Now, a new study by scientists at the University of Michigan Comprehensive Cancer Center reveals one crucial key to that deadly process.
In the June 18 issue of the Journal of the National Cancer Institute, the U-M team reports strong evidence that a protein called RKIP, for Raf kinase inhibitor protein, governs the ability of prostate cancer cells to leave their original location and enter nearby blood vessels – which then act as superhighways to the rest of the body.
The findings, made using a broad range of techniques, show that RKIP is vital to this process, called vascular invasion. Such invasion is the first in a cascade of events leading to metastasis. Tumors that produce a normal amount of RKIP appear unable to make the jump to the vascular system, the researchers report. But if a tumor cell lacks RKIP, metastasis can take place.
"The gene encoding RKIP appears to be a novel metastasis suppressor gene, involved in blocking the cell-signaling processes that allow cancer cells to enter the bloodstream," says senior author Evan Keller, DVM, Ph.D., an associate professor of Comparative Medicine and Pathology at the U-M Medical School. "If there is RKIP expression in a tumor, this first important step appears to be less likely."
Because the researchers made part of their discovery using tissue samples from patients treated for metastatic prostate cancer at the U-M Cancer Center, they believe the laboratory findings are especially relevant to "real life" cases of cancer.
They also feel that RKIP could be involved in other forms of cancer. "Understanding the basic biology means we may eventually be able to help patients with any form of disease that depends on RKIP to keep metastasis from happening," says Keller.
The newly published study may eventually lead to tests that could tell dangerous forms of cancer from those less likely to spread, since the presence of RKIP can be detected by a simple tissue-staining method. Or, it could lead to gene therapy to replace RKIP -- and prevent metastasis -- in those whose tumors don't produce it.
But Keller and his colleagues, including first author and U-M postdoctoral researcher Zheng Fu, Ph.D., note that it could be several years until RKIP-based tests or treatments will be available. Continued study of RKIP's role is needed, they say.
The study was funded by the U-M Cancer Center's Stuart and Barbara Padnos Endowed Research Fund, by the Association for the Cure of Cancer of the Prostate, by the Department of Defense, and by the National Cancer Institute through the U-M's SPORE (Specialized Program in Research Excellence) grant for prostate cancer research. The team used the U-M's unique library of prostate cancer tissue samples and sophisticated gene and tissue array facilities.
Keller and his team first suspected RKIP's key role several years ago, when they compared gene expression patterns in two lines of human cancer cells: a non-metastatic line called LNCaP and a metastatic line called C4-2B that was derived from the LNCaP cells. The RKIP gene was expressed far less in the metastatic line, so the team set out to study its role.
The new paper documents three and a half years of their work, using various techniques to thoroughly assess RKIP's function. The resulting evidence may be the first time a metastasis suppressor gene has been studied so thoroughly and the mechanism of its action determined so precisely. About 20 genes involved in the metastatic cascade have been found, but most have yet to be correlated with a specific action or function in the body.
The researchers started by confirming their initial finding of decreased RKIP expression in metastatic cancer cells, by measuring actual levels of the RKIP protein and the messenger RNA that carries instructions for making the protein inside each cell. They found that RKIP levels were three times lower in the metastatic cell line than in non-metastatic cell line.
They also looked at RKIP levels in tissue samples from U-M patients with prostate cancer who had agreed to donate their prostates after death. The men were autopsied within hours of their deaths, yielding tissue samples that still contained the fragile messenger RNA and protein molecules that can degrade quickly after death.
RKIP was detected at high levels in all 10 samples of non-cancerous tissue, and at slightly lower levels in all 12 samples of non-metastatic cancerous tissue. But the protein was not detected at all in any of the 22 samples of tissue from prostate cancer metastases.
Next, the researchers looked at the function of RKIP, in a sort of "gene therapy" approach designed to alter the amount of RKIP produced by the cells. Before and after transferring the genes, they assessed how well cells "invaded" a membrane that mimics a blood vessel wall.
The metastatic cancer cells that had been given extra RKIP showed an average 48.5 percent decrease in their invasive ability, while the same ability more than doubled in the non-metastatic cancer cells whose RKIP production was stunted. "This told us that RKIP somehow keeps cancer cells from invading nearby membranes," explains Keller.
In addition to this in vitro test, the researchers also did an in vivo test by adding an RKIP gene to metastatic prostate cancer cells, and injecting those cells into mice while injecting control mice with regular metastatic cancer cells. Both mice got cancer, and the two kinds of injections resulted in the same sizes of primary tumors. But the mice that received the cells with extra RKIP genes had far fewer metastases than the mice that received regular metastatic cells.
And, RKIP protein levels were far higher in the tumors of mice that got the cells with added RKIP genes, while RKIP was undetectable in the metastases that grew in either group of mice.
The researchers also looked at how well the cancer in the RKIP-enhanced and control mice invaded nearby blood vessels. While invasion was seen in four of the 10 mice that received the cells with extra RKIP genes, the number of blood vessels that were invaded in these mice was far lower than the number invaded in the control mice -- all of which showed invasion.
In addition to controlling the cancer cells' ability to invade nearby blood vessels, the researchers found that RKIP may also play a role in angiogenesis -- the ability of cancerous tumors to grow blood vessels of their own to feed their continued growth. The primary tumors of the control mice, which had the normal levels of RKIP, had higher numbers of new tumor blood vessels than the mice that received extra RKIP genes.
Finally, the researchers looked at the actual cell-signaling pathway that RKIP is involved in -- the Raf kinase system. They found that the cells with lower RKIP levels had higher levels of the activated form of a molecule called MEK. They predicted that reduced RKIP levels pave the way for activation of MEK, which in turn is involved in vascular invasion and metastasis.
The researchers tested their theory by blocking MEK activity in some metastatic cells and blocking the action of other kinases in other cells. The cells with blocked MEK activity showed a much bigger drop in invasive ability than the cells whose other kinases were blocked.
This documentation of an association between RKIP expression and a signaling pathway, leading to metastasis, is the first of its kind, Keller explains. And, the results suggest that inhibiting the pathway involving MEK and its related molecule ERK could prevent metastasis.
As they were conducting their research, the U-M researchers heard of another piece of evidence that helped strengthen the case for RKIP as a metastasis suppressor gene. Members of the International Radiation Mapping Consortium, part of the Human Genome Project, reported that they had found that the chromosome region 12q24 -- the same region where RKIP was known to be situated -- could suppress cancer metastasis.
The discovery of RKIP's role in cancer metastasis fits well with what scientists had already learned about the same protein's role in the brain, where it serves a vital role in binding proteins involved in cell signaling.
Keller notes that vascular invasion -- the "first step" in metastasis apparently governed by RKIP -- is not the only action necessary for cancer to spread in a patient's body. "Many cancer cells that enter the bloodstream don't go on to form successful metastases," he says. But he predicts that the U-M team's discovery will help scientists and clinicians better understand the complex process by which some cancers kill. "These findings bring home the point that if you can stop even one gene in the cascade, you can slow the process down," he adds.
Now, Keller and his team are using tissue microarray techniques to screen for RKIP in other cancer tissue samples.
In addition to Keller and Fu, the research team includes Peter C. Smith, now of Yale University; Lizhi Zhang; Mark Rubin, M.D., now of Brigham and Women's Hospital and Harvard University; Rodney L. Dunn, a biostatistician; and Zhi Yao, a visiting professor at the U-M and dean of research at the Tianjin Medical University in Tianjin, China.
Fu is one of the first graduates of the U-M Program in Immunology, a new interdepartmental degree program through the U-M's Horace Rackham School of Graduate Studies.
Reference: Journal of the National Cancer Institute, Vol. 95, No. 12, June 18, 2003
Note to editors: JNCI will publish an editorial pertaining to this paper, which is available on request. Images from the study will also available.
Journal
JNCI Journal of the National Cancer Institute