactly what happened. The mice consistently developed precancerous lesions in the prostate and invasive squamous cell cancer in the forestomach (an organ in the mouse similar to the human esophagus).
Upon additional analysis, the researchers noted in the mouse forestomachs an elevation of a receptor for hepatocyte growth factor (HGF), which is known to be regulated by TGF-beta.
The findings suggest that when the TGF-beta pathway is active in the fibroblasts, they provide signals that keep growth of the neighboring epithelial cells in check; when the pathway is interrupted, cell growth is allowed to proceed unchecked, a result that may be related to an increase in HGF activity. "The seminal observation from crossing these mice is that you don't need to have mutations in the epithelial cells for a carcinoma to result," Neilson said.
Dr. Neil Bhowmick, assistant professor of Urologic Surgery, lead author on the paper and a former member of the Moses lab, agreed.
"This really suggests new causes of cancer and targets for intervention," Bhowmick said. "We were looking at TGF-beta because that's what we study. But it's likely that there are other molecules in the stroma (supporting cells) that are important, too, and this opens the door for that research."
Bhowmick, who recently joined the Vanderbilt faculty, said that the project was of particular interest to him because it allowed him to collaborate with many other departments in the medical center. "The partnership was critical because without the mouse from the Neilson lab, we could never have made these observations."
Other co-authors on the paper were Anna Chytil, and Agnieszka E. Gorska of the department of Cancer Biology; David Plieth and Nancy Dumont, of the department of Medicine; and Scott Shappell and Kay Washington of the department of Pathology.
The work was supported by the National Institutes of Health (National Cancer Institute), the U.S. Department of Defense, and the T.J. Martell Foundation for Leukemia, Cancer and AIDS Research.
Journal
Science