Report on Human Islet Research Network, May 2016 meeting
by Adam Burrack, PhD
I recently had the chance to attend the annual meeting of the human islet research network in Bethesda, Maryland. This is a group of researchers funded through the National Institutes of Health, specifically the NIDDK and the Special Diabetes Program. The human islet research network is divided into 4 working groups, investigating broad questions of beta cell development, death during diabetes onset (both type 1 and type 2), the immune system’s role in destroying beta cells, and technical advances in techniques to investigate all of the above. HIRN has developed a facebook page and a LinkedIn page to facilitate interaction both between researchers and interactions between researchers and the public – in particular people with type 1 diabetes.
This was a very full meeting: 2.5 days with 24 hours of various meetings, working groups, and social mixers. I never left the city block on which our hotel in Bethesda was located. Below I will summarize some of the highlights from the meeting which have since been published in peer-reviewed journals.
Markus Grompe presented data showing that there are 4 distinct sub-types of beta cells in human pancreas samples. This group demonstrated expression of 2 proteins on the surface of human beta cells which had not been previously shown, CD9 and the enzyme ST8SIA1. Both of these proteins play roles in cell-cell adhesion, for instance they might play roles in establishing pancreatic islet architecture. Interestingly, the authors found that one of these types responds most strongly to changes in glucose concentration, suggesting these are the beta cells most responsible for production of “bolus” insulin following meals. These beta cells do not express either of the above-noted cell adhesion molecules. In contrast, another sub-type (expressing both CD9 and ST8S1A1) produced the most insulin at low glucose concentrations and also responded the least to high glucose concentration, suggesting these beta cells are primarily responsible for production of “basal” insulin. This ‘double-positive’ beta cell type was the most rare in normal human pancreas samples. The other 2 sub-types of beta cells are intermediate between these two extremes by protein expression (they express one but not the other) and behave more like “bolus-producing beta cells” than like “basal-producing beta cells”. Most interesting from this presentation, pancreas samples from individuals with type 2 diabetes had significantly enhanced numbers of “double-positive” beta cells. What in a normal subject is the most rare beta cell sub-type becomes the most prevalent beta cell sub-type in several of these patients. This feels logical, since type 2 diabetic patients often have massively enhanced fasting insulin levels due to peripheral insulin resistance. The next question is a chicken-or-egg: does the beta cell distribution change first, or does the insulin resistance precipitate the change in beta cell distribution?
Another cutting-edge study from this meeting was published by Teresa DiLorenzo at Albert Einstein College of Medicine. In this study Dr. DiLorenzo utilized a humanized mouse to show that autoreactive T cells could traffic to the pancreas. To accomplish this goal, researchers in her group had to genetically engineer out mouse MHC class II (the molecule with which CD4 T cells interact) and engineer in human DR4 (the HLA molecule most associated with type 1 diabetes). This may sound straight-forward, but probably took several years’ worth of genetic engineering and mouse breeding. The other neat trick in this report is ‘transducing’ human CD4 T cells with the T cell receptor specific for a key target of autoreactive T cells in human diabetes, GAD65. ‘Transduced’ means they used a virus to deliver this T cell receptor into these cells, forcing the cells to express this and only this T cell receptor. This is key because any transduced DR4+ T cells could be used, rather than only T cell lines isolated from T1D patients. The researchers also know precisely the specificity of each autoreactive T cell in these experiments. Neat result from this study: 12 weeks after transfer of human T cells, there were significant numbers of human T cells within the pancreas of these humanized mice. This is a key proof-of-principle humanized mouse experiment. Human T cells can engraft in mice expressing human HLA II alleles, and autoreactive CD4 T cells traffic to the pancreas in these mice.
There have been at least 64 publications from HIRN members in the past 2 years alone. Between beta cell regeneration using iPS cell lines, controlling the autoimmune response through inducing tolerance, and developing more-and-more sophisticated tools to study the immune system, we can expect many, many more ground-breaking reports from the HIRN group in the future.