This week, San Diego–based Novocell Inc. showed that cell replacement therapy for type I diabetes mellitus seems to work, at least in mice. The Holy Grail for researchers working to treat diabetes this way is a reliable source of cells that produce sufficient quantities of insulin in response to glucose; starting with embryonic stem cells, a team led by Emmanuel Baetge have apparently developed just that1.

Type I diabetes occurs when insulin-producing β-cells are destroyed by the immune system. Treating patients by transplanting a pancreas or pancreatic cells is reasonably successful, but the tissues are hard to get, and the transplants require a difficult immunosuppression routine. Nonetheless, the potential of transplanting functional cells has fuelled attempts to derive insulin-producing cells in vitro using human embryonic stem cells (hESC). Though β-cells have previously been produced in vitro, they have been unable to respond well to glucose and produce high amounts of glucose in vitro2 or in vivo3.

Previous work has shown that, when transplanted into diabetic mice, immature fetal pancreatic tissue can differentiate and correct hyperglycaemia. Baetge's team decided that instead of trying to fully differentiate the cells in vitro using the protocol they'd previously developed, they would try using cells at an earlier stage, when they resembled fetal pancreatic tissue, and see if they could continue to differentiate in vivo. These 'pancreatic endoderm' cells were grafted into immunocompromised mice. Only 30 days after engraftment, the cells produced human insulin in response to glucose, and the effect became more robust over time. The grafts expressed endocrine hormones independently of nearby tissue, and their structure resembled the mature pancreatic islets where β-cells are usually found. However, perhaps the most convincing evidence of the technique's potential is that the engrafted mice still did not get diabetes after researchers treated them with a cytotoxin that specifically destroys endogenous mouse β-cells. It is still unknown how long the grafted cells will survive; work in heart and brain diseases show that grafted cells tend to die over time.

Still, this is exciting news and has already generated a flurry of media attention. Researchers have finally coaxed hESCs into giving rise to a mature pancreatic cell population that can respond to glucose stimulation and produce insulin in a regulated fashion. The authors feel this is a promising step in using hESCs to develop a renewable source of insulin-producing islets for diabetes cell replacement therapies.

Related articles

No stem cell is an islet (yet)