News Release

Fly cells on the move may reveal clues to cancer metastases

Peer-Reviewed Publication

Johns Hopkins Medicine

Using neat genetic tricks with fruit flies, scientists from Johns Hopkins School of Medicine have found the key signal that allows a group of normally stationary cells in the ovary to travel, they report in the current issue of Cell.

Because fruit fly genetics are similar to but vastly simpler than human genetics, understanding the signals that mobilize the ovary cells may help clarify how human cancer cells invade distant tissues, says Denise Montell, Ph.D., associate professor of biological chemistry in the school's Institute for Basic Biomedical Sciences.

"Cells usually hold on to their neighbors, so a lot of things have to change for a cell to become migratory," says Montell, who has been studying this process in flies for 12 years. "We've now found the first signal that seems to be sufficient to get cells moving."

With funding from the American Cancer Society and the National Institutes of Health, graduate student Debra Silver examined thousands of mutant fruit flies and identified a protein called "Unpaired" that activates stationary cells.

The fruit fly ovary consists of about one hundred egg chambers, each made of 16 cells -- 15 "nurse" cells and one oocyte, which becomes the egg -- surrounded by a layer of several hundred epithelial cells. At a certain point in development of the egg, a single small cluster of these epithelial cells detaches from the others and moves from the edge of the egg chamber to the center, sliding between nurse cells and coming to rest at the edge of the oocyte.

The cluster consists of two interior "polar" cells and a covering of four to eight "border" cells. Other researchers have shown that, while key to the migratory process, the polar cells can't move without the help of border cells. But until now it wasn't known how polar cells recruit border cells to be their vehicle.

Montell and Silver suspected the polar cells might send an initial, short-range biochemical signal, in the form of a molecule, to nearby cells. However, standard gene knockouts, in which a particular gene of interest is eliminated, were unlikely to reveal these answers because the missing gene would probably kill the fly.

"Genes involved in cell migration are very important," says Montell. "They are probably involved in early embryo development."

Instead, Montell's lab used mosaic flies, in which the vast majority of cells are normal, but a scattered subset of cells essentially lacks the gene of interest. "The knockout patches are large enough to reveal the effect of the missing gene but not so large to affect the organism's overall survival," says Montell, who has used the technique for the past three or four years to study different aspects of the cells' migration.

To make mosaic flies, the scientists induce some of the epithelial cells to misappropriate their genetic material as they divide. Before dividing, the cells each have one regular copy and one abnormal copy of the gene of interest, and during the process of dividing, they copy both.

However, thanks to the scientists' trickery, instead of getting one copy of each, the new cells end up with two normal copies or two abnormal copies. As the double-abnormal cells divide, they create mutant patches in an otherwise normal animal.

With the help of the mosaic flies, Silver discovered that polar cells make and release Unpaired, the only known trigger in the fruit fly of the so-called JAK/STAT pathway, which was already known to help regulate cell division and cell survival in both fruit flies and humans.

Upon leaving the polar cells, Unpaired attaches to its receptor on nearby epithelial cells, turning on the protein JAK (Janus kinase), which builds a docking point for another protein called STAT (signal transducer and activator of transcription). STAT can then enter the nucleus and activate certain genes.

Not only do polar cells normally make Unpaired, Silver found that any epithelial cell that produces Unpaired will be whisked across the egg chamber by its neighbors. "Releasing Unpaired and turning on the JAK/STAT pathway in neighbors is all that's needed for the cells to start moving," says Montell.

The next step is to examine ovarian cancer tissue to see if the JAK/STAT pathway regulates the mobility of human cancer cells. In general, tumors aren't particularly dangerous until they begin moving, a process called metastasis. While much is understood about how cancer cells travel (i.e. through the lymph system), it's not well understood how they let go of the original tumor or lodge elsewhere.


Related Web site:

Media Contact: Joanna Downer, 410-614-5105, Email:

Johns Hopkins Medical Institutions’ news releases are available on an EMBARGOED basis on EurekAlert at, Newswise at and from the Office of Communications and Public Affairs’ direct e-mail news release service. To enroll, call 410-955-4288 or send e-mail to

On a POST-EMBARGOED basis find them at and Quadnet at

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.