The stem cell community has generated incredible news flow and controversy over the past decade (see here), but new data published in Nature by Doug Melton’s group (Harvard) is the most surprising find so far.

Melton’s group altered cells within the body and changed their appearance and function to a completely different cell type. Most importantly, the cells acquired the ability to produce insulin. Consequently, this work heralds the possibility of a new and unexpected approach for treating diabetes.

The prevailing dogma has been that to create a cell with a new function it is necessary to take it back to an early stage of development (e.g. iPS cells) or simply start with an early stage cell (embryonic stem cells). The astounding thing about this latest discovery is that the cells were fully differentiated adult cells that were induced to transform directly into insulin-producing cells without any regression to an earlier stage of development.

The methodology involved injecting the pancreas with adenovirus containing three transcription factor genes (Ngn3, Pdx1 and Mafa). New insulin-producing cells were found outside of the pancreatic islets within three days and they functioned consistently throughout the length of the experiment (three months). These insulin-producing cells were indistinguishable from islet beta cells — the usual providers of insulin.

Transplantation of beta cells has been previously shown to control diabetes (see the Edmonton Protocol), but there are a number of serious limitations with this approach (e.g. need for immunosuppressive drugs, shortage of pancreas donors, duration of treatment effect, etc.).

In Melton’s work it was found that when diabetes was induced, the transformed cells were able to control blood glucose levels to a significant degree (although not quite as well as in non-diabetic animals).

This news has a number of important implications, one of the main ones being that it may not be necessary to use stem cells to generate functional adult cells. However, before getting too carried away a few more questions need to be answered including:

  • Can similar findings be achieved with other cell types?
  • Can this finding be replicated in other species?
  • What would be a safe gene delivery mechanism for human use?
  • Is the genetic control of insulin production a representative case, i.e. will other phenotypes be as easy to develop?
  • Can the induced insulin-producing cells provide sufficient glucose control to halt diabetes?
  • How long will the induced cells remain stable?

Despite these caveats, this work is outstanding and represents a major paradigm shift in the life sciences.

John McCulloch

John provides assistance to life sciences entrepreneurs in business strategy, management, intellectual property, financing and licensing. See more…