New Insight Into Cancer Metastasis

"As a result, cancer cells acquire in one fell swoop many of the abilities they need to execute the complex stages of metastasis," says Weinberg, who also is a professor of biology at MIT.

Metastasis is a highly inefficient, multi-step process that requires cancer cells to jump through many hoops. The cells first must invade a nearby tissue, then make their way into the blood or lymphatic vessels. Next they must migrate through the bloodstream to a distant site, exit the bloodstream, and establish new colonies. The entire operation involves so many steps that it raises an obvious question: How do cancer cells cobble these behaviors together and acquire the ability to do all this? According to the new study, they don't. Rather, they hijack an existing cellular process and use it to disperse throughout the body.

Reporting in the June 25 issue of the journal Cell, the research team headed by Weinberg describes how a breast carcinoma in mice misappropriates a protein called Twist. Twist is a gene regulator, meaning that it tells genes when to turn on and when to turn off. But Twist is mainly active in early embryonic development, where it enables cells to move from one part of an embryo to another and allocate these cells into different tissues. As an embryo develops, Twist's functions no longer are necessary, and it soon becomes dormant in most tissues throughout the rest of an organism's life.

Through a process that still is somewhat unclear to researchers, tumor cells reactivate this long-dormant protein and thereby acquire the ability to move throughout the body.

The research was conducted in two phases. First, Whitehead postdoctoral fellows Jing Yang and Sendurai Mani compared metastatic and non-metastatic cancer cells taken from mouse tumors. Using microarray technology to determine levels of gene expression, they found certain genes that were active only in the metastatic cells. The gene that stood out among all the others was one coding for the protein Twist.

Next, Yang isolated highly metastatic cancer cells and disabled the Twist gene. When she injected these cells into mammary glands of mice, the mice developed primary, localized breast tumors--but the tumors were unable to metastasize. Furthermore, their study showed that Twist caused breast cells to separate from one another, losing their cell-cell adhesion and scattering, thereby allowing these cells to travel to distant organ sites, like dandelion seeds scattered in the winds.

"Twist is probably the first gene regulator that has been tied so definitively to human cancer metastasis," says Yang.

Andrea Richardson, a pathologist at Brigham and Women's Hospital and coauthor on the paper, correlated these finding with her lab's data taken from human breast cancer studies--essentially re-analyzing her lab's data in light of Yang's. She found that Twist is highly expressed in invasive lobular carcinoma, a unique type of breast cancer where breast tumor cells completely lose their cell-cell adhesion and infiltrate other tissues.

"In many mouse studies you have great models and come up with something and it cures the mice, but then it never seems to work in people," says Richardson. "In this case we were seeing the exact same phenotype and gene expression correlation in human breast tumors."

Although clinical applications of this research are still unclear, Richardson can see the potential for developing a Twist inhibitor, a drug that wouldn't kill a tumor but rather stop its metastatic capabilities. "Something like that would turn cancer into a chronic disease, rather than a deadly one," she suggests.

Yang also can see potential to apply these findings diagnostically. "With breast cancer patients diagnosed with a primary tumor, it's hard to tell whether or not that tumor will generate metastases," she says. "Some small tumors will metastasize, some large ones won't. Here we might be able to discover if a tumor is ready to invade and metastasize or not by finding out if the Twist gene is on or off."

For now, the discovery of Twist is only the beginning. Says Weinberg, "There are a number of other regulatory proteins that have been studied in other labs and have properties very similar to those of Twist. The other regulators undoubtedly will play important roles in other types of human metastatic cancer."