Year 2

Type 1 diabetes (T1D) is caused by the destruction of insulin producing b-cells in the pancreatic islets by an overactive self immune system. Due to lack of ways of preventing T1D, insulin administration is currently most common method of treatment, but susceptible to developing serious, long-term complications. Islet transplantation provides an excellent alternative and can improve glycemic control and reduce the risk of further complications associated with diabetes. Transplantation of pancreas or isolated islets across immune barriers has been conducted with some efficacy but is limited by inadequate organ availability and the need for long-term non-specific immunosuppression. The ability to transplant a renewable supply of islets or β-cells would have a dramatic impact on the disease. Human embryonic stem cells (hESC) represent one of the most promising alternative sources of β-cells, given their pluripotent and self-renewing properties. However, we hypothesized that the immunocompetent recipients will recognize the product as foreign (and autoimmune-self) due to MHC-mismatch (allogeneic) between donor and recipient, leading to transplant rejection. Thus, in order to make stem cell-derived islet cell therapy both commercially and clinically viable, β-cell protective therapies need to be developed to combat the strong allogeneic immune response, as well as the recurrent autoimmune attack which caused the disease in the first place.

In previous studies, we have shown that a small population of CD4+ regulatory T cells (Tregs) plays a critical role in controlling self-reactivity thus maintaining immune tolerance. Adoptive immunotherapy with Tregs into diabetes prone non-obese diabetic (NOD) mice can restore immune tolerance leading to a long-term cure of T1D and have shown a lot of promise for the treatment of autoimmune diseases in the preclinical studies. A key parameter has been the enhanced activity of antigen specificity in directing the tissue-protective functions of Tregs. Thus, in this proposal, we proposed to exploit antigen-specific Treg to selectively regulate ongoing immune responses while avoiding the need for pan-immunosuppression. In the past year, we have made significant, albeit slow progress in this effort. We have taken significant steps to develop an approach to expand alloantigen-specific Tregs (those that recognize and suppress immune responses against foreign tissues). We have developed a process for isolating and expanding alloantigen-specific Tregs to determine their ability to suppress stem cell-derived islet transplants. We have established a humanized mouse model to test the hypothesis that alloantigen-specific Tregs will potently suppress rejection of stem cells. We set up a mouse model of islet transplantation by implanting human islets (from cadavers) in mice lacking a functional immune system. These islets were functional and produced insulin to control normal glucose levels in mice. However, when these mice were complemented with human immune system, the islet grafts were rejected by the allogeneic immune response. This provides an excellent mimic to the human islet transplant where a similar allogeneic immune response leads to the graft rejection. This will provide a setting for testing the allospecific Tregs.

This year we have begun to test the stem cell-derived islets in this setting. First, we have developed a more efficient and effective protocol for developing pancreatic endoderm from human embryonic stem cells. These cells have been transferred into immunodeficient mice and shown to produce human insulin after some period of time (3 months). In addition, we have developed in vitro modifications of the differentiation process that has enabled the formation of cells that express markers of fully matured insulin producing cells without extensive in vivo maturation whose functionality is currently assessed. Finally, we have been able to mimic this process using a genetically marked ES population (luciferase transduced). This allows for rapid and ongoing analysis of cell survival in transplanted animals to enable rapid and extensive therapeutic interventions.

Finally, we have validated a humanized mouse model that allows an analysis of alloreactive Tregs in this setting. Thus, in the coming year, we fully expect to be able to test the hypotheses posed in the original application, as all the tools are now available. In addition, we will work towards adapting the process to the analysis of iPS cells.