The researchers say their discovery demonstrates the importance of the tumor's environment and shows more precisely how the metastatic process occurs and might be stopped. Their study appears in the January 10 issue of Developmental Cell.
"What actually kills in cancer is not the primary tumor--it's metastasis," says senior author Ross L. Cagan, Ph.D., associate professor of molecular biology and pharmacology and a researcher with the Siteman Cancer Center at Washington University School of Medicine and Barnes-Jewish Hospital. "You can't study that in a laboratory dish. You have to look at the tumor cells in their natural environment--surrounded by normal tissues."
To do this, the research team created tumors in fruit fly eyes and wings that permitted them to study the behavior of individual tumor cells.
"We found that the tumor cells in direct contact with normal cells had a different behavior than cells further inside the tumor," says lead author Marcos Vidal, Ph.D., research associate in molecular biology and pharmacology. "They were exclusively the ones that tended to leave the tissue."
The tumors were generated by turning off an inhibitor of a major oncogene called Src (pronounced sarc), making the tumor cells high in Src oncogene activity. (An oncogene is a gene that increases the malignancy of a tumor cell.) This particular genetic change is common in human breast tumors.
Boundary tumor cells were shown to lose surface proteins that attach them to other cells and stabilize their position within tissues. T he study demonstrated that it was the difference in Src activity that led to the change in the attachment proteins. When a high Src cell (tumor cell) was next to a low Src cell (normal cell) the attachment proteins changed their characteristics, and the high Src cell became "unglued."
In addition, this change sent a signal that activated several other proteins in the tumor cell, one of which was an enzyme that dissolves the matrix surrounding cells. This enzyme makes it possible for a cell to move through tissues.
"Even though all the cells in the tumors we created were genetically identical, the proximity of the boundary cells to normal cells--their interaction with normal cells--made them special," Vidal says. "This is the first time the epithelial environment has been shown to play a role in metastasis."
The cells that left the fruit fly tumors eventually succumbed to the natural process of programmed cell death and were eliminated. According to Cagan, that was not unexpected.
"In a tumor, probably 99.99 percent of the border cells are raining out of the edges and dying," Cagan says. "But as oncologists have found, cancer stems from an accumulation of genetic mutations. If one of these wandering cells acquires a second mutation that prevents cell death, it could go on to establish a metastatic tumor."
Having created a model for studying metastasis of tumor cells, the research team has begun to look for ways to manipulate boundary cells to prevent their metastatic behavior. They have seen that disabling some of the genes in the pathway activated in boundary cells stops the cells from leaving the tumor.
Cagan's laboratory also has developed a robotic system for screening anticancer drugs, and they plan to use this system to look for drugs that will affect the metastatic process in their fruit fly model.
"A drug that can prevent metastasis would be an important adjunct for cancer treatments," Cagan says. "It co uld cut a patient's risk of having tumor cells leave the area before the primary tumor was eradicated. That's essential--metastatic cancer is far harder to treat than early-stage tumors."