Two weeks after transplant into a diabetic mouse, human pancreatic β cells made in the lab produce enough insulin (green) to cure the animal.

Two weeks after transplant into a diabetic mouse, human pancreatic β cells made in the lab produce enough insulin (green) to cure the animal.

Douglas Melton

For diabetes, stem cell recipe offers new hope

Douglas Melton is as impatient as anyone for a cure for diabetes. His son developed the disease as an infant, and his daughter was diagnosed at age 14. For most of the past 2 decades, the developmental biologist at the Harvard Stem Cell Institute has focused his research on finding a cure. This week, he and his colleagues report a potentially significant step toward that goal: a recipe that can turn human stem cells into functional pancreatic β cells—the cells that are destroyed by the body’s own immune system in type 1 diabetes patients such as Melton’s son and daughter. The cells the researchers produced respond to glucose by producing insulin, just as normal β cells do. And when implanted into mice with a form of diabetes, the cells can cure the disorder.

“The diabetes research community has been waiting for ages for this type of breakthrough,” says Jorge Ferrer, who studies the genetics of β cells at Imperial College London. The lab-generated cells should be a valuable tool for studying diabetes and, Melton hopes, could eventually be used to treat patients.

Throughout the day, the pancreas regulates the body’s blood sugar levels, responding to an increase in glucose after a meal by secreting insulin, which helps cells take up the sugar. In type 1 diabetes, the body’s immune system mistakenly kills the β cells for still-unknown reasons, and the body is left without insulin. People control their diabetes by injecting carefully calibrated doses of insulin. But matching the precise insulin control achieved by the healthy pancreas is almost impossible, so researchers have hoped for decades to find a way to replace the missing cells.

When scientists isolated human embryonic stem (ES) cells in 1998, hopes soared. ES cells are pluripotent, which means that in theory they can turn into any of the body’s cell types—including β cells. Indeed, one of the first things researchers tried to make from ES cells was pancreatic β cells. Later they tried with so-called induced pluripotent stem (iPS) cells, made by reprogramming adult cells into an embryolike state. Either way, “it’s proved to be an extraordinarily complicated undertaking,” says Mark Magnuson of Vanderbilt University in Nashville, who studies pancreatic development.

Several teams have turned stem cells into precursors of β cells, which mature when placed into experimental animals. But the cells take 6 weeks to become fully functional β cells, and they can’t be studied easily outside the body. Nevertheless, a clinical trial started last month to test their therapeutic use in patients.

In Cell this week, Melton and his colleagues report a complex recipe that can transform either human ES cells or iPS cells directly into functional β cells. The breakthrough is based on more than a decade of tenacious work in Melton’s lab. He and his colleagues have painstakingly studied the signals that guide pancreas development, applying what they and others have found to develop a method that turns stem cells into mature β cells. “There’s no magic to this,” Melton says. “It’s not a discovery so much as applied developmental biology.”

The protocol “is reproducible, but it is tedious,” Melton adds. The stem cells are grown in flasks and require five different growth media and 11 molecular factors, from proteins to sugars, added in precise combinations over 35 days to turn them into β cells. On the bright side, Melton says, the technique can produce 200 million β cells in a single 500 ml flask—enough, in theory, to treat a patient. Melton says the protocol seems to work equally well with ES and iPS cell lines.

Before the cells can be used to treat type 1 diabetes, researchers need to find a way to protect them from immunologic rejection. The same autoimmune response that triggered the disease would likely attack new β cells derived from the patient’s own iPS cells, and a normal immune response would destroy ES-derived β cells, which would appear foreign. (That has been a challenge for efforts to treat type 1 diabetes with received transplants of β cells from deceased organ donors.) Melton and colleagues are now exploring how to physically encapsulate their stem cell–derived β cells, as well as ways to modify the β cells to enable them to ward off immune attack.

In the meantime, the cells should help the study of the autoimmune disorder. The technique “potentially provides ways to create model systems for studying the genetic basis of diabetes, or to discover novel therapies to enhance existing β cells,” Ferrer says. Melton says his lab has iPS cell lines from people with diabetes—both type 1 and type 2, in which the β cells are not destroyed—and healthy controls. They are generating β cells from those cell lines to look for differences that might explain how the different forms of the disease develop. They will also screen for chemicals that can stop or even reverse the damage diabetes does to β cells.  

Melton says his son and daughter—now 23 and 27 years old—were pleased but unsurprised by his group’s progress. Reversing the parent-child role, they gently nagged him to “get going and solve the [immune-rejection] problem.”

Follow News from Science

A 3D plot from a model of the Ebola risk faced at different West African regions over time.
dancing shoes