For the first time, researchers have converted fully-differentiated cells in vivo into another type of cell without first reprogramming them to a pluripotent state. The conversion of pancreatic exocrine cells into fully functioning beta cells in living mice is described in a paper to be published tomorrow (August 28) in Nature.

"This paper is an important milestone on the road that hopefully leads to the generation of new beta cells from other, more available cell sources, at will," Harry Heimberg, from the Vrije Universiteit Brussels, who was not involved in the study, told The Scientist in an Email.

The researchers, led by Harvard Stem Cell Institute co-director, Douglas Melton, gave mice pancreatic injections of a combination of three transcription factors -- Ngn3, Pdx1, and Mafa -- to "flip" the cells into full-fledged beta cells. To find out which transcription factors were needed to reprogram the exocrine cells, the team first tried several combinations of nine transcription factors known to be developmentally expressed in the embryonic pancreas. They found that only three were required to convert the cells in beta cells.

Within three days after injection, the cells were converted into beta cells. The team confirmed this with genetic lineage analysis. The conversion was efficient in 20% of pancreatic cells -- a rate nearly 100-fold higher than methods used to create reprogrammed pluripotent (iPS) cells, Melton said in a teleconference. The new beta cells also secreted insulin and lowered blood sugar levels in the mice. But the big surprise, Melton added, was "how easy it was to do it. Of course for the people who did the lab work it wasn't all that easy -- it was a three-year project -- but it's not as hard as you might think to convert" one cell type to another.

Although exocrine cells are much more abundant than beta cells in the pancreas, they are far less easy to collect than fibroblasts, the popular cell source for making iPS cells, Heimberg said. "It would therefore be interesting to find out whether other cell types, more accessible than those in the pancreas and with a high capacity to differentiate, like intestinal stem cells, would be re-programmable without the need to be de-differentiated to an embryonic stem cell-like phenotype."

What's especially promising about the study, said Melton, was that the researchers delivered their transcription factors using adenoviruses, which don't integrate into the host DNA and are already used in some clinical trials. (iPS cells so far have only been created using retroviruses.) In addition, the genes used to reprogram the exocrine cells are not oncogenic, unlike those used to create iPS cells. "I think this approach could be broadly applicable," Melton said. "It represents a parallel approach [to somatic cell nuclear transfer and iPS cells] to make cells for regenerative medicine."