Embargoed for Release May 13, 1999,5 p.m. Eastern Time
Contact: Todd Ringler, (617) 632-5357
STUDIES FIND THAT GENETIC CHECKPOINTS AGAINST CANCER ARE SOMETIMES FRIEND, SOMETIMES FOE
BOSTON -- In a pair of studies that promise to rewrite scientists' understanding of genetic "tripwires" that kill cells before they become fully cancerous, researchers at Dana-Farber Cancer Institute have found that the same system that protects against cancer can, in some circumstances, actually promote cells on the road toward malignancy.
The studies published in the May 14 issue of Cell, shed new light on cells' built-in safeguards against cancer, but also promise to complicate the search for therapies that seek to quell cancer by targeting key structures within cells. The research was performed in the laboratory of Ronald DePinho, M.D., of Dana-Farber and professor of Medicine (Genetics) at Harvard Medical School.
"The studies confirm much of the conventional thinking about the nature of cells' natural defenses against cancer," says co-lead author Lynda Chin, M.D, of Dana-Farber and assistant professor of Dermatology at Harvard Medical School. "But they also show that some cancer cells have the ability to evade these defenses in ways that make them extraordinarily difficult to target with single therapies."
The studies deal with a key landmark in the life and death of cells - a point known as "crisis." Crisis occurs when cells are near the end of their natural life. Their lifespan is controlled by structures called telomeres, which are repetitive DNA strands that sit at the ends of chromosomes and protect them from damage. They serve as a meter of cell division, because each time a cell divides, it loses a little bit of its telomeres. When the telomeres become short enough - usually after 60-or-so divisions - they send a signal to the cell to stop dividing. This point is referred to as the "Hayflick Limit."
The Hayflick Limit is believed to serve as a checkpoint against cancer: by arresting the growth of cells, it prevents them from proliferating out of control, a hallmark of cancerous cells.
Sometimes, a genetic mutation enables cells to bypass the Hayflick Limit and continue to divide. With each division, the telomeres get shorter and shorter until they are severely eroded. With the loss of telomeres and their chromosome-protecting function, the chromosomes start fusing with one another, breaking and rearranging, resulting in a scrambling of the cells' genetic programming.
At this point, cells somehow sense that something has gone awry and activate a suicide program - called apoptosis - to kill themselves. It is here that cells are considered to be in crisis.
It was assumed that the cell-suicide mechanism ensures that abnormal cells that are growing out of control do not pass on their scrambled genetic programs to their offspring. "Cells are programmed to sacrifice themselves so that genetic abnormalities are not perpetuated," says Chin.
Since most cells die at crisis, it has been theorized that crisis serves as the ultimate "brake" against cancer. To test whether this is indeed the case, Chin, her co-authors Steven Artandi, M.D., Ph.D., and Roger Greenberg, of Dana-Farber and Harvard, and their colleagues bred successive generations of mice that lacked the ability to rebuild their telomeres and lacked one of two key tumor suppressor genes, p53 or INK4a. p53is the most common tumor suppressor gene to be found in mutated form in human cancers. The INK4a gene is second only to p53 in its involvement in human cancers. Losing either one of these genes predisposes an organism to develop cancers.
"We essentially removed one brake against cancer - p53 or INK4a - in order to determine whether crisis itself is also a brake," Chin says. The combination of a disabled telomere-repair mechanism and a lack of p53 or INK4a made it easier for cells to barge past the Hayflick Limit and head straight into crisis.
With each successive generation of mice that lacked INK4a, the percentage of mice with tumors decreased, as researchers had expected. In the first generation of "crisis" mice, 64 percent of the animals developed cancer; in the third generation, 50 percent did; and in the fifth generation, 31 percent did. At the same time, the survival rate of the mice rose with each generation - from 12 percent in the first generation, to 54 percent in the fifth.
"This represents the first proof that telomere-shortening during crisis inhibits the growth and creation of tumors," Chin says. "Without intact telomeres, cells have a harder time dividing and proliferating."
But if crisis is indeed a tumor suppressor, it is an imperfect one, the study found. While successive generations of mice had fewer tumors, some tumors did develop. Within those tumors, researchers found a high frequency of chromosomal fusions. The authors speculate that these fusions help preserve the integrity of the genetic materials within the chromosome.
"It's as if one had a pair of shoelaces that had lost the plastic caps at the ends," Chin says. "Tying the ends together would prevent them from unraveling. Cromosomal fusion provides cancer cells with a way of evading the death that crisis would normally bring about," she continues. "Our results show that while the loss of telomeres in cancer-prone mice impairs tumor formation, it does not prevent it."
Although the first study on INK4a and telomeres indicated that crisis, imperfect as it may be, did provide a brake against cancer development, the parallel study on p53 revealed some unexpected results, adding a new wrinkle to those of the first.
p53, known as the "guardian of the genome," is known to trigger apoptosis in response to genetic damage in a cell - a way of preventing genetic errors from being passed on to the next generation. It had been theorized that the erosion of telomeres during crisis represented such a trigger: telomere loss would activate p53 and set cell suicide in motion. The researchers found that this is indeed the case.
"This demonstrated for the first time that p53 responds to the erosion of telomeres in the same way that it responds to other kinds of DNA damage: by initiating the process of cell death," says co-author Steven Artandi.
When the researchers turned their attention to the most recent generation of p53-deficient mice - eight generations removed from the first set of "crisis" mice - they identified a stage of cancer development that they call "genetic catastrophe," a condition which Artandi describes as a "genetic furnace."
At this stage, cancer-prone cells are so genetically unstable as a result of the crumbling of telomeres, that the cells have only two courses open to them: death or further progress toward cancer. Cells that are unable to adapt to the disruption of their genetic material die. Cells that can adapt - by rebuilding their telomeres, by fusing their chromosomes in ways that enable them to still function, or by other unknown, responses - continue to divide relentlessly, and are one step closer to becoming fully malignant.
"This study shows that therapies that seek to arrest cancer by preventing cells from rebuilding their telomeres may be less effective in cells that lack functioning p53," Artandi says. "Telomere-targeting therapies will need to be studied in tumors that are p53-competant and those that are p53-deficient. This work, by showing the complexity of the mechanism for keeping cancer cell growth in check may result in more effective therapies that attack the problem from several directions at once," he continued. "A multi-pronged approach increasingly seems to be the one that's most likely to be effective."
In addition to Chin, Artandi, and Greenberg, other researchers who contributed to this work include: Qiong Shen, M.D., Alice Tam, Shwu-Luan Lee, Ph.D., and Kee-Ho Lee, Ph.D., of Dana-Farber and Harvard; Geoffrey J. Gottleib, M.D., of Quest Diagnostics in Teterboro, N.J.; Carol W. Greider, Ph.D., of Johns Hopkins University School of Medicine; and Andrea Femino, Ph.D., and Robert H. Singer, Ph.D., of Albert Einstein College of Medicine in New York City.
RSS