While diabetes can mostly be managed through a carefully-controlled diet and regular insulin shots, a longer term solution and perhaps even a cure could be found by replenishing a patient's own supply of beta cells, which naturally produce the hormone. Researchers at Washington University in St. Louis are a step closer to that kind of treatment, after tweaking the recipe for turning stem cells into beta cells.

For a while now, scientists have been able to turn skin cells from patients into what are known as induced pluripotent stem cells (iPS cells). These are effectively "blank slate" cells that can then be coaxed to become almost any other type of cell in the body. In this case, that's beta cells, which have in the past been made from iPS cells with mixed success.

"Previously, the beta cells we manufactured could secrete insulin in response to glucose, but they were more like fire hydrants, either making a lot of insulin or none at all," says Jeffrey Millman, principal investigator of the study.

Normally, scientists can make iPS cells differentiate into certain specialized cells by exposing them to specific growth factors at different stages of development. For the new study, the team made tweaks to the previous recipe for converting iPS cells into beta cells, and managed to make ones that function more effectively.

To test how well the new beta cells worked, the team then transplanted them into diabetic mice that couldn't produce their own insulin. Sure enough, within a few days they began secreting insulin at levels that were able to control the blood sugar levels in the animals, functionally curing their diabetes. Better yet, the cells were able to keep doing so for months on end.

"We've been able to overcome a major weakness in the way these cells previously had been developed," says Millman. "The new insulin-producing cells react more quickly and appropriately when they encounter glucose. The cells behave much more like beta cells in people who don't have diabetes."

Of course, as promising as this study may seem it was conducted in mice, so there's no guarantee that it will work the same way in humans. One hurdle is that the mice in the study had to have their immune systems suppressed so they wouldn't reject the new cells – a step that's too dangerous to perform in humans. But, the researchers say, there are a couple of potential ways around this problem.

"The first would be to encapsulate the cells in something like a gel – with pores small enough to prevent immune cells from getting in but large enough to allow insulin to get out," says Millman. "Another idea would be to use gene-editing tools to alter the genes of beta cells in ways that would allow them to 'hide' from the immune system after implantation."

The research was published in the journal Stem Cell Reports.