Efforts to replicate controversial diabetes therapy bring partial success

March 23, 2006

An effort by researchers at the University of Chicago to confirm the results of a high-profile study that brought extraordinary hope to diabetes researchers and patients worldwide has met with considerable, but not complete, success. The researchers were able to reverse type-1 diabetes in one-third of mice but were unable to find any evidence of insulin-producing beta cells derived from donated spleen cells, initially thought to be a crucial component of the therapy.

Two other research teams, one from Harvard University and one from Washington University in St. Louis, found similar results. All three papers are published in the March 24, 2006, issue of Science.

"Using a protocol that was identical to the original study, we were able essentially to cure 32 percent of treated mice, which was quite encouraging, even though it was less than the 67- to 92-percent cure rates previously reported," said transplant immunologist Anita Chong, PhD, associate professor of surgery at the University of Chicago and lead author of the study. "We saw no evidence, however, of spleen-derived beta cells in the pancreas, despite using very sensitive tests."

While the results provide a boost for efforts to reverse type-1 diabetes in recently diagnosed patients by manipulating the immune system, they come as a disappointment for those who hoped to cure established diabetes by using stem cells from donor spleens to help patients grow new pancreatic islets.

In type 1 diabetes, the body's immune cells mistakenly attack the insulin-producing beta cells found in the pancreatic islets. As the islets die, insulin production ceases and blood sugar levels rise, damaging organs throughout the body.

One of the most promising treatments is islet transplantation, in which islets are extracted from a donated pancreas and injected into the patient's liver, where they take up residence and begin making insulin, restoring control of blood sugar levels. Widespread use of islet transplantation is constrained, however, by the challenges of preventing the recipient's immune system from attacking the transplanted islets, and by the extreme shortage of pancreas donors.

A paper published by Denise Faustman and colleagues at Harvard Medical School in Science on November 14, 2003, on "Islet Regeneration During the Reversal of Autoimmune Diabetes in NOD Mice," appeared to provide a way around both hurdles. In that paper, Faustman argued that her team had been able to cure diabetic mice by disrupting the immune system's attack on the insulin-producing beta cells and injecting the mice with cells from the spleens of other mice, which would "rapidly differentiate into islet and ductal epithelial cells within the pancreas."

"It was immediately clear that this could be extremely important," said diabetes specialist Louis Philipson, MD, PhD, professor of medicine at the University of Chicago and senior author of the paper. "Patients wait years for pancreatic islets. We have hundreds of people on our waiting list, yet we perform only a few islet transplants a year. Having a plentiful source of potential islet cells is our dream, so the news was almost too good, which is why the diabetes community wanted to see this replicated before it moved into human trials."

The three papers in the current issue of Science attempt to replicate that work. With funding from the Juvenile Diabetes Research Foundation, the Chester Foundation and the National Institutes of Health, Chong, Philipson and colleagues set out to copy the Faustman protocol exactly--with one slight twist.

They worked with non-obese diabetic (NOD) mice, a common model for studies. When the mice developed severe diabetes, they injected one foot pad of each mouse with Freund's complete adjuvant, a potent mixture that can over-stimulate the immune cells that cause diabetes. To protect the mice from the ravages of their diabetes during the study, they temporarily transplanted 400-500 islets per mouse under one kidney capsule. And they injected the mice with spleen cells.

In the original study, the treated mice were all female and the spleen cells came from male donors. The presence of a Y chromosome served as a marker for donated cells, a technique that, according to Philipson, can be "fraught with technical artifacts." Instead, Chong and Philipson used spleen cells from mice that expressed the transgene for green fluorescent protein (GFP) driven by the insulin promoter. So if any of these spleen cells begin to produce insulin, they would glow bright green, making then easy to detect.

They treated 22 mice. Seven of those mice (32%) maintained normal blood sugar levels for the duration of the study, 90 to 100 days. Although one died from unrelated causes, the kidney containing the transplanted islets was removed from the remaining six mice. All six continued to maintain normal blood sugar levels for an additional 29 to 42 days. This suggested that the 'cured' diabetic mice now had enough beta cells to maintain normal blood glucose without the transplanted islets.

In none of those mice, however, was the pancreas normal. Mice of this age would ordinarily have about 2,000 small islets, scattered throughout the organ. These mice had about a dozen extremely large islets, five times normal size, composed almost entirely of beta cells and surrounded by white blood cells. Total beta-cell volume in treated mice was about 22 percent of normal for mice of that age, but it was enough to maintain normal blood sugar levels.

However, no beta cells expressing GFP were detected in any of the islets examined or in the organs outside the pancreas. "Our data," the authors note, "do not support a conclusion of beta-cell regeneration from spleen-derived stem calls."

"The question we are now addressing is the source of these beta cells," Philipson said. "They could be beta cells that survived the initial immune onslaught, recovered and replicated. Or they could be pancreatic stem cells, which had matured into beta cells once autoimmunity was controlled."

Despite the inability of any of the three teams to replicate that aspect of the study, the authors were encouraged by their other successes. "Our studies confirm that autoimmune diabetes can be reversed," they conclude, "and that sufficient beta-cell mass can be restored to cure diabetic NOD mice with the treatment protocol developed by Faustman and colleagues."

These GFP mice were generated by co-author Manami Hara, assistant professor of medicine at the University of Chicago. Additional authors include Jikun Shen, Jing Tao, and Andrey Kuznetsov of the University of Chicago, and Dengping Yin, who is currently at the University of Illinois at Chicago.