WEHI researchers using stem cells to prevent diabetes

Researchers at the Walter and Eliza Hall Institute are devising a stem-cell-based approach designed to prevent the development of type 1 diabetes in susceptible individuals by engineering the haematopoietic stem cells to express proinsulin after differentiation into antigen presenting cells.

The strategy, developed by Dr Raymond Steptoe, Janine Ritchie and Prof Len Harrison, has been published in the current issue of the Journal of Clinical Investigation.

Harrison said the strategy was based on a number of lines of evidence suggesting that proinsulin, produced primarily by the pancreatic beta cells but also by cells in the thymus, is the key autoantigen responsible for the self-directed immune response that destroys the insulin producing cells in both humans and in the non-obese diabetic (NOD) mouse model of type 1 diabetes.

Working on the principle that T cell tolerance, a process that removes self-reactive T cells during the thymic maturation process, can be generated by expression of antigen on antigen presenting cells, the researchers engineered haematopoietic stem cells to express a proinsulin transgene that was expressed when the cells differentiated to become antigen presenting cells.

After transfer back into the host, the stem cells repopulated the host immune system, and antigen-presenting cells expressing the proinsulin transgene induced the mice to become tolerant to the epitopes responsible for destroying insulin-producing cells. The engineered stem cells prevented the spontaneous development of diabetes in the NOD mice.

In addition, said Harrison, the proinsulin-producing antigen presenting cells continued to be generated in the host, promoting tolerance toward the proinsulin epitopes.

Harrison said the experiment was a good proof of concept study for the strategy, demonstrating that tolerance-inducing stem cells could be used to prevent the development of autoimmune disease in genetically or immunologically susceptible individuals.

"It's essentially an autologous bone marrow transplant, so the cells become a permanent part of the host," he said. Allogeneic bone marrow transplants were already being used to treat some autoimmune diseases, but these approaches had problems with rejection, requiring the use of potentially toxic drugs.

Not suitable for all

But the new approach has many hurdles to overcome before it can be attempted in the clinic, said Harrison, and would not be suitable for individuals whose pancreatic cells had already been destroyed by their immune response.

"First of all, you have to be able to do this under conditions of minimal or no prior conditioning of the host, such as irradiation to destroy the immune system," Harrison said. The mouse experiments found that the number of genetically altered stem cells required to make the immune system tolerant was very small, and because the cells would be autologous, irradiation to ablate the immune system might not be necessary, he explained.

The second important problem that would need to be overcome, said Harrison, was the need to develop a technique suitable for transferring the transgene into the stem cells.

"In the mouse model, we can use retroviral transfer systems, but this is potentially dangerous and therefore not feasible in humans," he said. "But we have some ideas -- it will probably bee a new form of molecular engineering that does not use retroviruses or lentiviruses at all."

Harrison said that the research was still at a very early stage and would probably be five to 10 years away from the clinic, requiring studies in larger animal models such as primates to refine the technology before it could be tested in humans.

Steptoe said the advantage of the approach over non-specific therapies such as general immunosuppressants was that it specifically targeted immune cells responsible for inducing diabetes, without compromising the rest of the immune system. And the approach could be used to prevent other autoimmune diseases including multiple sclerosis and rheumatoid arthritis.

Harrison has also been examining the proinsulin epitope that appears to be responsible for much of the immune response involved in diabetes. In a second paper in the same issue of the Journal of Clinical Investigation, WEHI researchers, in collaboration with others from the University of Melbourne as well as Italian and Greek teams, have demonstrated that a peptide spanning the junction of the insulin B chain with the connecting C peptide in proinsulin contains both T helper and cytotoxic T cell epitopes.

The researchers initially used the peptide to intranasally immunise NOD mice, hoping to inhibit spontaneous development of diabetes -- in effect creating a diabetes vaccine. But the treatment only reduced the incidence of spontaneous diabetes from around 80 per cent to 60 percent of NOD mice, and in addition to generating CD4+ helper T cells, a population of CD8+ cytotoxic T cells was also produced.

Using predictive algorithms, the lead researcher Dr Nathan Martinez found that the peptide not only contained an epitope that bound to MHC class II molecules, it also contained another epitope recognised by MHC class I, which resulted in the generation of the cytotoxic cells.

"We called it a combitope, because it had overlapping CD4 and CD8 epitopes," said Harrison.

Simple solution

The solution was simple -- the researchers simply truncated the peptide, retaining the ability to generate regulatory T helper cells, but removing the potential to generate the undesirable cytotoxic response. And according to Harrison, the strategy worked, with the number of mice developing diabetes after intranasal immunisation with the truncated peptide dropping to 25 per cent.

"It means you can't just use any peptide, you have to look closely at what it can do," said Harrison. Alternatively, he said, a more general approach would be to block CD40, a co-stimulatory molecule required for the cytotoxic T cell response, although this would be a more complicated procedure.

Harrison said that the WEHI researchers were using their strong base in immunology to come up with useful strategies to deal with autoimmune diseases like diabetes.

"We're trying to understand the natural history of diabetes from before birth," he said. "We need to go back to these early events in all autoimmune diseases, and diabetes is a good model to start with."

The team is funded in part by the US-based Juvenile Diabetes Research Foundation (JDRF), the NIH and Diabetes Vaccine Development Centre, a partnership between the NHMRC and the JDRF.