BY KRISTIN WEIDENBACH
By looking at which genes are switched on and off in a variety of cancers, Stanford researchers are learning what thousands of our genes do and why their behavior changes in malignant cells. Using the so-called DNA chip, the researchers can study tens of thousands of genes at the same time and see how they vary in different tissues and how they respond to different drugs.
Scientists in the laboratories of Patrick Brown, MD, PhD, a Howard Hughes investigator and associate professor in biochemistry, and David Botstein, PhD, professor of genetics, have mastered the technique of applying thousands of genes to a small piece of glass and deciphering the patterns that appear when they expose the genes to genetic material from human cells. Fluorescent tags make the genes glow red when they are highly active and green when they are idle. A computer program then sorts all the genes into color patterns that can be interpreted by the researchers.
Doug Ross, MD, PhD, a postdoctoral fellow in Brown's laboratory, examined 8,000 human genes in 60 different cancers including leukemia; cancers of the central nervous system; renal cell and non-small-cell lung carcinoma; melanoma; and ovarian, breast, colon and prostate cancer. Cells from these cancers have been cultivated in the laboratory and comprise a cell panel constructed by the National Cancer Institute (NCI) to test new anti-cancer drugs.
Ross, Brown and Botstein with their collaborators in the laboratory of John Weinstein, MD, PhD, at the NCI, describe the results of their studies in two papers published in the March 1 issue of Nature Genetics.
Scientists already know what some of the genes on the chip do; however, approximately half of them merely have a name or a code number and their function is unknown. When Ross and his colleagues analyzed groups of genes that are switched on or off in a similar way, they found that the groups often contained a mixture of known genes and unknown genes.
"The expression patterns can give you some information about the roles unknown genes might play," said Ross. "Those with the same pattern are very likely involved in the same function."
The researchers saw many complex patterns of gene expression among the cells, but it was clear to them that the predominant pattern corresponded to the tissue from which the cancer originally derived. For example, a large group of genes highly expressed in melanoma cancers contained many genes with known roles in melanocyte biology, and leukemia cells were distinguished by a small set of genes specific to white blood cells.
Ross's collaborator, Uwe Scherf, PhD, a postdoctoral fellow at the NCI, then looked at how the genes' expression patterns related to the cancer cells' sensitivity to different drugs. Researchers at the NCI have compiled a database that records the sensitivities of the 60 cancer cells to each of 70,000 chemical compounds. Scherf analyzed how the gene expression patterns correlated to the cancer cells' sensitivities to a subset of these compounds all the chemotherapy drugs commonly used to treat cancer.
"Since the variation in the way the drug works is determined by the gene expression patterns, we should be able to learn something about the role of different genes in determining the sensitivity of the cell to different compounds," said Ross.
An example of the type of information the researchers are trawling for is provided by a drug called L-asparaginase and a gene called asparagine synthetase (AS). Certain malignant cells lack this gene and are therefore more sensitive to the drug. Physicians already know that acute lymphoblastic leukemia cells are killed by L-asparaginase. When the researchers looked for the AS gene they found it was expressed at a very low level in the acute lymphoblastic leukemia cells and expressed at a high level in cells from a different kind of leukemia one relatively resistant to the drug. The researchers can now apply what they know about the AS gene expression pattern and L-asparaginase sensitivity to other cancer cells, hoping to find a matching pattern that will identify other cancers susceptible to this drug.
"We can characterize gene expression in different cell lines and we can characterize drug sensitivity in different cell lines, and correlate the two. We hope that will teach us something about the mechanisms of drug sensitivity," said Ross.
Members of the Brown and Botstein labs are continuously expanding their DNA chip, which is known scientifically as a DNA array. The arrays they are currently working with hold more than 20,000 genes, and the researchers are producing more than 400 arrays each month to meet the demands of the lab members and their Stanford collaborators. Soon they hope to be squeezing 40,000 genes onto each chip.
Other Stanford authors who contributed to the studies include Michael Eisen, PhD, now at the University of California, Berkeley, who contributed to the study when he was a research fellow in Botstein's lab; Charles Perou, PhD, and Christian Rees, PhD, postdoctoral fellows in Botstein's lab; Paul Spellman, PhD, and Alexander Pergamenschikov, a graduate student and research assistant in Botstein's lab; Vishwanath Iyer, PhD, a postdoctoral fellow in Brown's lab; Stefanie Jeffrey, MD, assistant professor of surgery; and Matt Van de Rijn, MD, PhD, assistant professor of pathology.
The research at Stanford was
supported by the NCI and the Howard Hughes Medical Institute. Ross
is a Walter and Idun Berry Fellow, Eisen is an Alfred E. Sloan
Foundation Fellow and Perou is a SmithKline Beecham Pharmaceuticals
Fellow. SR