Science Editor

Most scientists would think of stem cells as being on one end of a one-way street. A series of cell divisions can lead a stem cell down the path to a more restricted progenitor cell and ultimately, mature cell types. But a specific cell type cannot return spontaneously to its stem-like state.

But researchers from the Whitehead Institute recently found that that some breast cells can turn back into stem cells.

More disturbingly, at least from a medical point of view, when those breast cells were transformed into cancer cells with an oncogene, they could revert into cancer stem cells; in fact, the addition of the oncogene "enhanced their ability to spontaneously transition to a stem cell state," Christine Chaffer told BioWorld Today. Chaffer is a researcher at the Whitehead Institute and the first author of the paper reporting the findings, which appeared in the April 11, 2011, online issue of the Proceedings of the National Academy of Sciences.

The discovery, Chaffer said, was "a little bit serendipitous." It began with Chaffer's observation that when she cultured breast cells, most of them stuck to the culturing dish, but a few floated. The fact of floating cells was not in itself that unusual, but their state was: "They actually looked healthy and alive – which was strange," Chaffee said.

Chaffee and her colleagues started testing the cells to see how they were able to survive without sticking to their substrates, which is usually the kiss of death for cultured cells. And during the course of those studies, they found that some of the floating cells were stem cells – and some were not originally stem cells, but could give rise to stem cells.

Chaffee and her team next looked at whether, if such normal cells became cancer cells, that would be the end of their powers of transformation – or just the beginning. The team added an oncogene to the cells, and transplanted them into mice. If such cells were not able to turn into stem cells after they transformed, such cells should not be able to seed tumors; but they were. When Chaffer and her team injected oncogene-containing cells into mice, they formed tumors. And from those tumors, the team was able to isolate cells with the characteristics of cancer stem cells.

Medically speaking, the findings could be both good and bad news. In recent years, the notion that cancer stem cells are behind many cases of relapse has become increasingly prominent. Chemotherapies target rapidly dividing cells, the theory goes. But they are ultimately doomed to fail because they do not eradicate stem cells, which then go on to repopulate the tumor.

The new results suggested that stem cells are a moving target, since even if stem cells were completely eradicated at one time, regular tumor cells could turn into stem cells, leaving a patient back at square one.

"It's not going to be enough to target the cancer stem cell," Chaffee said. Cancer stem cells will need to be targeted in addition to regular chemotherapies, not as a replacement for them.

More generally, she said that her team's findings do imply that there is "a much larger pool of cells that could give rise to a cancer stem cell," which certainly makes both the task of eradicating a tumor and preventing metastases more challenging.

On the plus side, if cells can indeed return to a stem-cell state, and do so spontaneously, then harvesting such cells could possibly be a source for patient-specific stem cells. The relative advantages and disadvantages of spontaneous stem cells relative to induced pluripotent cells remain to be seen. Certainly, recent reports that induced pluripotent cells have high mutation rates and differences in methylation have been a reminder that for all their promise, iPSCs are still in their infancy.

On the other hand, it is unclear whether spontaneously converted stem cells can be a match for iPSCs in terms of the efficiency of generating them. Chaffee said that her team wants to test whether other cell types can also revert into stem cells, and whether the rate of such conversion is influenced by the microenvironment – becoming more frequent, for example, after an injury. But to date, the work is all on cultured cells. "It's something we haven't been able to demonstrate in vivo as yet," Chaffee said.