COLLEGE STATION - By looking at the cellular mechanisms underlying colon cancer - the second leading cause of cancer death in the United States - scientists at Texas A&M University have gained new insight into how diet affects colon cancer, which could lead to new ways of treating the disease.
The researchers, led by Joanne R. Lupton, a professor who holds appointments in animal science, food science and technology, nutritional sciences and veterinary anatomy and public health and holder of the William W. Allen Endowed Chair in Nutrition, and Raymond J. Carroll, Distinguished Professor of Statistics, Nutrition and Toxicology, have been working for four years to understand why colon cancer seems to be prevented by a fish oil-diet but not by a corn oil-diet.
"Fish oil seems to protect against colon cancer while corn oil not only does not protect against it but appears to promote it," says Lupton, who has spent the last 20 years working on the effect of diet on the colon, "so we are looking for a potential mechanism at the cellular level that would tell us why that might be the case."
Lupton has discovered that eating mostly corn oil or fish oil can affect the fatty acid composition of cell membranes, which in turn can affect whether a cell will become cancerous.
"Once the fatty acid composition of the membranes of colon cells is changed," Lupton says," that sets up a different fate for the cell. A signal can go from the cellular membrane to the nucleus of the cell and tell it either to divide and become a tumor or give up the ghost and undergo programmed cell death."
With help from the group of statisticians led by Carroll, the scientists have also determined for the first time how cancer spreads among cells in the colon. "Looking at how colon cancer damage depends on cell positions is completely new," says Carroll, one of the world's leading experts in statistics.
The scientists performed their studies on rats, which were injected with azoxymethane, a carcinogen known to induce colon cancer, and then the scientists looked at individual cells of the colons of the rats fed either fish oil or corn oil diet.
Cells in the colon replicate and grow within separate groups of cells called "crypts," which are little indentations at the surface of the colon. Within each crypt, the cells closer to the surface are the oldest and the ones at the bottom are the youngest. The scientists studied how cell damage varied from the bottom to the top of each crypt.
"Biologists used to assess the damage of the crypt as a whole or to group the cells in three groups: bottom, middle and top," says Jeffrey Morris, assistant professor of biostatistics at the M.D. Anderson Cancer Center in Houston, who participated in the study as a Texas A&M graduate student. "In this work, we looked at every cell along the crypt, from bottom to top."
The researchers also compared how cell damage spreads at both ends of the colon. "Within the colon, there are different degrees of cancer at different sites," Lupton says. "Indeed, most humans have left-sided tumors rather than right-sided tumors."
If two groups of crypts, one on the left side, the other on the right side, are equally damaged, the damage is likely to be diffuse, explains Morris. Instead, if one side has more damaged crypt cells than the other side, the damage is likely to be more localized and concentrated, which could in principle intensify its effects, he adds.
The scientists found that for a fish oil-rich diet, the young cells at the bottom of the crypts were equally damaged at both ends of the colon of a given rat. "For some unknown reason, the cells at the bottom of the crypts behave differently from the other cells in the middle and the top of the crypts," Morris says.
"Presumably, after the carcinogen is metabolized in the liver, it goes through the bloodstream. There may be more blood vessels in the bottom of the crypts, so carcinogens might get there more systematically than to the middle and top regions of the crypts."
For rats fed corn oil-diet, the scientists found an unexpected result: When the cells of the crypts on one side are heavily damaged, the ones on the other side are only slightly damaged, and vice versa.
"The result was surprising since such negative correlations are rarely encountered in biology," says Morris, adding that fish oil and corn oil diets might create different chemical environments in the colon.
To verify and better understand these findings, the researchers are now designing studies involving a larger number of rats. Though more work will be needed to better understand how diet affects the spread of cancer at the cellular level, Carroll says that such work has been possible only because of the collaborative effort between statisticians and nutritionists. So he decided to set up a program aiming at training statisticians in the biology of nutrition and cancer.
"A lot of times, when scientists from two different fields want to collaborate, the biggest problem is that they do not speak anywhere near the same language," he says. "We spent four years learning one another's language, so we want to teach highly trained statisticians enough biology so that the communication difficulties break down."
The program has been awarded a grant by the National Cancer Institute for six years from July 2001 to June 2006. "This project has the best funding score that any such grant has received," Carroll says. "There is no other program like this in the United States, which might explain why we got such a good score."
The program will provide training courses and seminars to four postdoctoral students, two students starting a doctoral thesis and five graduate students. Seven statisticians and five nutritionists will serve as mentors of the program.
The research training will take place in the Department of Statistics, the intercollegiate Faculty of Nutrition and the Biomedical Imaging group within the Department of Electrical Engineering.
"This program is going to help Texas A&M be part of the molecular biology revolution by raising the visibility of biostatistics and bioinformatics here," Carroll says. "For me it is a huge commitment."
Contact: Raymond J. Carroll, 979-845-3141, carroll@stat.tamu.edu or Joanne Lupton, 979-845-2142, Jlupton@tamu.edu or Jeffrey Morris, 713-794-1720, jeffmo@odin.mdacc.tmc.edu