As cells specialize, they lose their natural abilities to form new cell types. At least, that's what everyone expects. Now, work from Molly Weaver and Mark Krasnow of Stanford University defy these expectations, demonstrating that differentiated cells within fruit flies naturally return to a flexible, multipotent state as larvae grow into adults. This might point to ways to reprogram human cells to more flexible states without requiring genetic modification.

During fly metamorphosis, most larvae cells die. Before this publication, none of the differentiated cells that survive were thought able to produce new daughter cells, let alone new tissues. Instead, all tissues within the winged adult were thought to regrow from unspecialized "imaginal cells" residing in clusters within the larvae. A fly's breathing apparatus, the trachea, grows from imaginal cells that are found on only one side of a structure known as the dorsal trunk.

By genetically modifying tracheal cells to glow green at a certain point in their development, the researchers hoped to watch imaginal cells migrate across the trunk to form the far branch of the trachea. That didn't happen. Though the imaginal cells do form the near branch of the trachea; they did not give rise to another branch on the far side of the trunk.

Puzzled, the researchers used genetic marking tools to observe other cell types and their progeny. When they did so, they found that one branch of the trachea is re-built from a surprising source of cells. Rather than dying, more-specialized cells from larval tracheal tissue revert to a less-specialized state. Originally tube-shaped, these cells resume a typical round form, and then redifferentiate into at least three cell types1.

Those observations mean, weirdly, that a single tissue is naturally remodeled by two sources of multipotent cells: the long-known imaginal cells, and highly-specialized cells that revert and respecialize.

Linheng Li, who studies the establishment of mammalian niches at Stowers Institute in Kansas City, Missouri, says the work will “add to the debate as to whether every tissue has its own stem cells.” Established work has shown that some tissues, notably liver and pancreas, heal because differentiated cells reactivate and self-renew in response to injury. The function of these cells (i.e. hepatocytes and beta cells) could be comparable to those in the fly's dorsal branch. The fact that the fly trachea is built from two sources of cells, says Li, should lead researchers to search for similar possibilities in mammals, both bona fide stem cells and so-called facultative stem cells.

That search could also lead to new applications. Already, technologies to revert differentiated cells to a less specialized state are common to the breakthroughs behind Dolly the sheep and induced pluripotent stem cells. Now, the tiny fruit fly offers a natural, tractable system in which to study it.