Projects & Publications. Current nPOD Projects

Beta Cell Physiology and Dysfunction

Taro Akiyama Merck Confirmation of target gene expression in T1D and T2D pancreas

Characterization of expression of novel anti-diabetes targets in pancreas of Type 1 and Type 2 diabetics.

Graeme Bell & Manami Hara University of Chicago Beta-cell Death and Survival

Type 1 diabetes (T1D) is a progressive immune-mediated disease that is preceded by an asymptomatic preclinical period of highly variable duration. A better understanding of beta cell and islet function or dysfunction during this asymptomatic period and the nature of the “dialog” between islet endocrine cells and the immune system could lead to new therapeutic strategies to arrest the progression to autoimmunity. Here we propose to develop a framework for understanding the underlying biology of T1D by integrating histological and genomic (RNA-seq) analyses of the human pancreas, pancreatic islets and islet endocrine cells. We anticipate that these studies will lead to a new understanding of the pathophysiology of T1D and point to biomarkers of human beta-cell stress, death and survival.

Gary Cline & Richard Carson Yale University School of Medicine Imaging pancreatic beta-cells with PET neuroimaging agents

A clinically viable means to measure pancreatic beta-cell mass (BCM) is essential for evaluating the physiological basis of therapeutic approaches to restore deficient insulin secretory capacity. Major advances in imaging BCM have been made by taking advantage of receptor-specific imaging probes that have been successfully used for neuroimaging. Hence, Positron Emission Tomography (PET) imaging ligands that were originally developed to specifically bind to neurons may prove useful for imaging BCM.

The Yale PET Center has been at the forefront of imaging both brain receptors and BCM and proposes to evaluate its extensive library of human neuroimaging agents as in vivo probes to quantitatively determine BCM. To maximize the value of these studies, pancreatic imaging in humans will be obtained together with validation studies in healthy and type 1 and 2 diabetes mellitus pancreas. Imaging probes that show suitable in vivo specific uptake in pancreas and appropriate imaging properties in humans, and can distinguish between healthy and diabetic pancreas in vitro will then be tested in a limited clinical PET-imaging trial to assess whether there is a measurable decrease in radiotracer binding in the pancreas of T1DM patients.

In Specific Aim 1, we will evaluate whether the approved radiotracers currently in use at the Yale PET center for human neuroimaging can be used for imaging BCM in healthy individuals.

In Specific Aim 2, we will validate β-cell specificity of the radioligands in vitro in healthy, T1DM, and T2DM human pancreas tissue obtained from the network for Pancreatic Organ Donors with Diabetes (nPOD), and in human islets.

In Specific Aim 3, we will perform a clinical evaluation in healthy and T1DM volunteers of those agents that meet the criteria of 1) promising pancreas imaging characteristics determined in Aim 1 and 2) favorable β-cell specificity as determined in Aim 2. A limited clinical evaluation is necessary to establish 

Yuval Dor & Benjamin Glaser Hebrew University of Jerusalem Markers of stress in diabetic islets

Our project seeks to understand the molecular basis for beta cell failure in human diabetes. We have discovered that excessive glucose metabolism in beta cells may lead to double strand breaks in the DNA and activation of the tumor suppressor gene p53. With the help of nPOD, we are studying the significance if this response in human type 2 diabetes patients. In our experiments we utilize archived material from human pancreata to examine the expression of multiple markers of the DNA damage response, to understand the nature of this response.

Dieter Egli, Bjarki Johannesson & Ruddy Leibel Columbia University Funded by: Helmsley Foundation Generation of pluripotent stem cells from diabetics ADDENDUM: Generation of iPS cell lines from a majority of T1D cases

ORIGINAL PROJECT: Established different skin fibroblast cell lines, both from control and T1D nPOD donors. Graciously offered this as a resource available to current nPOD investigators.

ADDENDUM: The pancreata of nPOD cases are some of the best characterized type 1 diabetes cases. The availability of beta cells from stem cells would greatly enhance this resource by providing the ability to both study the primary pancreas, as well as living beta cells or other cell types in a defined environment with the genotype of the patient. This project aims to generate cell lines for as many of the nPOD cases as possible, and to convert selected cell lines to iPS cells, and stem cell derived beta cells to determine whether disease phenotypes are cell-autonomous effects of the genotype, or acquired through interactions with the environment or with other cell types.

Decio Eizirik, Ihsane Marhfour & Anne Op de beeck Universite Libre De Bruxelles, Belgium Funded by: JDRFI Characterization of beta cell ER stress associated to enteroviral infection in type 1 diabetes

Enteroviral infections are associated with an increased risk to develop type 1 diabetes. Several stains of enteroviruses are able to infect beta cells leading to both inflammation and cell death, two major factors for the onset of type 1 diabetes (T1D).  The cellular factors that modulate cellular permissiveness to enteroviral infection and beta cell death remain largely unknown. Since ER stress is involved in beta cell death and viral infections induce ER stress in other cell types, the aim of this project is to address the role of ER stress in virally induced beta cell death and local inflammation.

For this purpose, we will first determine whether components of the unfolded protein response (UPR) are induced during beta cell infection with a diabetogenic coxsackievirus (CVB5). Our preliminary data indicate an increase in CHOP, Bip and XBP1s mRNA expression in infected INS-1E cells. These results must be confirmed at the protein level and the activated UPR pathways further detailed.

Second, the role of the putative viral-induced ER stress in beta cell death and release of inflammatory mediators will be addressed by modulation of their expression through a silencing/overexpression approach and further evaluation of cell viability after CVB5 infection. These experiments, and the above described experiments, to be initially made in INS-1E cells, will be confirmed in primary rat and human beta cells. The accumulation of the involved ER stress markers in CVB5-infected beta cells, as well as co-detection of the capsid viral protein, will be evaluated by immunohistochemistry. To confirm the in vivo relevance of these observations, we will evaluate the co-localisation of the involved ER stress markers with viral capsid proteins in sections of pancreas from controls and coxsackievirus infected- and non infected-T1D donors. Altogether these results should shed light on the role of ER stress in beta cell death during coxsackievirus infection. 

Ivan Gerling University of Tennessee Health Science Center Funded by: JDRF Defining islet heterogeneity using single islet transcriptomics

The goal of this project is to obtain transcriptome data from human islets using either hybridization arrays or RNA sequencing technology.  We plan to obtain complete information about relative expression of all genes in islets from a number of different organ donor groups with an emphasis on: 1) donors with no autoantibodies or diabetes, 2) donors with autoantibodies but no diabetes, 3) donors with newly diagnosed type 1 diabetes, 4) donors with long-standing (over 5 years) type 1 diabetes, and, 5) donors with type 2 diabetes.  The dataset will be analyzed to identify genes whose expression differs significantly between one or more of these donor groups.  The goal is to define what pathways and processes uniquely define islets in each donor group, in order to gain new insights into the pathophysiology of diabetes and discover possible new drug targets.  To further that goal each list of differentially expressed genes (distinguishing islets in one group from one or more of the other groups) will be subjected to a data mining process designed to define if the list is significantly enriched in genes that: 1) belong to specific pathways or gene ontologies; 2) are regulated by specific transcription factors; 3) are known to be influenced by specific drugs, proteins or other molecules.  This will provide a global (systems level) definition of gene expression patterns in islets from each of these donor groups.  However, we also intend to make this dataset available to other investigators who may want to examine the data for expression of genes relevant to their specific studies.

Manami Hara University of Chicago Funded by: JDRF Beta-cell destruction and preservation in early and late onset type 1 diabetes

Type 1 diabetes (T1D) is one of the most common chronic diseases of childhood with onset peaking between 5-7 years of age and around puberty. Early childhood-onset T1D often exhibits severe beta-cell depletion compared to late adult onset T1D. We propose to test the hypothesis that immature beta-cells in early childhood are more susceptible to autoimmune destruction, whereas the cell death of mature beta-cells in adult onset T1D undergoes different kinetics and its process is substantially slower than previously considered. A better understanding of distinct beta-cell destruction and preservation during childhood could lead to new therapeutic strategies to treat young children with T1D.

Gokhan Hotamisligil & Feyza Engin Harvard University Funded by: JDRF Examination of ER stress markers in type 1 diabetes samples in humans

The endoplasmic reticulum (ER) is an organelle that is responsible for the proper folding of proteins and biosynthesis of lipids and steroids. Disruption of ER homeostasis leads to ER stress and activates a highly conserved adaptive network called the unfolded protein response (UPR). ER stress has been implicated in several diseases associated with protein folding defects and low grade chronic inflammation including neurodegenerative diseases, type2 diabetes, and atherosclerosis. The ability of UPR mediators to activate inflammatory and apoptotic signaling suggest ER stress might also be critical in development or progression of type1 diabetes. Despite increasing evidence that points to a critical role for ER stress in the pathogenesis of type1 diabetes, to date alterations in this pathway in human diabetic patients has not been demonstrated leading to a great scientific gap in understanding how organelle stress and related stress responses function in the pathogenesis of this disease. Our specific aim is to investigate the ER stress marker expression in the pancreata of human type1 diabetes and control samples by using immunofluorescence (IF) technique.

Pamela Itkin-Ansari, William Balch & Randal Kaufman UCSD Funded by: NIDDK HIRN Insulin interacting proteins and stress markers in T1D

We have developed novel methods for proteomic study of beta-cells within native islets. Using
antibodies to proinsulin or insulin we reported the first Biosynthetic Interaction Network for insulin
(Pottekat, Cell Reports, 2013). Based upon our finding that many proteins which interact directly with
insulin are T1D GWAS candidates or proteins required for ER homeostasis, we propose to determine
how these proteins and others in their class are altered during progression to T1D. Moreover, based
upon recent evidence of ER stress in T1D islets and our identification of compounds which stabilize
the ER protein folding environment in vivo, we propose to test new strategies for preserving or
restoring beta-cell mass in T1D.

Mitchell Knutson University of Florida Funded by: NIDDK and NIH Molecular mechanisms of iron uptake by pancreatic beta cells and their contribution to the development of diabetes

Human disease states that result in excess iron accumulation, e.g., hereditary hemochromatosis and beta thalassemia major, are associated with an increased incidence of diabetes. In these disorders surplus iron overwhelms normal transport and storage mechanisms leading to the appearance of excess free iron in circulation and the subsequent deposition of iron in tissues such as the pancreas. Excess iron in tissues can lead to the formation of free radicals, which can damage beta cells and lead to cell death or dysfunction. Despite the long-established connection between iron overload and diabetes, exactly how iron is transported into pancreatic beta cells is not yet known. Currently only three mammalian proteins are known to be able to transport free iron: Divalent Metal-Ion Transporter-1 (DMT1), ZRT/IRT-Like Protein 14 (ZIP14), and ZRT/IRT-Like Protein 8 (ZIP8). The aim of our project is to determine the expression and subcellular localization of these established iron transporters in the cell types of the human pancreas, focusing primarily on beta cells. Identification of potential routes for beta cell iron uptake will provide targets for future mechanistic studies concerning the relationship between beta cell iron loading and function.

Etienne Larger, Marc Diedisheim & Roberto Mallone Université Paris Descartes Beta-cell mass in Antibody-positive non-diabetic subjects

The question of the B-cell mass of non-diabetic autoantibody-positive (Ab+) subjects is crucial to understand both common forms of type 1 diabetes and latent autoimmune diabetes (LADA). It is well established that the presence of GAD Abs is not sufficient to predict diabetes or insulin dependency. Knowing the B cell mass of Ab+ donors could be crucial to decipher the history of diabetes between the initial lesion, marked by the appearance of Abs, and the onset of diabetes. To this aim, we propose to use the insulin-stained virtual slides of the nPOD project to quantify the B cell mass of Ab+ subjects. We will first test whether the insulin-stained virtual slides of the nPOD database can be used for this analysis. To this end, we will compare the B cell mass, islet size distribution and B cell distribution throughout the pancreas in control donors of the nPOD database with published data of the literature. We will then compare the Ab+ subjects to control donors and patients with recent-onset, i. e. <3 years T1D.

Clayton Mathews University of Florida Funded by: JDRF Islet resistance to T1D

     Our proposed studies will address a major gap in the understanding of T1D, namely, to identify the mechanisms of beta cell failure from a metabolic as well as an immunological perspective. This specific project has high potential to reveal significant, novel, and translatable information regarding beta cell dysfunction that will be of therapeutic benefit for individuals with both established T1D as well as persons with pre-diabetes.  The goal is to identify components of the glucose sensing and insulin secretory machinery pathways that are altered during the natural history of disease progression in type 1 diabetes (T1D), and to characterize the metabolic and immunological signals responsible for the functional loss in beta cell activity that are specifically associated with this disease.  This aim will be achieved by performing two lines of experimentation.  First, gene expression arrays will be utilized to analyze islet RNA isolated from pancreatic sections of JDRF nPOD cases (representative from the natural history of T1D, alongside of appropriate matched controls) following laser capture microdissection (LCM). These analyses will be undertaken to determine the specific gene expression changes that result in beta cell dysfunction, as well as to identify the immunological signals which are responsible for beta cell failure in T1D.  In the second we will use primary human islets obtained from cadaveric donors with T1D as well as non-diabetic subjects, to confirm and extend our associations of phenotypes reflective of beta cell dysfunction. Also, using extended in vitro culture systems to allow for a recovery of beta cell insulin secretory function, followed by expression analysis, we will identify changes in signaling that are reflective of both cellular damage and recovery.  These experiments are posed to test the overall hypothesis that that following the appearance of autoantibodies, inflammatory signals act to increase islet inflammation and negatively impact beta cells through downregulation of genes essential for glucose sensing and insulin secretion. Based on previously published work, we expect that the effects of these signals are reversible.  Elucidating the immune signals that induce beta cell dysfunction would have a major impact on future treatment strategies as they should provide a means to revive the endogenous beta cell mass.

Raghavendra Mirmira & Teresa Mastracci Indiana University School of Medicine Funded by: JDRF Deoxyhypusine synthase: a novel target for beta cell protection

The cellular processes giving rise to type 1 diabetes (T1D) and type 2 diabetes (T2D) involve the activation of inflammation, which leads to the eventual death of islet β cells.  An urgent priority in diabetes research is the discovery of biomarkers (simple blood tests) that can assist in the identification of persons at-risk for disease for the purposes of early therapy.  Recently, our laboratory has been investigating the involvement of a specifically altered form of the protein eIF5A (known as  eIF5AHyp) in the progression of diabetes in mice.  In previously published studies, we showed that eIF5AHyp is responsible for the production of a subset of inflammatory proteins in β cells and immune cells of the mouse.  However, the specific types of cells that exhibit eIF5AHyp in human diabetes have never been examined, largely because specific reagents for tissue analysis have heretofore been unavailable.  The purpose of this study is to determine the cell-type distribution of eIF5AHyp in the human pancreas and spleen from persons with T1D, T2D and controls.  The results of this study will uncover whether the accumulation of eIF5AHyp-expressing cells are a marker for diabetic disease in the pancreas.

Jerry Nadler, Margaret Morris & Kaiwen Ma Eastern Virginia Medical School Funded by: JDRF and NIH 12/15 Lipoxygenase expression in type 1 diabetes

12/15-lipoxygenase (12/15-LO) is an enzyme responsible for the metabolism of arachidonic acid to pro-inflammatory responses by immune cells in a variety of tissues, including pancreatic islets and fat tissues.  Our extensive research has shown that type 1-like diabetes development is significantly diminished in the absence of 12/15-lipoxygenase.  By using 12/15-LO deficient C57BL/6 mice, we were able to show that these mice were protected from developing diabetes following multiple low-dose streptozotocin treatment.  We then developed a non-obese diabetic (NOD) mouse line congenic for this deletion, and saw an extremely significant decrease in the rate of spontaneous diabetes as compared to wild-type NOD mice.  We are moving forward to develop 12/15-LO inhibitors that might be used therapeutically.  In order to maximize the productivity of these compounds, it would be extremely beneficial to garner information from type 1  and type 2 diabetic human samples.  Namely, we are interested in testing the expression of 12/15-LO in human pancreas, lymph node, and spleen samples, so we can determine which cells produce this enzyme in the requested tissues. 

Edward Phelps, Steinunn Baekkeskov &Jeffrey Hubbell École Polytechnique Fédérale de Lausanne Funded by: JDRF How does induction of ER-stress in pancreatic beta cells affect the processing and trafficking of beta cell autoantigens?

In humans, GAD65 is one of the first targets of autoimmunity and 70-80% of human diabetics have antibodies against this protein. In late-onset patients, circulating GAD65 autoantibodies may even be more predicative of future disease onset than antibodies to insulin. The early appearance of GAD65 antibodies may hold a clue that GAD65 is implicated during the earliest phases of the disease, which makes it an attractive target for interventional therapies to prevent clinical progression. One of the peculiarities of GAD is that this protein is expressed in two isoforms, and only the lighter weight form, GAD65 is immunogenic, while the heavier one, GAD67, is not. Mice predominantly express GAD67 in their beta cells, while rats express both GAD65 and GAD67 and humans only express GAD65. Because the majority of diabetes research is conducted on NOD mice where GAD65 is not well expressed in beta cells, the central importance of GAD65 in human diabetes may be overlooked in NOD mouse research. In this proposal we plan to study GAD65 in human tissues to better understand how this protein contributes to T1D. There is considerable evidence to suggest that protein folding stress in beta cells plays a central role in the loss of beta cells that precedes clinical onset of T1D. Our preliminary data in rat islets links beta cell stress with abnormalities in GAD65 intracellular trafficking dynamics as potential sources of increased immunogenicity. If confirmed in human tissues, these results represent a novel and potentially ground-breaking finding into the basic mechanisms of diabetes autoimmunity.

Jon Piganelli & Eddie James University of Pittsburgh Funded by: JDRF Disruption of tolerance by ER-stress in type 1 diabetes

Type 1 diabetes (T1D) is an autoimmune disease in which insulin-producing islet cells are targeted and destroyed by the immune system, especially by T cells. The events that lead to immune recognition of islet proteins in people genetically predisposed to autoimmunity are poorly understood. Many of the environmental triggers associated with T1D including viral infection, exposure to chemicals or free radicals, and dynamic glucose sensing, however, cause cellular stress within islets. In addition, the natural function of islet cells (to produce and release insulin) makes these cells especially susceptible to stress. The cells respond to stress by activating a program that protects them from death, but this results in mishandling of proteins leading to changes in there normal structure. In healthy individuals, this mishandling of proteins is not a problem; however, in individuals prone to autoimmunity, these proteins and the cells that make them may be targeted by the immune system. We hypothesize that the environmentally related events leading to this cellular stress in islet cells causes abnormal modification of proteins which in genetically predisposed patients, leads to abnormal immune system targeting causing the destruction of the islets. We have tested this hypothesis using a model system of T cell recognition of islets, in particular an islet protein called chromogranin A (CHgA) in a well established mouse model of T1D, demonstrating that T cells respond much more strongly to islets exposed to environmental triggers of cellular stress. These data support the hypothesis that cellular stress changes the proteins in the islet, and these changes facilitate the recognition of islets by T cells. We are now proposing to translate these findings further using human islet-reactive T cells, human islets and a humanized mouse model of T1D. The overarching goal of these studies is to determine in the humanized T1D animal model if the same modifications are demonstrated under conditions of cellular stress, and to determine how these modifications lead to the recognition of islets by the immune system. These studies are important because understanding how islet proteins come to be recognized by T cells will reveal targets for therapeutic intervention to prevent incorrect modification of islet proteins, protect islets from the immune system, and ultimately prevent T1D. 

John Andrew Pospisilik & Tess Tsai-Hsiu Lu Max Planck Institute of Immunobiology and Epigenetics Epigenetic regulation of beta-cell function and identity

Diabetes is a disorder of complex interaction between genetic susceptibility and environmental perturbation. Various epigenetic models, such as intra-uterine growth retardation, have been shown to contribute to pancreatic beta-cell dysfunction. More recently, beta-cell dedifferentiation has been discovered as a mechanism for beta-cell dysfunction in diabetic mouse model. The project aim to understand the role of classic epigenetic modules in regulating beta-cell identity and function in human context. We’ll survey changes in histone modifications across human pancreas samples over various age groups and diabetic conditions. Understanding how histone modification changes with age and disease state in human will enable the research community to identity factors that underlie aging related decline in beta-cell function and identity.

Al Powers, Marcela Brissova and Rachana Haliyur Vanderbilt University Funded by: JDRF and NIH Pancreatic islet biology and vascularization

Our group is interested in pancreatic islet development and function and how beta cells are affected in diabetes. Critical discoveries from nPOD samples and resources are changing our understanding of type 1 diabetes. We are using or hope to use nPOD samples for the following studies:

 1) Define and describe the normal human pancreas and islet over the human lifespan. In terms of normal pancreatic islet development, we are studying the fetal human pancreas, the juvenile human pancreas (less than age 18), and the adult human pancreas. We are interested in how islet morphology, gene expression, and function change as humans age and adapt to different physiologic conditions such as puberty and insulin resistance. In addition to studying islet gene expression (hormones, key islet-enriched transcription factors, and differentiation markers), we are also studying the innervation and vascularization of islets. Islets are richly innervated and vascularized and we are interested in interactions between these processes and islet cell function.

 2) Describe the features of insulin+ cells remaining in the pancreas of individuals with Type 1 diabetes (T1D). We seek to understand the functional characteristics, and molecular signatures of the insulin+ cells remaining in the pancreas of individuals with T1D. Little is known about the remaining insulin+ cells and their functional status, or how those cells have survived in the inhospitable metabolic and immunologic environment. It is not known whether these insulin+ cells are ‘normal’ β cells.

 3) Describe the features of islets and beta cells in the pancreas of individuals with cystic fibrosis-related diabetes (CFRD). This form of diabetes is becoming increasingly common, but the pathogenesis is incompletely understood. By studying samples in the nPOD repository, we hope to determine how these beta cells compare to normal beta cells.

 We are using or hope to use nPOD samples from both the control and diabetic pancreas. We will combine information from nPOD samples with our results from samples from human pancreatic tissues that our group has recently collected (fetal and juvenile). We will examine these tissues by immunohistochemistry and other methods such as tissue imaging mass spectrometry. We will use antibodies that are widely available and also new antibodies that are being developed by investigators in the Beta Cell Biology Consortium.

July 11, 2014

Sasanka Ramanadham University of Alabama-Birmingham Funded by: NIDDK, UAB Comprehensive Diabetes Center and Iacocca Family Foundation Role of Calcium-Independent Phospholipase A2β (iPLA2β) on β-Cell Apoptosis

Beta-cell apoptosis contributes to loss of β-cells and decreases in β-cell function in both types 1 and 2diabetes mellitus. It is therefore important to understand the mechanisms underlying β -cell apoptosis if this process is to be prevented or delayed. Our hypothesis is that the group VIA Ca2+-independent phospholipase A2 (iPLA2 β) participates in β -cell apoptosis. We observed that (a) ER stress, proinflammatory cytokines, and prolonged hyperglycemia promote expression and activity of iPLA2 β  inislets, (b) iPLA2 β  activation increases ceramide generation via neutral sphingomyelinase (NSMase)- catalyzed hydrolysis of sphingomyelins triggering the mitochondrial apoptotic pathway and β –cell apoptosis, (c) these outcomes are suppressed by inhibition of iPLA2 β  or NSMase, (d) iPLA2 β –null islets are less and iPLA2 β -transgenic (Tg) islets more sensitive to ER stress, (e) NSMase expression is unaffected in iPLA2 β -null islets and amplified in iPLA2 β -Tg islets, and (f) β -cell iPLA2 β  and NSMase messages are higher in the Akita mouse model of spontaneous β -cell ER stress that leads to diabetes. We also find that iPLA2 β  participates in human islet β -cell apoptosis induced by hyperglycemia and cytokines and that islet iPLA2 β  and NSMase messages are elevated in prediabetic NOD mice. Further, we recently found evidence for inhibition of iPLA2 β  leading to decreased incidence of T1D in diabetes-prone NOD mice. These findings strengthen our hypothesis thatiPLA2 β  participates in the onset and progression of T1D. We propose to validate our hypothesis in human T1D by correlating expression levels of iPLA2 β  with indices of beta-cell apoptosis in pancreas samples obtained from control and diabetic donor subjects.

Shiva Reddy University of Auckland, New Zealand Immunohistochemical identification of molecular markers of oxidative and nitrosative stress in β-cells during the early stages of human type 1 diabetes mellitus

This research seeks immunohistochemical evidence of beta cell oxidative and nitrosative stress during various stages of T1D and whether such stressors are also present preceding and during the early onset of the disease. Antibodies to nitrotyrosine, a marker of peroxynitrite-mediated cell damage and to 8-hydroxy-2-deoxyguanosine (a marker of DNA oxidation) are being employed in triple-label immunohistochemistry to determine if such processes are activated within the beta cell. Combined immunohistochemistry and histochemistry are being applied to correlate the immunolabelling of such deleterious markers with the extent and spatial distribution of leukocytic infiltrates within the islets and with beta cell co-localization. Studies to date have shown weak to moderate immunolabelling of nitrotyrosine restricted to a majority of the remaining beta cells from 5 T1D cases, being also present in islet beta cells with minimum insulitis. Nitrotyrosine was absent in glucagon cells and the islet leukocytes. It was absent or weakly expressed in islet cells of non-diabetic donors (2 cases). DNA oxidation was observed within islet endocrine cells but was not restricted to beta cells from T1D cases.

The expression of nitrotyrosine and DNA oxidation are being extended to non-diabetic cases with single or two autoantibodies.  In line with previous in vitro studies with human islets in culture, the current research is also exploring the possible activation of the pro-inflammatory cytokine-nitric oxide axis within the islets of T1D cases, likely under-expression of anti-oxidant enzymes and whether stressors within the pre-diabetic beta cell lead to beta cell apoptosis and attract the early immune response.

Protocol development, and search for suitable primary antibodies and optimization studies are in progress for the immunohistochemical localization of interleukin-1β, inducible nitric oxide synthase and interferon-α.

Charmaine Simeonovic, Christopher Parish, J. Dennis Wilson & Andrew Ziolowski Australian National University Funded by: JDRF Heparan sulfate levels mark the health status of human islet β-cells

Heparan sulfate (HS) is a sulfated glycosaminoglycan that consists of repeating disaccharides (composed of uronic acid and glucosamine), forming a linear polysaccharide. The HS chains are assembled onto the core proteins of heparan sulfate proteoglycans (HSPGs). We recently reported that HS is localized in the peri-islet basement membrane (BM) in mice and is intensely expressed and widely distributed throughout the intra-islet cell mass, being localized selectively inside insulin-producing beta cells. This intracellular distribution of HS is unique for beta cells, since normally HS is localized extracellularly in BMs and the extracellular matrix (ECM). Using in vitro studies, we have discovered that isolated mouse beta cells need intracellular HS for their survival and protection against oxidative damage. During T1D in NOD/Lt mice, HS is progressively lost from islet beta cells, due to the local production of heparanase (a HS-degrading enzyme) by insulitis leukocytes, resulting in beta cell death. Using nPOD human pancreas specimens, we have demonstrated that normal human islets also express high levels of intracellular HS in their beta cells. In parallel, our immunohistochemical analyses of the distribution of HSPG core proteins revealed strong intra-islet staining for collagen type XVIII and syndecan-1 HSPG core proteins. These findings provide support for the high levels of HS in human beta cells in situ. Like T1D in NOD/Lt mice and in contrast to normal human pancreas, nPOD pancreas specimens from donors early after T1D-onset (≤ 8 years post-onset) have shown selective loss of HS but not HSPG core proteins in remaining insulin-positive islet beta cells. In fact the HS content of T1D islets was only 10% of normal controls. In addition, peri-islet insulitis mononuclear cells in the T1D pancreas specimens strongly expressed heparanase. Our findings strongly support the testing of dual activity heparanase inhibitors/ HS replacers as new therapeutics for preventing T1D progression in new onset T1D individuals.

Roland Stein Vanderbilt University Medical Center Funded by: NIDDK Analysis of Islet Enriched Transcription Factor Expression and Activity in Normal & Type 2 Diabetic Islets

The therapeutic treatment of type 1 and type 2 diabetes mellitus (T1DM, T2DM) has vastly improved over the past decades; however complete normalization of glucose remains elusive without proper functioning pancreatic β cells. Truly, sustained and effective treatment of T1DM and T2DM must involve retaining/replacing the function of pancreatic β cells. To this end, we must understand the molecular mechanisms that govern β cell development and function along with understanding the molecular progression of the disease.

The Stein lab and others have focused on the large Maf transcription factors, MafA and MafB, and their role in the β cell. Loss of MafB impedes normal β cell development in mouse models whereas loss of MafA restricts normal adult β cell function. In recent published data, we have shown that MafA and MafB are lowered in db/db mouse models and T2DM donor samples. Collectively, this data reveals the importance of MafA and MafB in normal β cell function and presents an area of research that needs to be further examined. 

Using nPOD, we are currently characterizing transcription factor levels and investigating coregulator interactions in T1DM and T2DM donor samples. This data is crucial to the understanding of the human β cell situation and vital for β cell replacement therapy protocols. 

Peter Thorn University of Queensland Funded by: NHMRC and Diabetes Australia Beta cell polarity

Insulin secretion is principally regulated by the beta cell response to glucose. This secretory response is multiphasic with an initial burst of insulin secretion followed by sustained or increasing secretion. This complex timing of secretion is important for body function and is lost in type 2 diabetes. Beta cells normally function, and are normally regulated, within the islets of Langerhans which are composed of many hundreds of cells. However, due to technical limitations, beta cell function within islets is not well understood. To this end we have developed new methods that study single beta cells within an islet. Our new evidence shows that beta cells are structurally polarised and secretion of insulin is targeted to the vasculature.

Emil Unanue Washington University School of Medicine Funded by: JDRF The immunogenicity of abnormal catabolites of insulin metabolism

We are searching for abnormal products of insulin metabolism that appear in beta cell as a result of the increased demands for insulin. These abnormal catabolites in NOD mice select for very unique CD4 T cells that initiate this chronic autoimmune reaction. The plans are to search for these abnormal products in islets from early diabetic subjects. Their presence in human islets may point to key antigens that may be the target of future immunotherapy.

Hongju Wu & Vivian Fonseca Tulane University Pancreatic GLP-1 vs glucagon production in healthy and diabetic patients

Glucagon-like Peptide 1 (GLP-1) and glucagon shares the same precursor molecule proglucagon, but each arises from a distinct posttranslational process in a tissue-specific manner. Recently GLP-1 has been shown to be co-expressed with glucagon in pancreatic islet cells. Our preliminary data showed GLP-1 was progressively up-regulated in pancreatic islets during type 2 diabetes development. These data suggest intra-islet GLP-1 production may have an impact on the clinical and pathophysiological processes of these diseases and may be a target for therapeutic approaches. This project is thus designed to investigate this matter in human type 1 and type 2 diabetic patients. In addition, since GLP-1 based drugs, incretins, have been developed and approved for clinical use, it is thus essential to investigate their impact on intra-islet GLP-1 production. The main objectives of this project are thus 1) To investigate whether GLP-1 vs glucagon production in pancreas of human has changed with the development of type 1 and type 2 diabetes; and 2) To investigate whether the GLP-1 based therapies (i.e., incretin therapy) had any impact on the relative production of GLP-1 vs glucagon. To accomplish the goals, pancreatic tissues from normal, type 1 diabetic, type 2 diabetic without incretin treatment and type 2 diabetic with incretin treatment will be analyzed by immunofluorescence staining. The GLP-1 vs glucagon expression will be quantified and analyzed for statistical significance. In addition, the matching blood samples will be used to obtain physiological data related to glucose regulation. The results will help us understand the complete effect of incretin drugs, and identify potential safety issues. Therefore, the study will not only advance our knowledge on the regulation of intra-islet GLP-1 production in diabetes, but also have significant impact on the clinical practice regarding current use of incretins.

Beta Cell Development, Differentiation & Regeneration

Domenico Accili & Chutima Talchai Columbia University Funded by: NIH Beta cell dedifferentiation in type 2 diabetes

Diabetes is associated with β-cell dysfunction. But it remains unclear whether the latter results from reduced β-cell number or function. Purpose of this study is to investigate the hypothesis that β-cell failure is the result of β-cell dedifferentiation, rather than apoptosis or degranutaion. Transcription factor FoxO1 integrates β-cell proliferation with adaptive β-cell function. The rationale for the proposed project arises from the observation that mice lacking FoxO1 in β-cells develop hyperglycemia with reduced β-cell mass following physiologic stress, such as multiparity and aging. Surprisingly, lineage-tracing experiments demonstrate that loss of β-cell mass is due to β-cell dedifferentiation, not death. Dedifferentiated β-cells revert into progenitor-like cells expressing Neurogenin3, Hdac1,and L-Myc. They also partly convert into non-β-cells, resulting in hyperglucagonemia. Strikingly, we identify the same sequence of events as a feature of different models of murine diabetes. We intend to investigate whether humans with type 1 and 2 diabetes develop the same abnormalities observed in Foxo1 β-cell-specific knockouts.

Justin Annes Stanford University Funded by: NIDDK Beta cell replication

Recent decades have seen an incredible growth in the prevalence of diabetes. While it is important to pursue multiple therapeutic strategies, harnessing the regenerative capacity of islet β-cells to increase an individual’s insulin secretion capacity is among the promising approaches.

Recently, discovery-oriented unbiased chemical screening led to the identification of the metabolic enzyme adenosine kinase (ADK) as a regulator of rodent and porcine β-cell replication. Importantly, small molecule ADK-inhibitors selectively stimulate replication of β-cells, but not alpha-cells, in vitro and in vivo. The identification of ADK as a regulator of β-cell regeneration has raised several important questions.

Does long-term inhibition of ADK in vivo promote β-cell mass expansion and protect against developing diabetes? How does a broadly expressed metabolic enzyme selectively control the growth of a specific cell-type?

Is human β-cell growth controlled by ADK? Using the nPOD resources, we are investigating the expression of ADK in normal and diabetic pancreatic tissues. We are interested in weather this molecule is expressed in human β-cells and whether its expression changes with diseased states.

Mark Atkinson University of Florida Funded by: NIH Quantitative and qualitative evaluation of the endocrine & exocrine pancreas in type 1 diabetes

In the most widely recognized natural history model for type 1 diabetes, there is a presumption that all individuals begin life with the same number of beta cells. In combination with this, a consensus exists that the so-called “normal” adult pancreas possesses approximately one million islet cells. These concepts have drawn little attention or question over time as much of the research interest in terms of the disorder’s natural history has focused on genetic susceptibility and the trigger period, where the beta cells begin a process of linear loss.  However, a limited body of largely unnoticed data draws question to the notions of a normal pancreas as well as the number of islets and/or beta-cells that may be resident, both in early life through adulthood, in humans.

We (in association with Dr. Martha Campbell Thompson and Dr. John Kaddis) recently published data (JAMA, 2012) involving nPOD tissues suggesting that the pancreas of individuals with type 1 diabetes is smaller, consistent with other studies over the previous two decades. But, until recently, the observation drew little investigator attention.  Moving away from pancreatic size, the situation becomes even more intellectually diffuse when one considers quantification of islet and/or beta-cell numbers.  In studies of pancreatic development in both humans and rodents, it appears that a wave of beta-cell death occurs as a normal physiological process following birth.  The implications for this process may influence T1D’s pathogenesis in at least two ways. First is how it might influence T1D relates to the initiation of autoimmunity. A second notion as being of equivalent if not an even greater potential relates to the aforementioned concept of “100%”.   We believe it conceivable that a major and yet highly unappreciated risk factor for T1D resides in the number of beta-cells present in early life.   Taken collectively, these findings support a hypothesis that immune and perhaps non-autoimmune mechanisms (e.g., developmental, environmental, genetic, etc.) influence a loss of beta-cell mass as well as loss of pancreatic mass in T1D, possibly even before full-blown disease; thus, forming clear objectives for our studies.  

Susan Bonner-Weir & Anmar Khadra Joslin Diabetes Center Funded by: Rumsey Funds Origin of composite islet structure in humans

In the adult human pancreas the majority of islets have a clear non-random organization of mantle of non beta cells around beta cell cores, with larger islets usually having multiple such subunits (composite) whereas smaller islets resembled the classic rodent pattern. There is far more variability in islet composition both within each human pancreas and among different human pancreas than in rodent pancreas. The intrapancreatic variability in human pancreas raises important questions about how islets evolve and function throughout life and how this might relate to the pathogenesis of diabetes. We will examine the origin of the composite structure of the human islets by histological analysis of pancreas of donors from newborn to 25 years old. Determining the islet organization pattern (composite vs simple) with the distribution of islet size over the age range of expansion of the beta cell mass will provide data for mathematical modeling, which will predict the mechanism of formation of the composite islets.

Alexandra Butler, Sangeeta Dhawan University of California Los Angeles Funded by: National Institutes of Health/NIDDK and Helmsley Charitable Trust George S. Eisenbarth nPOD Award for Team Science Growth and Development of the Endocrine Pancreas in Humans: Extrapolating Normal Physiology to the Regenerative Capacity of the Endocrine Compartment in Type 1 and Type 2 Diabetes

Chromogranin A is a global marker of endocrine cells. Most endocrine cells in the pancreas, in addition to Chromogranin A, also express other hormones, Insulin and Glucagon being most frequent, with Pancreatic Polypeptide, Somatostatin and Ghrelin in smaller numbers of endocrine cells. We have found that in adult human pancreas a small number of pancreatic endocrine cells (<1% of the total number) express Chromogranin A but not any of the other known pancreatic hormones; we have termed these cells Chromogranin Positive Hormone Negative [CPHN] cells.

We found that there is a very high percentage of CPHN cells in sections from human fetus and newborn pancreas, which suggests that these cells are a precursor of mature endocrine cells. We also found that the percentage of CPHN cells is higher in pancreas from both type 1 and type 2 diabetic subjects, suggesting that this may be an attempt at restoring the beta cells lost as a consequence of disease.

Our initial observations are that these cells occur mostly in small clusters of 1-3 cells rather than in established islets, suggesting that these cells may be newly formed endocrine cells. The source of these CPHN cells is of great interest, as they could potentially occur de novo (neogenesis), by transformation of existing cells (transdifferentiation) such as from cells in the ducts or exocrine pancreas, or by replication of existing endocrine cells.

Our goal is to characterize these cells by looking at the frequency of replication and apoptosis , and the transcription factors Nkx6.1 and Nkx2.2, in the CPHN cell population. Our findings will have significance with regard to the regenerative capacity of human endocrine cells in human pancreas.

Martha Campbell-Thompson, Alberto Pugliese, Clive Wasserfall & Mark Atkinson University of Florida ORIGINAL: Beta cell regeneration for Type 1 Diabetes ADDENDUM: Neuro-insular pathways regulating beta cell regeneration

ORIGINAL: Understanding basic human beta cell physiology is crucial to targeted therapies and prevention strategies for type 1 diabetes. Clinical signs in type 1 diabetes start when beta cell function and/or numbers are unable to maintain adequate blood glucose levels. After onset of type 1 diabetes,  life-long insulin replacement therapy is needed. Factors regulating human beta cell regeneration are poorly understood. The purpose of these studies are to: 1) define pancreatic beta and alpha cell populations and how diabetes alters their function and numbers, 2) characterize potential autoimmune mediators of  islet inflammation (insulitis) and resultant beta cell death, and 3) correlate lymphocytic populations in pancreatic and nonpancreatic lymph nodes and spleen between  pre-diabetes and type 1 diabetic patients.

ADDENDUM: The goal of this addendum is to utilize large scale 3D human pancreas imaging methods for quantification of islet neuroendocrine cells and intra-pancreatic neurons and fibers. The autonomic neuro-insular network may be an unrecognized co-factor in the loss of functional beta cells during the progression to type 1 diabetes and following an islet transplantation. This proposal is expected to provide key data regarding neuronal pathways regulating beta cell regeneration.

Juan Dominguez-Bendala, Giacomo Lanzoni & Luca Inverardi University of Miami Funded by: Diabetes Research Institute Foundation Progenitors of the beta cell lineage in the pancreas and biliary tree of diabetic patients

Stemming from a collaboration with Lola Reid (University of North Carolina), we have described stem and progenitor cells in niches of the human biliary tree and pancreatic ducts. These cells can mature into pancreatic endocrine cells, including functional insulin-producing cells. Our data suggest that bile ducts, peribiliary glands, pancreatic ducts and pancreatic duct glands may comprise a continuous ramification of cells that are related in maturational lineages. We have identified an extended network of stem cell pockets that branch out from the more naïve stem cells in the biliary tree to more and more mature cells in the pancreas. The presence of these progenitors in adults strongly suggests that the human pancreas has the potential to replenish lost cells in normal homeostasis and disease. Preliminary data obtained through our nPOD collaboration with Ivan Gerling (University of Tennessee) suggest that islets at the early stages of Type 1 Diabetes (T1D) may be undergoing regeneration or neogenesis. We are: (1) investigating the presence of pancreatic endocrine stem/progenitor cells in T1D and T1D Medalist cases; (2) testing if the stem/progenitor cell pool is affected by (or responsive to) the pathology and (3) studying whether these progenitors are subject to autoreactive aggression or not. We are also studying progenitor cells in transplanted pancreata in the nPOD-T study group. The overall aim of the project is to understand if regenerative processes are ongoing in T1D, to clarify the grade of maturation of beta cells in T1D, and to determine the feasibility of autologous approaches aimed at beta cell regeneration in T1D patients. Our team is open to new nPOD collaborations; we offer expertise and technologies in the following fields: stem cell markers, pancreatic islet development, beta cell maturation, immunology, imaging, laser-capture microdissection.

Matthias Hebrok & Long Cai University of California, San Francisco Funded by: JDRF Investigating the de-differentiated state of the β-cell in human diabetic patient tissues

It is well accepted that compromised ß-cell function is an integral part of diabetes development.  Several recent studies, mostly using complex genetic mouse models, have supported the notion that a change in the ß-cell state or identity due to varying insults causes diabetes in the absence of ß-cell death.  Our work has focused on the involvement of the von Hippel Lindau/Hypoxia Inducible Factor (VHL/HIF) pathway, a master regulator of cellular response to diminished oxygen (hypoxia).  In a mouse model that stabilizes the HIF pathway despite normal oxygen levels within the ß-cell, we observe the development of diabetes which correlates strongly with a) a loss of the ß-cell differentiation state as determined by markers such as Pdx-1, Nkx6.1, MafA, Glut-2, and Insulin, and b) inappropriate activation of embryonic signaling pathways and genes.  Immunofluorescence analyses on islets from diabetic animals displayed strikingly reduced level of insulin as compared to control, normoglycemic littermates, which indicate the cause of diabetes to be insulin insufficiency.  The broad disruption of gene expression observed in diabetic mice suggests a loss of ß-cell identity (or dedifferentiation) as the underlying cause.  We have also shown that ectopic expression of a key transcription factor Sox9 (normally expressed during embryogenesis) in the ß-cell causes dedifferentiation and diabetes.  Our goal is to translate our findings from the mouse models of diabetes into the human condition.  It is becoming apparent that diabetic patients can be stratified into distinct cohorts (based on onset of disease, age, weight, race) and it is critical to identify what sub-population of patients develops the disease due to ß-cell dedifferentiation.  To this end, samples available from the nPOD provide an invaluable resource to a) identify patients who have compromised ß-cell identity, and b) correlate the findings from mouse models into human patients by identifying what signaling pathways or genes are erroneously regulated in these samples.  Mouse tissue allows us to conduct genome wide expression analyses that can lead to identification of important regulators to be tested in the human samples.  Our focus is on the maintenance of identity of the ß-cell, and discovering important regulators of this process may enable the development of intervention therapies that could block or delay the onset of ß-cell damage.

Pedro Herrera University of Geneva Medical School Funded by: NIDDK and JDRF Cell conversion in the human pancreas

The overall goal of my proposal is to identify means for improving β-cell regeneration in the adult pancreas. We previously developed a transgenic model of inducible total or partial β-cell ablation (termed RIP-DTR). We have reported that in these mice there is spontaneous reconstitution of new β-cells from heterologous (i.e. non-β) cells after near-total β-cell loss.

The RIP-DTR model has revealed an unsuspected degree of cellular plasticity in the pancreas of juvenile and adult mice, including aged individuals, regarding the spontaneous inherent capacity of islet a-cells to switch to insulin production upon β-cell loss.

During the next years we want to address the following fundamental questions:
1. What is (are) the signal(s) driving a-cell reprogramming upon near-total β-cell ablation?
2. Can the a-to-β-cell conversion be fostered? Why only a small fraction of a-cells engages into conversion? What is the nature of the epigenetic modifications in reprogrammed a-cells?
3. Can human acells reprogram to insulin production?
4. What is the influence of ageing on islet cell plasticity?
5. Can other islet cells, i.e. besides a-cells, reprogram to insulin production?
6. Can in vivo a-cell conversion be facilitated by means of compounds mimicking the effect of instructive signals?
7. Are human islets endowed with cell plasticity capabilities? How diabetic conditions influence i) the capacity of human/mouse a-cells to reprogram, and ii) the maintenance of the β-cell phenotype?
Seung Kim Stanford University Funded by: HHMI Investigating islet alpha-to-beta cell conversion in Diabetes

Diabetes occurs when insulin producing pancreatic β-cells are impacted: type 1 diabetes due to destruction of β-cells, and type 2 diabetes due to a decrease in their number and/or potency. Consequently, therapy that aims to replace β-cells is a promising long term alternative to the currently prevalent insulin injections. However, the difficulty in finding viable sources to generate functional β-cells remains a crucial bottleneck. Herrera and colleagues recently discovered that under conditions of near-total depletion of β-cells in mice, only α-cells within pancreatic islets spontaneously convert into β-cells1. Encouragingly, there is in fact an overabundance of α-cells in the pancreas of diabetes patients. However, today we lack the knowledge to initiate and maintain α-to-β-cell conversion and are thus unable to exploit the availability of abundant α-cells to orchestrate this process reproducibly in humans.

Exciting preliminary results from our laboratory and others reveal that a subset of pancreatic α-cells in T1D juvenile donors lose α-cell gene expression and acquire the expression of several key β-cell genes including Insulin. These data suggest that α-to-β-cell conversion may have occurred in these T1D donors. Further, we identified genes that likely control α-cell fate by recapitulating our T1D findings in mice that were engineered to delete specific genes (Dnmt1 and Arx) in their α-cells. The work proposed here aims to extend our findings in multiple T1D donor samples with long or short disease durations. To our knowledge, the question of whether α-to-β-cell conversion occurs in T2D has not been previously studied.

We therefore request pancreas samples from T2D donors to analyze them for evidence towards α-to-β-cell conversion. We will use state-of-the-art technologies to enable us to isolate α-cells undergoing conversion and probe them for changes in their genetic and epigenetic landscape. These studies are expected to a) Elucidate the similarities and differences between the fates of pancreatic α-cells in early and late stage T1D, and between T1D and T2D, and b) Identify new molecules and pathways that can be targeted for developing drugs to enhance the generation of functional β-cells.

Fred Levine Sanford-Burnham Medical Research Institute Funded by: CIRM Beta-cell regeneration by transdifferentiation

We have been interested in the mechanisms by which beta-cells regenerate in diabetes. Our laboratory is interested in how beta -cells regenerate under normal and pathophysiological conditions, with the goal of developing new therapies for diabetes that result in an increased number of those cells. There are two possible mechanisms for endogenous regeneration: beta-cell replication and beta-cell neogenesis. In 2013, we published that potent small molecule HNF4a antagonists induced beta-cell replication in mice and rabbits.

With respect to beta-cell neogenesis, we have developed a new model of pancreatic damage, combining pancreatic exocrine cell damage and chemical beta -cell ablation. In the model, high efficiency transdifferentiation of beta-cells to other islet cell types occurred. All of those experiments were done using mouse models. It is of interest to determine whether the findings in mice can be translated to humans. Thus, we propose to use samples from nPOD to examine for evidence of islet cell transdifferentiation.

Dorene Markel, Momenico Accili, Peter Arvan, Mark Atkinson, Jeffrey Bluestone, Matthias Hebrok, Kevan Herold, Chris Rhodes & Matthias von Herrath Brehm Coalition, University of Michigan Funded by: JDRF De-differentiation during progression of beta cell loss in type 1 diabetes

The overall goal of this proposal is to test the hypothesis that β-cell de-differentiation is an early stress-related response that may account for the early loss of β-cells in T1D. Furthermore, we believe that a small but significant population of long-term T1D patients might still have de-differentiated β-cells in their pancreatic islets, which creates hope for future re-differentiation paradigms. The ultimate goal is to address questions whose answers could lead to treatments that prevent β-cell loss and promote re-differentiation of progenitor-like cells into β-cells.

Maike Sander University of California, San Diego Preventing beta cell dedifferentiation in type 1 diabetes

Recent evidence suggests that beta cell dedifferentiation is a key early mechanism that underlies the pathogenesis of type 2 diabetes (T2D) [1]. This includes the loss of transcriptional regulators of beta cell function as well as other critical components of beta cell function. The study further suggests that beta cell dedifferentiation and acquisition of progenitor markers precedes beta cell death in T2D. Whether the same sequence of events and loss of factors occurs during progression of type 1 diabetes (T1D) is currently unknown. Interestingly, the analysis of pancreata from T1D individuals suggests that T1D patients retain insulin-producing cells even years after the onset of disease [2]. However, preliminary, still unpublished evidence suggests that these remaining beta cells have impaired health and have lost expression of some critical beta cells genes (Mark Aktinson, ADA meeting, June 2012). This observation raises the question of how function and mass of these remaining beta cells could be improved. We hypothesize that similar factors as lost in T2D are also lost in T1D and result in impaired function of remaining beta cells. 

The overall goal of this proposal is to identify pathways that can prevent and/or reverse beta cell dedifferentiation in type 1 diabetes. The strategy is to define upstream pathways that maintain the expression of critical beta cell differentiation factors. It is predicted that sustained expression of these differentiation factors will restore beta cell function and mass in type 1 diabetes.

Andrew Stewart, Nathalie Fiaschi-Taesch Icahn School of Medicine at Mount Sinai Funded by: JDRF and NIDDK Multidisciplinary approaches to driving human beta cell replication

Understanding and controlling human beta cell replication and regeneration are major goals of the JDRF specifically, and of the diabetes community in general. Whereas the field of beta cell replication is a relatively new one, and much new data has recently accumulated regarding the molecular control of rodent beta cells, very little is known regarding the molecular mechanisms that govern human beta cell replication. We have received support from the JDRF and the NIDDK to specifically define the presence and subcellular location of all human cell cycle control molecules in the human beta cell. The major cell cycle control point in all cells is the “G1/S checkpoint”, or the “pRb pathway”, which contains the cell cycle proteins pRb, p107, p130, p15, p16, p18, p19, p21, p27, p57, p53, cyclins A,D,E, and cdks 1,2,4,6. Until recently, very little has been done to elucidate the presence, absence or role of these molecules in physiologic human beta cell regeneration, nor in attempts at therapeutic human beta cell replication. We are therefore trying to develop a “Human G1/S Proteome” Atlas, a task we have largely completed using human cadaveric islets provided by the JDRF and by the IIDP. We have demonstrated that when we overexpress these cell cycle molecules using adenovirus, we can induce both robust human beta cell replication, as well as nuclear translocation of these molecules. Thus, this is an observation of great significance to beta cell regeneration investigators: it suggests that the proximate reason that human adult beta cells do not replicate and cannot be induced to replicate as that the requisite nuclear machinery does not, and apparently cannot, access the nucleus where it would drive beta cell replication. These are paradigm-changing observations, with major implications for therapeutic human beta cell replication and regeneration.\

ADDENDUM: Type 1 Diabetes results from the loss of the insulin producing beta cells. It is now clear that beta cell replacement can reverse diabetes in humans. One way to replace beta cells is to induce the regeneration of the endogenous beta cells or to expend beta cells from human donors for transplant. However, we know little about human beta cell replication. Therefore, information on human beta cell replication is critical for the development of new therapies for diabetes. Adult human beta cells proliferate at an extremely slow pace and are mysteriously resistant to the induction of replication. Rare examples of physiological replication of human cells are during the first years after birth and during pregnancy. Recently, we have made the novel and unexpected observation that many key cell cycle accelerator proteins are excluded from the nucleus in adult human pancreatic beta cells. It is possible to induce some replication of adult human beta cells by manipulating the cell cycle. However, this induction of replication is dramatically reduced during diabetes, such as Type 2 Diabetes. We propose to define the expression of cell cycle molecules in neonatal human beta cells, or human beta cells from pregnant donors or Type 2 diabetic donors. The overall goal of our research is to identify therapeutic targets to promote human pancreatic beta cell proliferation.

Bone Marrow Studies

Reza Abdi & Marwan Mounayar Brigham and Women’s Hospital Funded by: JDRF Intrinsic immunoregulatory defect of bone marrow mesenchymal stem cells (BM-MSC) from T1D patients

Type 1 diabetes (T1D) represents a major public health burden demanding innovative treatment strategies. Mesenchymal stem cells (MSC) have profound immunomodulatory effects. Numerous clinical trials have focused on the immunomodulatory capacity of MSC to treat various immune-mediated diseases.  We were amongst the first to show the protective effects of allogeneic MSC in autoimmune diabetes in NOD mice. More studied need to be carried out in particular on human MSC to unravel the mechanisms of immunomodulation of MSC to help in the design of MSC therapy in T1D.  We are fortunate to collaborate with nPOD to receive samples from T1D patients to isolate and grown MSC for our studies to examine their immunomodulatory capacity. Data to be generated using these samples are crucial in better understanding as how MSC suppress autoimmunity which would support the bases for These data provide important preclinical data supporting the basis for further development of MSC-based therapies for T1D and potentially, for other autoimmune disorders.

Maria Grant Indiana University Bone marrow progenitor cell (BMPCs) dysfunction in diabetes is mediated by reduced bioavailability of NO

Despite advances in understanding the pathogenesis of diabetic retinopathy (DR), the nature and time course of the molecular changes associated with DR remain incompletely understood. We propose to investigate the mechanism of bone marrow (BM) failure in diabetes and test diverse therapeutic approaches to correct BM and bone marrow progenitor cell (BMPC) dysfunction in order to restore retinal vascular health. Healthy BMPC participate in the process of re-endothelialization and also contribute indirectly to vascular repair by the secretion of paracrine factors. In health, distinct populations of BMPC work in concert to orchestrate vascular repair. The small pool of CD34+ cells has robust reparative function and modulates, via paracrine signaling, the function of the much larger pool of CD14+ cells which, depending on the local milieu, have the potential to also become efficient reparative cells. Freshly isolated BMPC can be expanded in vitro to generate two additional populations of reparative cells: early endothelial precursor cells (eEPC) or late outgrowth endothelial cells (OEC) that may have more therapeutic utility than freshly isolated cells. To date the optimal progenitor cell for vascular repair has not been identified.

In diabetes, BMPC demonstrate deficient reparative function and may even have a deleterious effect on the repair process. We demonstrated that: 1) diabetic CD34+ cells do not repair acellular capillaries in the diabetic retina, whereas CD34+ cells from nondiabetic individuals do (a similar observation was also made using CD14+ cells); 2) compared to nondiabetic CD34+ cells, diabetic CD34+cells have reduced bioavailable nitric oxide (NO) which results in their reduced migratory ability that can be corrected by restoring NO levels to normal; and 3) diabetes is associated with the loss of circadian rhythmicity of BMPC release from the BM into the circulation. Decrease in blood flow through the vasa nervorum, the small arteries that provide blood supply to peripheral nerves, has been implicated in the development of diabetic neuropathy. We postulate that in diabetes reduced NO initiates changes in the vasa nervorum and is a mechanistic contributor to reduced sympathetic neurotransmission which ultimately results in BM dysfunction. Loss of this neurotrophic support leads to changes in the BM microenvironment which contribute to, not only reduced release of these cells, but also to their impaired function. This dysfunction is further exacerbated by elevated reactive oxygen species (ROS) in these cells. Based on our findings that diabetic retinopathy is linked to peripheral neuropathy, we put forth the following hypothesis:

Loss of circadian rhythmicity of release of BMPC from the BM (due to dysregulated NO in the vasa nervorum and subsequent BM neuropathology) and increasing levels of intracellular ROS, increasing endogenous NOS inhibitors and inappropriate secretion of pro-inflammatory cytokines and proangiogenic growth factors compromise the reparative ability of BMPC leading to development of diabetic retinopathy.

Kevan Herold & Alfred Bothwell Yale University Funded by: JDRF Reconstitution of HHLS mice with bone marrow from patients with T1DM

The focus of our JDRF Autoimmunity Center is to create and study a humanized mouse model of Type 1 diabetes. The approach that we are developing involves use of immune deficient (γc-/-) mice with “knock-in” of human cytokine and other genes that are needed for optimal reconstitution with human cells. In addition, we have created transgenic mice that express human proinsulin in beta cells and in addition, have human MHC+ mouse MHC- mice in which the presentation and responses to human antigens (e.g. proinsulin, insulin) can be studied. We hope to reconstitute the immune deficient mice with human CD34+ bone marrow cells from patients with diabetes. The duration of diabetes should not be an issue so the use of nPOD tissue would be appropriate. We plan to study the ability of the reconstituted mice to develop diabetes – spontaneously or with provocateurs such as low dose streptozotocin, immunization with antigens, or even transgenic expression of B7.1 on beta cells. (adopted by nPOD staff)

Paolo Madeddu & Costanza Emanueli University of Bristol, UK Funded by: Wellcome Trust Microangiopathy in diabetic bone marrow

We hypothesize that the alterations seen in diabetic BM are a major cause of impaired homeostasis in other vascular beds. Main objectives (1) to obtain a deeper insight into the cellular and vascular composition of diabetic BM using IHC, flow cytometry, and expressional studies (2) to determine whether BM microangiopathy is associated to poor metabolic control and increased susceptibility of cardiovascular complications. (adopted by nPOD staff)

Shannon Wallet University of Florida Characterization of Mesencymal Stem Cells in T1D

The development of MCS-based therapies for autoimmune diseases, including T1D, is a new field with a lot of promise, but also a lot of uncertainty. While successful treatment of disease allogeneic MSCs has been achieved, some MSC treatment failure has also been reported. The significance of this proposal is that it will fill a significant gap in the current understanding of mechanisms associated with human MSC-based therapies for T1D. These studies raise critical issues such as which patients could benefit from MSC treatment and when autologous or heterologous MSCs are best to be utilized. Here we propose to investigate whether T1D-derived MSCs are as effective in suppressing inflammation as non-T1D derived MSCs. Importantly, we are proposing to evaluate this using human samples as there is murine data to suggest that T1D-derived MSCs are genetically altered which affect their efficacy as immunosuppressive therapies.

Core Lab

Mark Atkinson University of Florida Funded by: JDRF, NIH Pancreatic immunologic and metabolic parameters

The Organ Processing and Pathology Core (OPPC), located at the University of Florida, receives nPOD donor tissue directly from OPOs, processes the tissue, ships samples directly to investigators and stores case samples and data.

Graeme Bell University of Chicago MODY mutation testing

Perform Maturity Onset Diabetes of the Young (MODY) testing for select nPOD samples.

Anil Bhushan, Mark Atkinson, Todd Brusko & Bita Brodbar University of California, San Francisco Funded by: JDRF Epigenetic analysis of nPOD islet cells

Type 1 Diabetes (T1D) is a complex autoimmune disease affecting more than 30 million people worldwide. T1D is caused by a combination of genetic and epigenetic factors leading to immune destruction of insulin-secreting islet cells. Although significant progress has recently been made in elucidating the genetics of T1D, the non-genetics component has remained poorly defined. Multiple lines of evidence including studies of migrant populations and monozygotic twin pair discordance for T1D underline the role of epigenetic factors in this disease. Thus, the characterization of epigenome is critical for the development of new T1D therapeutics and for appropriate staging of disease. DNA methylation is one of the main epigenetic modifications involved in human disease. Using epigenome-wide analysis such as reduced representation bisulfite sequencing (RRBS), we aim to identify unique differentially methylated regions (DMRs) within the genome that accumulate in PBMCs isolated from circulating blood, that may reflect the immune insult that occurs with T1D onset. Then we are interested to study the role of these hypo or hyper methylated regions in T1D.

Santica Marcovina University of Washington Northwest Lipid Research

The Marcovina lab performs C-peptide testing for all nPOD samples.

Janelle Noble Children’s Hospital Oakland Research Institute High resolution HLA typing

The Noble lab produces HLA genotyping for all nPOD samples.

Stephen Rich University of Virginia Funded by: NIDDK Immunochip Assays

nPOD participated in the ImmunoChip consortium, that was established to design a cost effective genotyping array in order to fine map well established GWAS reported risk loci in immunologically related human diseases, including type 1 diabetes.

Clive Wasserfall University of Florida UF AAb Core

The Autoantibody Core at the University of Florida serves as the coordinating center for nPOD screening lab partners (LABS,Inc and Viracor-IBT Laboratories). Combi-kit manufacturing and quality control/assurance measures are managed here. Additionally, the UF Autoantibody Core tests all nPOD samples for autoantibodies to GAD, IA-2, and ZnT8 by ELISA.

Liping Yu University of Colorado-Denver Denver AAb Core

The Autoantibody Core at the University of Colorado works with the Autoantibody Core at UF for quality control/assurance and tests all nPOD samples for autoantibodies to GAD, IA-2, Insulin, and ZnT8, by RIA.


Reza Abdi & JianJun Cheng Brigham and Wowen's hospital Use of T1D pancreastic and lymphoid tissue to assess the expression of adhesion molecules

While immunoregulatory therapies hold great promise for the treatment of T1D, their inadequacy and serious toxicity have limited research efforts to find a lifelong immunosuppression approach. This has motivated investigators to search for alternative approaches such as the use of nanotechnology to deliver a wide variety of drugs and biomolecules. In recent years, there have been major strides to address various obstacles in order to bring nanotechnology closer to clinical practice. Nanotechnology has been moving forward on two fronts: 1) synthesis of a NP which efficiently loads, retains and releases drugs and 2) development of an innovative strategy to deliver NP to specific sites.   Much progress has been made on NP surface properties and size to reduce burst release, blood opsonization and clearance to enhance the circulation half-life of NP from several minutes to tens of hours.

On the front of delivering NPs to specific tissue sites, we have witnessed less progress. To make our nano-delivery approach applicable to Type 1 diabetes (T1D), we (Abdi and Cheng) have made major strides as follows: (A) we have successfully designed nanomedicines with the desired surface properties and controlled particle sizes to improve their pharmacokinetics, (B) demonstrated that loading and controlled release of the drug is markedly improved by developing a polylactide-drug conjugated nanoparticle, or nanoconjugate, via drug-initiated polymerization of lactide followed by nanoprecipitation, (C) our published data indicate that our nanotechnology platform is highly efficient in delivering immunosuppressive drugs, and importantly..

(D) Importantly, our preliminary data indicate selective PNAd dependent delivery of the nanoparticles to the pancreas and pancreatic lymph nodes in NOD mice. We are interested in examining the expression of PNAd in pancreatic lymph nodes (PLN), pancreas, non draining LN and spleen by immunhistolgoy to assess the feasibility and potential success of our targeted therapy in humans.

Mark Stuart Anderson University of California, San Francisco Funded by: JDRF Identification of novel Tissue-Specific Antigens expressed by human Extrathymic Aire-Expressing Cells, and determination of their potential contribution to the prevention of type 1 diabetes

Aire+ cells were readily identified by immuno-staining in both PN and nPLN sections. Like in mouse, human eTACs were relatively rare but localized outside the B-cell follicles and stained with the hallmark “nuclear speckling” pattern. Furthermore, the eTACs were ubiquitously positive for MHC class II but lacked high expression of CD11c, calling into question their identification as a known resident DC population. Quantitative PCR confirmed the expression of Aire in whole PLN and in sorted CD45+, MHC class II+ populations from fresh human spleen. No difference was found in the relative expression of Aire between T1D, no diabetes, and pre-T1D PLN specimens. However, the expression of ladinin-1, a putative Aire-regulated TSA in mouse eTACS, was positively correlated with Aire expression, suggesting that it might be an Aire-regulated TSA in humans as well.

Overall, our data confirm that Aire is indeed expressed outside the thymus in humans and that human eTACs appear analogous to the more fully characterized murine eTACs. Their potential for promoting peripheral tolerance in a therapeutic setting remains an exciting possibility.

Manuela Battaglia San Raffaele Scientific Institute, Italy Funded by: JDRF Dissecting the effector/regulatory compartments in the target tissues of T1D

An altered balance between pathogenic and regulatory pathways in autoimmunity has been hypothesized and at times demonstrated in peripheral blood of patients with type 1 diabetes. However, our knowledge in humans at the sites of the autoimmune attack (i.e., the pancreas) is scanty. Studies generated in our lab. demonstrate that FOXP3+ T regulatory (Treg) cells, identified based on their epigenetic imprinting, are present within the pancreatic lymph-nodes of patients with type 1 diabetes but they lack a suppressive activity towards both polyclonal and insulin-specific effector responses. We have found that there is an important discrepancy in Treg cells of T1D patients between their epigenetic imprinting, FOXP3 expression and regulatory function when these cells reside in the draining lymph nodes but not when they circulate in the blood.

Given the crucial role that microRNA (miRNA) genes have demonstrated in modulating immune responses, it is possible that dysregulation of miRNA-expression contributes to the pathogenesis of autoimmune diseases, type 1 diabetes included, as recently demonstrated in other autoimmune diseases (e.g. multiple sclerosis).

Objectives of our work are: (i) to characterize the miRNA expression profile in Treg cells residing in different tissues of the same donor (healthy and with type 1 diabetes), and (ii) to understand whether Treg cells residing in the lymph-nodes of T1D patients are post-transcriptionally regulated in a different way than those isolated from healthy donors. Results from this study may be instrumental for our comprehension of type 1 diabetes and for the development of novel therapeutics to target miRNAs to modulate immune responses.

Jeffrey Bluestone & Armando Villalta University of California, San Francisco Vascular endothelial growth factor receptor signaling regulates the pathogenesis of type 1 diabetes

1. To test the hypothesis that vascular endothelial cells in the pancreas of recent-onset diabetic patients have increased expression of VEGF receptors (VEGF-R).
2. To examine the expression of VEGF or VEGF-R on islet-invading CD4+ T lymphocytes.
3. To examine the utility of TCR repertoire between different T cell subsets as a marker of T1D disease initiation and progression by high throughput sequencing.

Todd Brusko University of Florida Funded by: University of Florida Molecular Signature of Autoimmune T cells in Type 1 Diabetes and Treg TCR deep sequencing. Detect Cd226+ Treg and Teff ininsulitis patient samples

The Brusko Lab is generally interested in the cellular immune response involved in the disease process leading to type 1 diabetes (T1D). A strong genetic association between T1D and the MHC class II gene region of the adaptive immune system has implicated coordinated T and B cell responses as a central mediator of the disease process. To date, the majority of studies in humans have been limited to investigations using peripheral blood samples. While these studies have identified a number of T cells reactive to autoantigen targets (e.g., insulin, glutamic acid decarboxylase (GAD), islet antigen-2 (IA-2), zinc transporter-8 (ZnT8), among others), they have been unable to identify clonal signatures unique to T1D patients. In an effort to generate a disease specific signature, our lab initiated studies to isolate CD19+ B cells, CD8+ T cells, CD4+ T conventional cells (Tconv), and CD4+CD25+CD127-/lo Tregs from nPOD-donor tissues including spleen, pancreatic draining lymph nodes, irrelevant draining lymph nodes, and peripheral blood. From this donor material, we will generate a molecular profile of T1D including a characterization of the T cell receptor (TCR) and B cell receptor (BCR) repertoire, global DNA methylation profile, and transcriptional signature.  We believe that this effort will lead to novel insights into the specificity and functional properties of immune cells that lead to autoreactive beta cell destruction. Of particular interest, we have been focused on a T cell costimulatory pathway involving CD226 or TIGIT interacting with CD155. We believe the molecular signatures identified in cell subsets isolated from the site of autoimmune attack (e.g., the islets and draining lymph nodes) will provide insight into the cause of immunoregulatory defects leading to T1D. 

John Cambier & Peter Gottlieb University of Colorado, Denver Tracking the loss of anergic B cells in pre-diabetic and new onset T1DHLA Testing

Insulin-specific B lymphocytes play an important role in development of autoimmunity in Type 1 Diabetes (T1D). Since potentially offensive autoreactive cells are silenced by mechanisms of immune tolerance, participation of insulin-binding B cells (IBCs) in T1D must reflect escape from this silencing. Examining this question, we found that those IBCs bearing antigen receptors with high affinity for insulin occur only in the anergic BND (naive IgD(+)IgM(-) B cells [Duty, 2009 J Exp Med. 16: 139-51]) B cell compartment in peripheral blood of healthy subjects. Importantly, these cells are absent from this compartment in some “at risk” first degree relatives and all pre-diabetic and new-onset (<1yr) T1D patients. BND cells were found at normal levels in individuals diabetic for >1 year. Interestingly, these changes were associated with loss of the entire BND compartment. By extrapolation from parallel studies of in diabetic NOD mice, we believe that under these circumstances B cells relocalize in tissues rich in their autoantigen, where they become activated and can present antigen to T cells. These findings suggest that environmental events such as infection or injury may, by disrupting B cell anergy, dispose individuals to autoimmunity, the precise tissue target of which is specified by genetic risk factors.

Martha Campbell-Thompson, Lizette Duckworth-Vila, Ralph Hruban, Gunter Kloppel, Chen Liu, Nickolas Papadopoulos & Mark Atkinson University of Florida Diabetes and Pancreatic Cancers: Pilot study with molecular profiling

Our study aims are the following:
1. Histopathology review of pancreata from patients without diabetes (no diabetes and autoantibody+) and
those within the diabetic and other groups (type 1, type 2, gestational, other (those with unclassified diabetes,
excludes cystic fibrosis)).
2. Molecular profiling of PanNET for neuroendocrine hormones and other genes (Isl1, p53, Ki67), FISH
(telomerase shortening), MEN1 mutations (DNA), and gene expression (RNA). Assays will be conducted at the
University of Florida and Johns Hopkins laboratories.

Barbara Coulson The University of Melbourne at the Peter Doherty Institute for Infection and Immunity Funded by: Agency: Australian Government-National Health and Medical Research Council Analysis of type 1 interferon responses to rotavirus in patients with type 1 diabetes

Rotavirus infection of children at-risk of type 1 diabetes is associated with disease progression. Similarly, infection of adult non-obese diabetic (NOD) mice with Rhesus monkey rotavirus RRV accelerates diabetes onset. This acceleration correlates with RRV spread to the mesenteric and pancreatic lymph nodes, where virus associates with and activates dendritic cells and induces B and T cell activation. We also have shown that RRV exposure ex vivo induces generalized B and T cell activation by triggering toll-like receptor 7 (TLR7) signaling and type I interferon production by plasmacytoid dendritic cells (pDC) from NOD mice. As this activation includes islet autoantigen specific T cells, we hypothesize that type I interferon-mediated bystander activation contributes to diabetes acceleration by RRV in NOD mice. Notably, RRV induces significantly more activation of NOD mouse than C57BL/6 mouse cells, providing evidence that these responses to RRV are heightened in mice at-risk of T1D. Influenza virus and coxsackievirus B, which also signal through TLR7, have been shown to induce heightened pDC-dependent type I interferon responses in PBMCs from T1D patients compared to controls. We hypothesize that rotavirus will also induce heightened pDC-dependent lymphocyte activation and type I interferon secretion in PBMCs and other immune cells from patients with T1D compared to controls. Therefore, the overall aim of our study is to evaluate activation and type I interferon expression following stimulation with rotavirus in immune cells from patients with T1D compared to controls.

Howard Davidson University of Colorado Health Science Center Funded by: JDRF Characterization of autoantigen-specific T cell receptors from PLNs

Type 1A Diabetes mellitus (T1D) is a predominantly T cell mediated autoimmune endocrine disease. Both CD4+ and CD8+ T cells reactive to diabetic autoantigens can be identified at low frequency in the peripheral blood of diabetic subjects and their at-risk relatives, but the extent to which these peripheral responses reflect those in the pancreatic draining lymph nodes (PLN) and insulitic lesions is currently unclear, mainly due to the relative inaccessibility of these tissues in living subjects. Answering this question is of central importance to the development of mechanistically relevant biomarker assays for staging disease and monitoring therapeutic efficacy, as well as for identifying appropriate targets for antigen specific immunotherapy. The goal of our project is to isolate and characterize T cells reactive to the major T1D autoantigens (with special focus on insulin, GAD65, IA-2, IGRP, and ZnT8) from the PLN and spleen of nPOD donors. The frequency, phenotype, epitope specificity and clonal diversity of islet-reactive T cells from these tissues will be compared with those present in islets and peripheral blood from the same donors (when available), and in PBMCs from HLA-matched living subjects at various stages of disease. 

Mark Davis & Yu Wong Stanford University Funded by: JDRF and Howard Hughes Medical Institute Analysis of the antigen-specific T cell repertoire in T1D patients

The majority of human T cell antigens relevant to T1D have been defined using algorithms which predict high affinity peptide binding to MHC.   However, direct evidence showing that these antigens are presented to T cells by human islets and APCs in T1D patients has not been established for the vast majority of these peptides.  A major goal of our project is to identify T cell peptide antigens relevant to T1D which meet the following criteria: 1) the peptides are naturally processed and presented by human islets/APCs and 2) there are T cells specific for these antigens in the PLN and/or islets of T1D patients.   To achieve this goal, we are eluting peptides from class I MHC on HLA-A2+ human islets and identifying them by mass spectrometry to define naturally presented antigens.  We are also verifying natural antigen presentation by using antigen-specific T cell lines as “biosensors” of antigen presentation on islets and APCs.  Islet-specific peptides that are shown to be naturally presented will be used to make MHC-peptide tetramers and analyzed by CyTOF technology to measure the frequency, phenotype, and function of T cells specific for these antigens in T1D pancreatic draining lymph nodes obtained from the nPOD tissue bank.  By verifying the natural antigen presentation of islet-derived peptides and the presence of activated T cells specific for these peptides in T1D patients, we will be able to differentiate between peptide antigens with bonafide relevance to T1D from islet peptides that are irrelevant to T1D.  CyTOF analysis of the frequency, phenotype, and function of pathogenic T cells specific for disease-relevant islet antigens in T1D patients should be informative with respect this disease, potentially enabling sophisticated and early diagnoses and the ability to assess the efficacy of therapeutic intervention in human patients.

Marc Donath, Thierry Nordmann & Elise Dalmas University Basel, Switzerland Identification of islet associated immune cells in type 2 diabetic patients and further exploration of their role in disease progression and severity

Growing evidence suggests inflammation as a major underlying mechanism in the pathogenesis of type 2 diabetes mellitus (T2DM). Supporting the involvement of the immune system in T2DM, cross-sectional and prospective studies have associated elevated circulating levels of acute-phase proteins. Furthermore, clinical studies have demonstrated that anti-inflammatory drugs may improve glycaemia. Morphological and therapeutic intervention studies have uncovered an inflammatory process in islets of patients with type 2 diabetes characterized by the presence of cytokines, macrophages, β cell apoptosis, amyloid deposits, and fibrosis. However, the immune cells involved in pancreatic islet inflammation during T2D are insufficiently described. In human tissue samples, only CD68 positive macrophages have been documented. The level of activity of these cells as well as the occurrence and role of other immune cell types remain to be investigated.

Using the pancreatic tissue samples supplied by nPOD, we plan on investigating the presence of innate and adaptive immune cells within or around the islets of T2DM patients versus healthy donors. Defining the inflammatory milieu within the islets of T2DM patients will help us target further inflammatory treatments. 

Garry Fathman Stanford University Microarray analysis of PLN from autoantibody positive donors

As with many autoimmune diseases, T1D is initiated by some unknown inciting event (or events) that results in the appearance of disease-relevant biomarkers.  Serum autoantibodies to islet antigens or changes in whole blood gene expression that appear before the onset of glucose intolerance can serve as biomarkers to identify pre-diabetic individuals. As we enter the era of intervention trials to block progression of T1D before beta cell destruction, it is important and beneficial to identify relevant biomarkers that can be used to stage early disease progression in T1D. Recent studies from my lab have demonstrated that changes in gene expression of pancreatic lymph node (PLN) tissue samples from AA+ individuals provided by nPOD, can be seen before the onset of abnormal glucose tolerance.  These data also demonstrated that gene expression in the PLN is similar between T1D patients who have had hyperglycemia for two or more years and normal subjects, suggesting that the inflammation accompanying progression to hyperglycemia in AA+ subjects can be identified by changes in gene expression in AA+ subjects, and may be used to identify biomarkers, and/or suggest targets of, or response to therapy.  Because PLN tissues cannot be readily biopsied from AA+ individuals, our rationale is to examine gene expression in PBC samples provided by the TrialNet Natural History tissue repository to ask if there are distinct changes in gene expression seen in the PBCs (similar to those seen in the PLN) that occur with the progression of disease. As a second project arising from these data, we will ask if there are changes in gene expression that accompany intervention therapy with CTLA4-Ig in a TrialNet intervention trial being run by Dr. Bill Russell at Vanderbilt.  These approaches may help identify early biomarkers of disease that will allow sub-setting of AA+ patients based on stage of disease or rate of progression, and could identify prediction of, or response to, therapy.

Brian Fife & Marc Jenkins University of Minnesota Funded by: Minnesota Partnership for Biotechnology and Medical Genomics, JDRF and NIH Tracking islet autoantigen specific CD4+ T Cells

The goal of this research project is to identify self-reactive T cell clones and monitor the frequency and activation state during the progression of autoimmune diabetes.  Identification of T-cell reactivity to beta-cell antigen epitopes is important to study pathogenesis and monitor antigen specific interventions in T1D.  We have successfully generated and validated human tetramer reagents to diabetes relevant target antigens to determine frequency and activation state of autoreactive CD4+ T cells in patients with diabetes. Tetramer reagents could provide much needed biomarker reagents to predict T1D earlier in ‘at risk’ patients, stage clinical disease, and determine if therapeutic approaches have an effect on the CD4+ T cell level during T1D pathogenesis.  If successful, the tetramer biomarkers and enrichment approach we are testing could precede autoantibody production and provide an earlier diagnosis of T1D, before irreparable damage has occurred to the islets.  Therefore, these reagents would have great impact to provide a new in vivo diagnostic approach to identify T1D T cell responses as an early diagnosis for high risk patients.  With a better understanding of beta cell specific targets we will be able to develop novel therapeutic approaches to cure diabetes.  

Georgia Fousteri San Raffaele Research Institute Funded by: Ministero della Salute, Giovani Ricercatori T follicular helper (TFH) and regulatory (TFR) cells in T1D

Type 1 Diabetes (T1D) is characterized by the gradual loss of insulin-producing beta cells, which are eliminated by autoreactive cells infiltrating the pancreas. In humans and mouse models of the disease, T1D susceptibility is determined by genetic factors and is greatly influenced by environmental triggers (1). Progression to T1D is usually preceded by a period of anti-islet autoantibody production, which often can last for years (2). Today, autoantibodies are most widely used as serum biomarker (3), but T-cell readouts and metabolome studies are underway as they may strengthen diagnosis, prognosis and most importantly bring forward effective protocols for prevention and cure. T follicular helper (TFH) cells are essential for the formation of germinal centers (GCs), specialized structures in secondary lymphoid organs where maturation of B cells into high-affinity plasma cells and long-lived memory B cells takes place. It has become clear that precise control of the GC response is important for the production of optimal antibody responses that are devoid of selfreactivity. The exact mechanism by which autoreactive B cells are eliminated during this process is not fully understood, however the number of TFH cells together with a specialized subset to FOXP3+ regulatory T cells, namely T follicular regulatory cells (TFR), finely control the survival of pathogen-specific high-affinity B cells. It is currently not known whether TFH or TFR cells are involved in the production of islet-reactive, autoantibody-producing B cells in patients with T1D. However, given that autoantibodies emerge prior to disease onset, the effect of TFH cells in T1D progression can be inferred. With this study we aim at identifying whether TFH and TFR cells are implicated in autoantibody development in T1D. Investigating this mechanism poses a cardinal opportunity to increase the pipeline of therapeutic targets and to improve the number of disease biomarkers.

Rachel Friedman National Jewish Health Funded by: Helmsley Charitable Trust George S. Eisenbarth nPOD Award for Team Science Mononuclear phagocyte populations in T1D islets

Little is known about how T cell pathogenesis is modulated in the islets during T1D, yet this is a critical obstacle for preventing destruction of remaining islets or beta cell grafts. Therefore, we must elucidate mechanisms that promote and prevent the pathogenic autoimmune response in the islets. Our data in the NOD mouse model show that different populations of CD11c mononuclear phagocytes have countervailing roles in restimulating or inducing tolerance in islet-infiltrating T cells as islet infiltration progresses. While macrophages have been identified in human pancreatic islets, these likely represent a complex assortment of several mononuclear phagocytic cell populations that play a role in islet destruction. As a result, our goal is to identify the populations of mononuclear phagocytes within the islets of normal, autoantibody positive, and type 1 diabetes subjects.

Kathleen Gillespie University of Bristol Funded by: Diabetes UK Maternal microchimerism in T1D pancreas

Maternal microchimerism (MMc) is acquired by an infant during pregnancy and these cells are maintained in some individuals for decades.  Encountering maternal antigens during pregnancy represents the first immunological challenge for the fetus although maternal cells are protected from detection by induction of suppressive fetal regulatory T cells.

In type 1 diabetes increased levels of maternal microchimerism have been demonstrated in peripheral blood DNA and maternal cells have been identified in human pancreas. The aim of this study is to phenotype these maternal cells in well characterised male pancreases in order to help understand how the presence of maternal cells could contribute to risk of T1D. The methodology used will utilise using X/Y chromosome FISH with concomitant immunofluorescence.  Once the phenotype is established, we will attempt to confirm the maternal origin using single cell laser capture with DNA profiling. 

Kathleen Gillespie & Abby Willcox University of Bristol Funded by: Diabetes UK Molecular insights into the type 1 diabetes pancreas

Pancreatic lymph nodes (PLN) are the central location for the autoimmune response that leads to type 1 diabetes (T1D), yet no histological studies of this crucial step in pathogenesis of human diabetes exist. In this study, we request sections from T1D and matched control PLN to allow us to follow up our preliminary observations showing that replicating germinal centers are diminished in recent onset T1D. In particular, we will focus characterizing the structure and cellular content of PLN in T1D and will test the hypothesis that lack of germinal center replication reflects attempted regulation of the immune response.

Dale Greiner & Leonard Shultz University of Massachusetts Diabetogenic function of autoimmune donor splenocytes in humanized mice

Our understanding of type 1 diabetes (T1D) has been advanced greatly by studies carried out using mice and rats. However, rodents are not humans and progress in the understanding of the pathogenesis of T1D in humans has been impeded by the a lack of assays that permit the direct in vivo analysis of diabetogenic human T cell populations without putting individuals at risk. The availability of spleen, bone marrow, and thymus samples from T1D and islet autoantibody-positive donors provides an opportunity to analyze the diabetogenic function of these cell populations following engraftment in newly developed humanized mouse model systems. This work is based on our recently developed immunodeficient mice that express human HLA molecules and are optimized for human immune cell engraftment. Our goal is to use unique tissues available from nPOD to recapitulate autoimmune T1D in humanized mice. We will engraft human hematopoietic stem cells and immune cells into HLA-matched NSG mice and determine the ability of cells from T1D donors to attack islets and induce the destruction of insulin-producing cells. The knowledge obtained from these studies will permit us to elucidate the pathogenesis of T1D and to develop new approaches for preventing or curing the disease.

Abdel Hamad & Thomas Donner Johns Hopkins University Fas Ligand: Unorthodox target for prevention of type 1 diabetes

Despite improvement in insulin delivery, maintaining tight control of glucose homeostasis continues to be a challenge that results in bouts of severe hypoglycemia and hyperglycemia and serious long term complications in many patients. Therefore, developing an immunotherapy for type 1 diabetes (T1D) remains a major goal. Reaching this goal requires detailed knowledge of the diabetogenic process. Particular knowledge of the molecules that play pivotal roles in driving the pathogenic mechanisms but are not required for mounting normal immune responses could lead to uncovering of safe therapeutic targets. Fas ligand (FasL) is a rare example of such molecules. It is an apoptosis-inducing member of the TNF family that is not required for immune responses. Yet it plays a critical role in driving the diabetogenic process as non-obese-diabetic (NOD) mice bearing spontaneous loss-of-function mutation of FasL (gld) of Fas (lpr) are completely protected from the disease. We hypothesize that aberrant FasL-mediated apoptosis of IL-10-producing regulatory B cells is critical for initiating the diabetogenic process. This hypothesis is supported by strong published and unpublished data in the NOD mouse model. By working with JDRF/nPOD, we are investigating whether a dysregulated Fas pathway is compromising the role of IL-10-producing regulatory B cells in humans. We have collected strong evidence, so far, indicating that, as in NOD mice, aberrant FasL-mediated apoptosis may a pathogenic mechanism in T1D. Such results will not only provide new insights into the disease process, but also a platform for novel therapeutic modalities that do not impair the patient host defense.

Dolores Jaraquemada, Mercè Martì & Carme Roura-Mir University of Barcelona, Spain Tracing effector and regulatory T cell populations in type 1 diabetes

Our group is studying T cell phenotype and antigen presentation and recognition by T cells in T1D, with three main objectives. (i) The characterization of role of T cells involved in immunoregulation, i.e. regulatory T cells (nTregs or iTregs) and NKT cells. We are studying the mechanisms of regulation by analyzing the role of soluble and membrane bound molecules on the regulatory effect of each cell type; and the cooperation between regulatory populations to achieve a higher degree of suppression of T effector cells. (ii) The characterization of the TCR repertoire both on the pancreas infiltrate and on PBMCs from T1D patients. TCR expressed by autoreactive T cells is important as a tracer of the different clonotypes involved in T1D pathogenesis. We are also working on the hypothesis that some of the T cells involved in the maintenance of the chronicity share the TRBV-CDR3 as public TCRs. If proven, TCR repertoires prone to autoimmunity could be defined when accompanied with the HLA-associated alleles. Expanded iNKT cells will be analyzed as a model of public TCR in those patients. (iii) The third aim addresses antigen recognition by autoreactive T cells by the characterization of peptides presented in situ by HLA molecules. We are now able to retrieve a good number of peptides from small samples (i.e. islets). We will compare the T-cell epitopes already identified in T1D with those naturally presented in the islets by HLA class I and class II molecules. Because of the low number of HLA-II positive cells in the islets, we expect results for class I and are working on alternative approches for class II.

The first part of this project was in part funded by the JDRF and we are now funded by the Spanish Department of Science.

Sally Kent University of Massachusetts Medical Center Funded by: JDRF Investigation of B cells in human islets and PLN in T1

Our goal is to understand the role of autoreactive B cells in tissues draining the pancreas and their role in the type 1 diabetes (T1D) autoimmune response. While B cells are responsible for secreting autoantibodies, B cells have other functions including secretion of effector cytokines, antigen presentation and costimulation of T cell responses. Using a multiplexed, single cell interrogation assay called microengraving, we saw, ex vivo, that there is a statistically more frequent number of immature (IgM secreting) CD20 B cells secreting antibodies reactive with islet antigen GAD65 in the spleen from 50% individuals examined with recent onset (0-7 years) T1D as compared to those from healthy control: these cells could also be seen on the spleen of some GADA+ individuals without disease and in some individuals with long-term T1D. We saw this reservoir of GADA-secreting B cells of a mature phenotype (secreting IgG1, IgG3 or IgA) in the spleen of two individuals with recent onset T1D. We saw a low frequency of immature CD20 B cells secreting antibodies reactive with proinsulin in all groups. We wished to interrogate the autoreactive B cells for their cytokine secretion, but due to cell demise in the single cell assay at time points greater than 24 hours, we developed a 7-day culture assay that allows for survival of B cells and identification of these cells by their autoantibody secretion and cytokine profile in a two-color ELISpot assay. We are interrogating spleen samples from control individuals, individuals with autoantibody, but without disease, individuals with recent onset T1D (0-7 years), and individuals with long-term T1D (>7 years) for co-secretion of GAD65 autoantibody and IL-6/IL-10 or LTa/TNFa. These studies will provide a profile of the function of autoreactive B cells from individuals with T1D and their role in the autoreactive disease process.

David Leslie University of London Risk of autoimmune disease and human self-antigen expression

We have identified type 1 diabetes (T1D)-associated differential DNA methylated variable positions (T1D-MVPs) in CD14+ from peripheral blood mononuclear cells (PBMCs) from monozygotic (MZ) twins discordant for T1D (Rakyan V et al.; Plos Genetics 2011). We aim to determine the role of T1D-associated MVPs in the mechanisms that underpin immune dysfunction in T1D. We have therefore studied whether the same disease signature is detectable in T1D tissue compared with normal tissues relevant to the disease, namely: pancreatic lymph nodes (site of self-reactive T cells priming) compared with the spleen (as control). In addition we are examining the methylation state of T1D-MVPs in selected cells from human thymus and we intend to make these cells available to nPOD in due course.

Sara Michie & Baohui Xu Stanford University Tissue-selective chemokines and adhesion molecules in human T1D

Migration of lymphocytes from blood vessels into pancreatic lymph nodes (PanLNs) and pancreas is critical for the development of islet inflammation and beta cell destruction in type 1 diabetes (T1D). Adhesion of lymphocytes to the luminal surface of blood vessel endothelia in PanLNs and pancreas is an important step in this migration. However, the adhesion molecules that control migration of lymphocytes into human PanLNs and pancreas during the development of T1D are not well defined.

The goals of our project are to determine which endothelia adhesion molecules, and their lymphocyte receptors, are highly expressed in human PanLNs and pancreas during development of T1D, as compared to pancreas and PanLNs of controls (Aims 1 and 2). The physiologic importance of adhesion molecules that are expressed on lymphocytes and endothelia of PanLNs and pancreas during development of T1D will be determined using Stamper-Woodruff in vitro binding assays (Aims 3 and 4). Successful completion of these studies will help define the immunological mechanisms that are important for the development of islet inflammation and beta cell destruction in human T1D.

Maki Nakayama & Joe Larkin, III University of Colorado Dynamic Survey of T Cell Repertoire Targeting Pancreatic Beta Cells

Our project aims to define islet-specific T cell receptor (TCR) sequence repertoires directly from pancreas to pursue two goals; 1) to identify antigens targeted by T cells contributing to the development of type 1 diabetes (T1D); and 2) to pursue the potential of TCR sequences to be used as biomarker.   We hypothesize that common antigens targeted by T cells negatively (Treg) and positively (Tpath) contributing to the development of T1D in multiple patients are present.  Identification of such antigens and TCR repertoires will lead us to develop antigen-specific immunotherapy and immunodiagnostic tools.  We directly sequence TCR alpha and beta chains of T cells in the pancreas as well as in different T cell subsets in the immune organs and peripheral blood of patients having T1D utilizing the high-throughput Illumina and 454 sequencing technologies.  TCRs detected in the pancreas are expressed on TCR-null T cell hybridoma cells using the retroviral expression system and are tested for the response to known antigens as well as the islet cells.  Once TCR pairs responding to the islet cells are determined, we will further assess pathogenicity in inducing diabetes by generating retrogenic animals.  In addition, presence and frequency of pancreas-derived TCR sequences in the peripheral organs of the same patients as well as those having the same HLA alleles will be evaluated to “trace” T cells detected in the pancreas.  Given evidence that a large number of public TCR sequences detected from multiple individuals are present, we believe that TCRs associated with the development of T1D exist and are able to be used for diagnosis in combination with other T cell assays (e.g. tetramer analysis).  Common antigens associated with the pathogenesis and regulation of T1D development will gain our insights to develop antigen-specific immunotherapy for T1D.  

Thomas Pieber Medical University of Graz Linking IHC immune profiles of duodenum and pancreas in T1D

Our project “Linking IHC profiles of duodenum and pancreas in T1D” aims to intensify our understanding of how the “gut immune system” contributes to the development of type 1 diabetes.

Previous studies have shown that factors like the gut microbiome, gut permeability and the mucosal structure play a role in T1D pathogenesis. Studies are now more and more focusing on intestinal immune cells and their contribution to the onset and disease course of T1D. Regulatory T cells (Tregs) are of special interest in this regard, since they are capable of suppressing auto-reactive effector T cells, and thereby prevent autoimmune reactions. The role of intestinal Tregs in T1D development has been described controversially. While some papers report of increased numbers of Tregs in the intestine of T1D patients, other publications find no difference in Treg numbers between patients and healthy subjects. With our new immunohistochemical method, which allows us the identification of CD4+CD25+FoxP3+ putative regulatory T cells with a specificity that has not been reached before, we hope to shed light on regulatory T cell profiles of duodenum and pancreas in parallel in patients at different disease stages compared to healthy subjects. Additionally we aim to investigate CD8+ and CD4+ Th17 effector T cells in the same tissues in order to get comprehensive information on regulatory and effector T cell subsets. Furthermore, we will investigate T cell function by using single-cell analysis of positively stained T cells regarding their expression profiles and methylation status.

Thus, using the precious nPOD samples, we hope to identify immunological processes in the duodenum and pancreas that play a crucial role in the development and pathogenesis of type 1 diabetes.

Massimo Pietropaolo & Michael Morran Baylor College of Medicine Funded by: NIDDK Identification of T1D-specific Fab fragments of IA-2 dominant conformational epitope

One of the primary characteristics of Type 1 Diabetes (T1D) is the presence of autoantibodies directed to self-antigens expressed in the pancreatic islets, specifically those associated with the insulin secretory machinery. Although T1D is well characterized as a T-cell mediated disease, a mechanistic role of islet autoantibodies generated during the progression of T1D have not been elucidated. The tyrosine phosphatase-like protein (IA-2) is a major neuroendocrine autoantigen associated with the progression of T1D. We previously identified a new candidate biomarker within the extracellular domain of IA-2 (IA-2ec) in humans, which is associated to rapid acceleration of T1D onset compared to conventional IA-2 antibody biomarkers of T1D. Little is known about the clonality and the V region structures of IA-2 autoantibodies, an important consideration given their diagnostic and predictive value. Thus, our objective is to perform de novo sequencing to determine the clonality and V region structures of human autoantibodies directed against the islet autoantigen IA-2.  This method can be applied to multiple serum samples and has been validated by the discovery of a clonotypic autoantibody specific for an immunodominant determinant of target antigens of a number of autoimmune diseases including systemic lupus erythematosus (SLE) rheumatoid arthritis (RA). In particular, we aim to characterize the IA-2 autoantibody sequence in the sera of the well-characterized population of T1D patients of the nPOD network in an effort to determine if these autoantibodies express a highly restricted B-cell clonotype repertoire which may be shared (public) across different disease phenotypes. We will then identify unique V region peptides to interrogate human sera for IA-2 by a targeted mass spectrometry (MS) protocol. Mass spectrometry-based detection of specific autoantibody motifs may provide a powerful new tool for analysis of humoral autoimmunity in T1D.

Alberto Pugliese & Francesco Vendrame Diabetes Research Institute, University of Miami Funded by: Assessment of memory T cells in the insulitis lesion Assessment of memory T cells in the insulitis lesion

There is growing evidence that memory T cells are associated with a variety of autoimmune conditions, including type 1 diabetes (T1D). While autoreactive T cells are detected in both patients and healthy control subjects, CD45RO+ memory autoreactive T cells are usually found in patients but not in controls. Our own studies in several pancreas transplant recipients with T1D recurrence have shown that autoreactive memory T cells can be found in the circulation and pancreas transplant lymph nodes. Moreover, the examination of pancreas transplant biopsies from two of such patients shows that most (~80%) of the islet infiltrating T cells have a memory phenotype. Based on this findings we hypothesize that memory cells should be present in the islet infiltrates of nPOD patients. Specifically, we plan to study pancreatic tissue obtained from patients with proven insulits for the presence of the T cells bearing the memory marker CD45RO. It is plausible that the frequency of the memory cells may be related to disease duration, and perhaps this could be higher in the lesions of patients with longer disease duration who still have insulitis. It is also possible that there will be a difference compared to pancreas transplant recipients with recurrent disease. Understanding whether memory T cells represent a significant component of islet infiltrating T cells is relevant to a more complete characterization of the insulitis. A recent clinical trial in new onset patients has tested Alefacept, an anti-memory T cell drug. Demonstrating memory T cells in the lesion would provide further rationale for regimens that target memory T cells, with Alefacept or with new drugs that have similar effects. Furthermore, if memory cells are demonstrated, we can propose additional work to better define the memory phenotype by staining for additional markers.

Helena Reijonen Benaroya Research Institute at Virginia Mason Correlation of Islet Autoantigen-Specific T-Cell Repertoire in the Pancreatic Lymph Nodes and Peripheral Blood in T1D Autoimmunity

Islet autoantigen specific T cells can be found in both T1D patients and healthy individuals.  However, these T cells with the same antigen specificity are expected to differ in their phenotypic characteristics. We will investigate CD4 and CD8 T cells in nPOD samples from T1D patients, prediabetic and healthy subjects allowing us to compare T cell specificities and phenotypes at different stages of the disease. We have generated tetramers representing the most common HLA-DR types associated with T1D risk (DR401, 404 and 405, 301, DQ8) or protection (DR1501, DRB5 0101 and DQ 0602) containing immunodominant peptides from the major diabetes autoantigens GAD65, Proinsulin/insulin, IA2 and Zn-T8.  The approach is based on the utilization of tetramers to identify and isolate autoreactive T cells from pancreatic lymph node tissue, peripheral blood (and possibly bone marrow) offers an advantage that phenotypic analyses are performed on T cells responding to a single, known antigen mediated with a defined MHC restriction.  Tetramer-based technology also allows determination of the functional phenotype of autoantigen specific T cells, for example memory, avidity/affinity, cytokine profiles, homing properties, and these cells can be sorted for gene expression studies and TcR analysis. In the cases when both peripheral blood lymphocyte and pancreatic lymph node samples are available the proposed study would provide valuable information on the predictive value of blood samples as biomarker in the evaluation of the disease process.  Collaboration with other nPOD investigators allows our project to generate new knowledge on the key features and the population dynamics of CD4+ T cells directly associated with islet autoimmunity and beta-cell biology in human diabetes. 

Suparna Sarkar & Dirk Homann Mount Sinai Hospital Funded by: JDRF and NIDDK Pancreatic expression of chemokines in human type 1 diabetes

Human chemokine superfamily of molecules mediate the recruitment of leukocytes to sites of inflammation. Many chemokines and chemokine receptors have emerged as key contributors and regulators of autoimmune disorders, including Type 1 Diabetes and may serve as potential surrogate biomarkers of inflammation or possible drug targets. More than one-half of all human and rodent chemokines have been implicated in the pathogenesis or complications of Type 1 diabetes. Most of the previously published work however is limited to the genetic association studies and chemokine and their receptor analyses in the peripheral blood. To delineate the complete spectrum of chemokine mRNA species transcribed by human islets in response to inflammatory stimuli, we used an established in vitro culture system in which islets purified from healthy organ donors were cultured in the absence versus the presence of IL-1β, TNFα, and IFNγ and subsequently subjected them to gene expression microarray analysis. Based on our microarray data and in collaboration with Homann laboratory, we study the expression and localization of chemokines in pancreatic tissue sections obtained through the Network for Pancreatic Organ Donors with Diabetes (nPOD) and included healthy control subjects.

Nora Sarvetnick & Kamiya Mehla University of Nebraska Medical Center Funded by: NIH Analysis and comparison of IL-18, IL-18R, CCL19 and CCL21 expression in T1D, T2D, autoantibody positive and control samples

Type 1 diabetes (T1D) has devastating effects on the health and quality of life of those affected. Several studies have associated the proinflammatory cytokine IL‐18 with an increased risk of T1D. Significantly higher IL-18 cytokine levels are seen in individuals at high risk for developing disease, increased serum levels of IL‐18 have been observed in T1D, and the presence of multiple islet autoantibodies in new‐onset diabetics was found to be associated with increased serum levels of IL‐18. Furthermore, our recent investigations demonstrate a role for IL‐18 in expanding the pool of islet‐destructive T cells during pre‐diabetes.  Taking the elevated serum levels of IL-18 in T1D and enhanced survival properties of IL-18 receptor (IL-18R)-expressing CD8 T cells into account, there might be a connecting link between IL-18R+ CD8 T cells and T1D pathogenesis. Hence, in the present study we propose that IL-18R+ CD8 T cells may underlie the direct or indirect islet destruction associated with T1D. Furthermore, IL-18 expression has been linked to the C-C chemokine receptor type 7 (CCR7).  Specifically, IL-18 has been shown to induce expression of CCR7, and IL-18-activated cells can stimulate the production of the CCR7 ligand CCL19.  Based on this data and our preliminary data showing an increase in proportion of a transitional CD4 T cell population expressing CCR7 in new-onset T1D patients, we postulate that the CCR7 ligands CCL19 and CCL21 will be overexpressed in the pancreas and spleen of T1D patients when compared to control, autoantibody positive, and T2D patients. Overall, the goal of our studies is to fully understand the role of IL-18R+ CD8 T cells and associated cytokines and soluble factors in T1D pathogenesis. With the aid of nPOD, human spleen and pancreatic tissue specimens will be utilized for immunohistochemical studies to determine islet infiltration of IL-18R+ CD8 T cells, as well as differential expression of cytokines and soluble factors important for the development of T1D.

David Serreze & Kevin Mills The Jackson Laboratory Targeting AID expressing B-lymphocytes as possible clinically applicable T1D intervention

Diabetogenic B-lymphocyte activity may be enhanced by SHM induced increases in Ig antigen
-SHM requires AID induced double strand DNA breaks repaired by RAD51 complex proteins.
-AID is expressed at higher levels in B-lymphocytes from NOD mice than control strains
-A small RAD51 blocking molecule specifically induces apoptosis of AID expressing B-lymphocytes
from both NOD mice and humans.
-RAD51 blockade is being tested as a new B-lymphocyte directed late disease stage T1D intervention
approach in NOD mice.
-For possible future clinical translation purposes it will be important to assess if diabetogenic Blymphocytes
in humans are undergoingAID induced SHM, and thus may be susceptible to depletion
by RAD51 blockade.
-To address this need nPOD is being requested to supply five cryopreserved PLN samples each from
T1D patients and controls.
-We will use these nPOD samples to determine if B-lymphocytes within PLNs of T1D patients express
higher AID levels than those from controls and thus more susceptible to RAD51 blockade induced
apoptotic death.
-Previous published work by Jax investigators demonstrate they have an ability to assess AID
expression by human B-lymphocytes, and determine their sensitivity to RAD51 blockade induced

Luc Teyton The Scripps Research Institute Funded by: NIH/NIAID, NIH/NIDDK Analysis of Immunocytes in Human Islets of Type 1 Diabetes Patients

CD4+, CD8+ T cells and B cells will be isolated from islets of patients suffering from type 1 diabetes. Laser dissection will be used to localize and isolate single cells after tissue staining and confocal imaging at high resolution. T cell receptors and B cell receptors will be isolated from single cells and expressed to characterize antigen specificity. T cells will also be characterized for their expression profiles of cytokines, chemokines, co-receptors, and transcription factors. These studies are aimed at discovering important new antigenic targets in type 1 diabetes and understanding the pathogenic process leading to T1D.

Hubert Tse University of Alabama-Birmingham Funded by: NIH/NIDDK Redox regulation of anti-viral responses in type 1 diabetes

Type 1 diabetes (T1D) is a T cell-mediated autoimmune disease resulting in the destruction of pancreatic beta-cells. Infiltrating leukocytes generate an inflammatory environment consisting of reactive oxygen species (ROS) and proinflammatory cytokines that collectively participate in beta-cell destruction and enhance the adaptive immune effector response of islet-specific T cells. Environmental factors that include viral infections are hypothesized to directly lyse pancreatic beta-cells and generate a pro-inflammatory milieu necessary to enhance T cell autoreactivity. Currently, a gap exists in our understanding of why the immune system attacks itself in Type 1 diabetics and the signals involved in initiating immune system activation in T1D. Our previous work has demonstrated the importance of oxidative stress and the generation of ROS as essential to the pathogenesis of T1D. ROS production is required for autoimmune responses during the development of T1D in Non-Obese Diabetic (NOD) mice, a mouse model that is widely used to study T1D. Mice deficient in superoxide synthesis (NOD.Ncf1m1J) are resistant to the development of T1D and our proposed studies to further define the synergy of ROS on immune responses have the potential to provide novel insight on the mechanisms associated with pancreatic â-cell destruction in T1D. Our proposed studies will test the hypothesis that ROS synthesis is necessary for efficient anti-viral responses in T1D. Specifically, we will define the role of redox-dependent signals involved in the activation of human innate immune receptors such as Toll-like receptor 3 (TLR3), melanoma differentiation-associated gene 5 (MDA5), and retinoic acid-inducible gene I (RIG-I) in response to Coxsackie virus infection with PBMCs from recent onset T1D and healthy patients. The studies in this proposal will further expand our knowledge of how innate immune-derived pro-inflammatory signals synergize to mediate efficient autoreactive T cell responses in Type 1 diabetes.

Matthias von Herrath and Teresa Rodriguez University of California, San Diego Funded by: NIH In situ detection of pro-inflammatory cytokines within pancreatic islets from diabetic subjects

In Type 1 Diabetes (T1D), the complex interplay of immune cells shifts the balance in favor of the development of lymphocytes, that attack structures of our own body: autoimmunity. These attacks resulting in an inflammation in the pancreas and, eventually, to the destruction of insulin producing beta cells.

Cytokines are small proteins that are released by cells and affect the behavior of other cells. They are oftentimes involved in promoting inflammation. We believe, that a network of “cytokines” determines the direction of the immune response towards the destruction of beta cells, and, as recent findings suggest, that this inflammatory milieu precedes the onset of the disease. However, it remains elusive so far, which cytokines are produced and when, and how they interact in such a complex network to sustain inflammation.

In this project, we use highly sophisticated chemical and bioinformational tools to detect important cytokines that are known to promote inflammation: Interleukin 1beta, Interleukin 6, and Tumor Necrosis Factor alpha in tissue pieces from diabetic patients, who have donated their pancreas for research.

This allows us to analyze both the presence of cytokines as well as their interactions during the very early days of diabetes at the single cell level and to create a so called “cytokine map” of the disease. Knowing which cytokine is released by which cell and when is a first step towards the development of very specific therapies for the treatment of diabetes and of other autoimmune diseases.

Shannon Wallet University of Florida Funded by: NIH/NIDDK Role of mucosal epithelium in autoreactivity

One of the major questions we still don’t have an answer to is why some people who are genetically predisposed for development of T1D don’t progress and why others do. In addition, why there are different timings for the progression of this disease (i.e. why are some faster than others).  Our thoughts are that these differences are due to differences in environmental experiences.  It is important to note that we feel that these different environmental experiences can be big or small (i.e. living in different areas, eating different foods, or infection with different organisms). Thus, my laboratory studies the communication of the immune system with the environment at the largest environmental interface within our body, the gastrointestinal track.  We feel that if we can understand the difference in this communication that occurs in T1D and how it shapes the immune system, we can not only figure out the initiating events of this disease process, but design better therapies and even preventive treatments to curb the disease process.

Emmanuel Zorn, Yufeng Shen, Ivan Gerling Columbia University Funded by: Helmsley Charitable Trust George S. Eisenbarth nPOD Award for Team Science Immunoregulation of type 1 diabetes

Type 1 Diabetes (T1D) is a disease that results from the autoimmune destruction of insulin-producing beta cells in the pancreas. The current treatment for T1D does not cure the disease or prevent the possibility of other serious complications. The mechanism that leads to the loss of immune tolerance still remains unclear; scientists believe that both genetic and environmental triggers play a role, making it challenging to develop treatments that address both. Investigators at the Columbia University Center for Translational Immunology (CCTI) conduct many research studies aimed to determine factors involved in T1D. The purpose of our study is to investigate genetic variants limited to immune dysregulation in T1D through next generation sequencing of genetic material from tissues of T1D donors. Elucidating the genetic factors involved in T1D will contribute to our knowledge of this disease and autoimmunity.

Novel Biomarkers

Ken Coppieters, Ditte Tornehave & Charles Pyke Novo Nordisk A/S IL-21/IL-21R and GLP-1R staining on pancreas tissues from (pre-)T1D, T2D and control donors

Our aim is to assess the protein expression level in conjunction with histological and (sub)cellular localization of the cytokine IL-21, its receptor IL-21R and GLP-1R in normal controls, (pre-)type 1 and type 2 diabetics (with history of incretin treatment for GLP-1R). (adopted by nPOD staff)

Nika Danial & Ulupi Jhala Harvard Medical School Funded by: NIH-HIRN Molecular signatures of islet inflammation in type 1 diabetes

The process of β-cell decline in the lead-up to the overt diabetes could provide critical clues for timely therapeutic intervention. However the complexity, cell-type specificity, rapidly changing dynamics and the cumulative effects of inflammation all inform the outcome of the autoimmune attack on beta cell.

This proposal is designed to use multiple in vitro models systems of beta cell inflammation, along with human tissue samples as well as human serum samples from T1D children, to arrive at a temporal proteomic and metabolomics profile of human islet inflammation and death. These data are likely to contribute towards our understanding of biomarkers that would assist in staging the progression of human type 1 diabetes. The ultimately goal is to be able to use the information gleaned from the grant to identify windows of opportunity when therapeutic intervention could preserve beta cell mass and function to delay the progression towards full-blown diabetes.

Zhiguang Guo, Alex Rabinovitch & Romano Regazzi Sanford Research/Sanford Health Funded by: NIH Identifying and Validating Noncoding RNAs as Human Beta Cell-Specific Biomarkers

Direct identification and validation of human β cell-specific biomakers is critical for monitoring the change of human β cell mass at prediabetes and the new-onset diabetes stage and during the therapeutic intervention. Due to technical and ethical constraints, our ability to prospective studies of human β cell biomarkers in response to β cell stress/apoptosis and its correlation with early β cell loss in human is limited. The objective of our study is to develop biomarkers of human β cell stress/apoptosis correlating with β cell loss that can be used to non-invasively monitor the development of diabetes pathogenesis and the outcome of treatments aimed at preserving or restoring functional β cell mass. The goal of this study is to identify potential biomarkers including a panel of noncoding RNAs in human islets that are correlated with β cell stress and apoptosis and to determine whether novel β cell biomarkers in circulation are correlated with early human β cell loss. We hypothesize that β cell ER stress and apoptosis induction increases a panel of noncoding RNAs in human islets in the pancreas of patients with diabetes and dead or dying human β cells can release them into circulation and the levels of these noncoding RNA in circulation are associated with β cell loss. We plan to collect islet cells for miRNA expression profiles by laser capture microdissection (LCM) and determine whether a panel of β-cell specific noncoding RNAs in islets of human pancreas specimens from patients with type 1 and type 2 diabetes are up-regulated by real-time PCR array. We will confirm some up-regulated noncoding RNAs in human islets are β-cell specific by miRNA fluorescence based in situ hybridization (FISH). We will validate this panel of noncoding RNAs, as β cell-specific biomarkers, in plasma of patients with type 1 and type 2 diabetes. Success in our proposed experiments would help us to further develop a panel of β cell biomarkers, which are correlated with early β cell loss, for clinical evaluation of disease stages including pre-diabetes and of therapeutic interventions to promote β cell health and survival.

Mark Huising University of California, Davis Funded by: JDRF Urocortin 3 as a sensitive marker for beta cell function

Urocortin 3 (Ucn3) is a neuropeptide that was discovered by our group (Lewis et al., 2001). Ucn3 is specifically expressed in mouse beta cells, but is in human islets also expressed by alpha cells (Li et al., 2003; van der Meulen et al., 2012). Ucn3 is a member of the same neuropeptide family as corticotropin-releasing factor (CRF) and both peptides activate related receptors; CRF predominantly activates the type 1 CRF receptor, while Ucn3 exclusively activates the type 2 CRF receptor. We have in recent years discovered that CRF and Ucn 3 both contribute to the regulation of insulin secretion and the maintenance of functional beta cell mass by virtue of locally expressed CRF receptors in mouse and human islets (Huising et al., 2010, 2011). Interestingly, Ucn3 is not detectable in the developing mouse pancreas until embryonic day 17.5 and is not expressed by all beta cells until postnatal day 14 (van der Meulen et al., 2012). This is considerably later than the onset of expression of transcription factors that mark mature beta cells, such as Pdx1, Nkx6.1 and MafA as well as markers of functional beta cell maturity, including Pcks1, Znt8 (Scl30a8) or Glut2 (Scl2a2; in mouse).

In line with findings in mouse pancreas, Ucn3 expression in human embryonic pancreas is not detectable until the end of the first trimester and trails the onset of expression of insulin and glucagon by several weeks (van der Meulen et al., 2013). However, in marked contrast to murine pancreas, where Ucn3 demonstrates exquisite selectivity for the beta cells, Ucn3 in both human and macaque (Macaca nemestrina) pancreas and in freshly isolated human donor islets is expressed in both beta and alpha cells (van der Meulen et al., 2012). Moreover, and perhaps to emphasize the species differences that exist regarding the local network of CRF-family neuropeptides, CRF expression is selective for human alpha cells and is not detected in mouse islets. In agreement to the above observations, Ucn3 in human embryonic stem cell-derived pancreatic endocrine cells is expressed relatively late and only after an extensive period of in vivo maturation in recipient mice and is then detectable in both mature beta and alpha cells (van der Meulen et al., 2012). These findings have established Ucn3 as a marker of beta cell maturity. We therefore hypothesize that Ucn3 is a biomarker of functional, mature beta cells, and is down regulated during diabetic conditions that are associated with beta cell dysfunction and dedifferentiation. Supported by nPOD, we will establish if Ucn3 expression is reduced in alpha or beta cells of T1D and T2D donors compared to non-diabetic controls.

Richard Kitsis Albert Einstein College of Medicine ARC, a novel beta cell survival factor in type 2 diabetes

The progression of type 2 diabetes is mediated to a large extent by the deaths of β-cells, the cells in the pancreas that synthesize and secrete insulin.  We discovered that ARC (Apoptosis Repressor with CARD), an endogenous inhibitor of cell death, is expressed at high levels in β-cells of the mouse pancreas.  Using cell culture studies and genetic mouse models, we have further demonstrated that ARC functions as a critical β-cell survival factor during type 2 diabetes. However, little is known about the expression patterns of ARC in human β-cells and other cells within the endocrine pancreas.  This information is critical for understanding the role of ARC in the pathogenesis of diabetes in humans.  We have assessed whether ARC levels in islets decrease during diabetes, and found no significant changes in ARC abundance in old versus young ob/ob mice.  Unexpectedly, however, we observed that the subcellular localization of ARC differs in diabetic versus non-diabetic mice. In this study, we propose to characterize ARC expression in islet cells of human pancreatic tissue from non-diabetic individuals and those with type 2 diabetes.  

Richard Lloyd, Heikki Hyoty, Joseph Petrosino & Ivan Gerling Baylor College of Medicine Funded by: JDRF and NIH-HIRN Beta-cell stress signatures and infection in T1D islets

This project will be the first to examine stress granule responses and the expression of long non coding RNAs in donor islet cells, a model beta cell line and the role of enterovirus infection in modulating these stress responses. This information will be used as a benchmark to examine nPOD tissues for these same stress responses, mobilization of innate immunity markers. The above experiments will also be performed in conjunction with analysis of the replication of terminally-deleted defective enteroviruses in the cell lines, and the stress responses they produce. This will inform secondary studies of actual nPOD tissues for the level of stress responses, key lncRNA expression profiles present. The nPOD tissues will also receive the most stringent analysis for virus capsid proteins and particularly viral RNA with single molecule sensitivity on a single cell basis that has yet been attempted.

Anna Moore & Amol Kavishwar Massachusetts General Hospital Funded by: NIH In vivo imaging of islets

The aim of our study is to develop an imaging agent for in vivo determination of beta-cell mass in humans. Such an imaging agent will not only provide information on progression and management of diabetes but it will also help in monitoring an individual’s response to exercise, diet and anti-diabetic therapy. Unfortunately, there aren’t any proteins/epitopes that are known to be uniquely present on insulin producing beta cells that can be targeted for imaging. Unfortunately, imaging agents that are currently in use or are under development lack specificity. These imaging agents highlight other cell types as well on PET and MRI images making it difficult to estimate beta-cell mass.

In past we have shown the utility of an IgM antibody (IC2) for ex vivo determination of beta-cell mass. This antibody was around for several decades but lacked full characterization. We recently showed that this antibody binds to unique sphingomyelin-rich patches present on the surface of insulin producing beta-cells. We did not see such patches on any other cells or tissue that we tested. Further characterization using human diabetic samples from nPOD revealed that these unique sphingomyelin patches represent insulin-secretory status of beta-cells. These patches were present on beta-cells from pancreatic tissue from normal individuals. There were fewer patches on beta-cells from pancreatic tissues from diabetic individuals and they were completely absent from beta-cells from type-1 diabetic patients.

IgM is a very big molecule and owing to its propensity to activate complement and slow blood clearance, it is not an ideal molecule for use as an imaging agent. Currently, we are developing a shorter version of this affinity agent such that it can carry an imaging payload to beta-cells for in vivo imaging and estimation of beta-cell mass. In the future we plan to combine this affinity imaging agent with therapy delivery.

Interesting papers:
Diabetes. 2001 Oct;50(10):2231-6.
J Lipid Res. 2011 Sep;52(9):1660-71.
J Histochem Cytochem. 2013 Dec;61(12):910-9.
Jerry Nadler, Margaret Morris & Julius Nyalwidhe Eastern Virginia Medical School Protein based biomarkers for type 1 diabetes

We have previously used Imaging Mass spectrometry (IMS) to identify Insulin within β-cells in the islets of Langerhans. The analysis utilized specially prepared pancreatic sections from the Network of Pancreatic Organ Donors with Diabetes (nPOD). In these studies, a comparative analysis of T1D MS spectra to those of Non-T1D revealed several uniquely expressed proteins between the two groups.  In the current study, we propose to perform inter- and intra-sample qualitative and quantitative analyses to verify protein expression between normal and T1D tissue samples in an expanded subset of samples.  We have developed methods that can distinguish proteins that are differentially present in T1D and non-diabetic samples. To this end, we have recently acquired a new instrument with a higher mass accuracy, resolution and sensitivity (Q-Exactive mass spectrometer, Thermo Fisher Scientific) on which we will perform LC-MS-MS analysis of additional samples in order to verify and validate the previously defined panel.  Therefore, we propose to identify and develop quantitative assays to determine the utility of these molecules as potential diagnostic biomarkers for T1D by continuing our studies in additional pancreas samples. 

Bruce Seligmann BioSypder Technologies, Inc In Site TO-Seq Profiling of Gene Expression and Proteins in T2D

We will select T2D patients, among them subjects with DNA variants, and profile gene expression, expressed mutants, and proteins from sub-areas down to the level of single islets and single cells within archived pancreas FFPE tissues using a novel in situ TO-Seq assay coupled with immunohistochemistry staining of protein biomarkers, retaining the morphologic context of the profiling data. Among the set of genes measured will be biomarkers of every known molecular pathway in addition to genes associated with the development, presentation, dysregulation, and pathogenesis of T2D. By obtaining in situ gene expression profiles defining for all the cell types involved in diseased vs normal tissue within both their morphologic context and the context of genetic variants we will be able to identify molecular phenotypes and mechanisms and correlate these to cell-cell interaction, tissue homeostasis, and genetic variants. Because TO-Seq is a targeted sequencing-based assay that can measure thousands of genes, and multiplex many samples within a single sequencing run it will enable profiling data to be obtained from many separate cells at an affordable cost/cell in order to achieve the statistical power required to produce quality data supporting reliable interpretation and bioinformatics analysis. No such data exists today, and there is no system capable of producing such data on a large scale at an affordable price. Laser capture microdissection (LCM) can produce single cell profiles within their morphological context, but at a high cost which severely limits the number of samples that can be profiled. We provide preliminary data demonstrating excellent correlation and sensitivity of TO-Seq measurement of gene expression from matched fixed and frozen tissue and successful implementation of the in situ TO-Seq protocol and its integration with immunohistochemical (IHC) staining of antibody biomarkers. We demonstrate the feasibility of performing single (prostate) gland profiling with this assay. Based on these preliminary data we will optimize the in situ TO-Seq/IHC assay, implement a single islet and cell sampling system benchmarked against LCM, implement the content for the assay, and then profile the cell types present in pancreas of normal vs T2D, and between areas of pancreas exhibiting a normal vs T2D histopathology. From these data we expect to gain an understanding of the impact of genetic variant(s) at the level of gene (and protein) expression, cell interactions, and upstream regulation, and molecular mechanisms and phenotypes as they potentially contribute to disease and its progression.

Bruce Seligmann, Klaus Pechhold & Horacio Rilo BioSpyder Technologies, Inc. In situ gene profiling and identification of T1D disease mechanisms

This CBDS program will establish an in situ RASL-Seq Islet Study Facility that will validate a novel in situ RASL-Seq platform for diabetes research using FFPE tissue from normal and Type 1 diabetes (T1D) diseased pancreas, and then apply this platform to address the gap in available methodology to take full advantage of T1D organ repositories such as nPOD in the effort to derive molecular profiles of individual -cells and cells of other types associated within the islet that are involved in the onset of silent T1D and disease progression. We will measure gene expression profiles from individual cells, and from this data identify biomarkers and gene expression phenotypes of silent T1D compared to more advanced stages of disease. We will exploit “big data” to apply pathway analysis to the gene expression results in order to identify putative molecular mechanisms of silent T1D and disease progression, providing a basis for the development of novel therapeutic approaches. We will test matched serum to identify useful biomarkers of silent T1D from this accessible tissue, and potentially to provide the capability to divide patients into subgroups based on prognostic outcome of whether disease will progress or not.

Terry Strom, Thomas Thornley & Maria Koulmanda Harvard Medical School To identify mRNA and miRNA biomarkers associated with exosomes that predict the onset of T1D

There are currently no clinically precise biomarkers that predict the development of type 1 diabetes (T1D). This limits the ability to identify patients who will become diabetic at a time when they have a greater beta cell mass and thus may respond more favorably to interventional therapeutics aimed at stopping T1D progression. To this end, we propose to identify biomarkers that predict T1D progression by examining exosomes in at-risk and nondiabetic control individuals. Exosomes are small particles released by cells during inflammation that communicate “danger” to other cells nearby and at a distance. Exosomes contain numerous mRNA and miRNA messengers that alter the biology of target cells, which include cells of the innate and adaptive immune systems. We propose to characterize the transcriptional (mRNA and miRNA) contents of exosomes from at-risk and non-diabetic control tissues in order to identify potential biomarkers that predict progression of T1D.

David Taylor-Fishwick Eastern Virginia Medical School The patterns of NADPH oxidase-1 expression in the course of human diabetes

Increased levels of reactive oxygen species (ROS) lead to dysfunction and subsequent loss of insulin-producing beta cells in pancreatic islets. Several lines of evidence support the importance of increased NADPH oxidase (NOX) activity as one of the contributing factors to elevated levels of ROS in beta cells. NOX-1 activity has been established as one of the sources of ROS in beta cells under experimental diabetic conditions. A contribution of NOX-1 to beta cell dysfunction in human diabetes is supported by preliminary studies. To further validate these observations as a first step we seek to establish the pattern of expression NOX-1 in normal and diabetic human islets. We hypothesize that NOX-1 will be elevated in diabetic islets. NOX-1 expression will be studied in relation to both the specific cells within pancreatic islets (beta cells, alpha cells) identified using well established specific protein markers for each phenotype as well as to the endocrine “sufficiency” status of such cells. In order to specifically examine qualitatively and quantitatively islets from 6 non-diabetic controls, 6 type 2 diabetes, 6 type 1 diabetes of short and 6 of long duration as well as 6 autoantibody positive would be employed; hence a total of 30 donors will be included in this study. Paraffin embedded 7 ìm thick sections of pancreas will be immunostained, using a protocol previously developed specifically for nPOD specimens. Commercially available antibodies raised against human NOX-1, insulin and glucagon will be used. Stained sections will be subjected to microscopic evaluation using both conventional fluorescent microscope and confocal microscope. A qualitative and quantitative analysis of the obtained images will be employed. We expect that an increase in expression NOX-1 will be observed in islets obtained diabetic and autoantibody positive donors.

Charles Thivolet University Claude Bernard Lyon 1 Characterization of new cell markers associated with beta cell failure in diabetes

The percentage of reduction of beta cell mass in patients with Type 1 diabetes (T1D) of recent onset cannot fully explain the reduction of endogenous insulin production that leads to hyperglycemia. In addition, functional defects of the beta cells may prevail from reduction of beta cell mass during Type 2 diabetes. In boty situation chronic inflammation may explain why beta cells cannot produce sufficient amounts of insulin to cope with insulin demands. The purpose of the project is to examine in situ both the mitochondria and RE-mitochondria interactions, key organelles of beta cells. Apart from being involved in the capacities of beta cells to secrete insulin upon demand, mitochondria have a central role in beta cell fate through ROS generation, capacities to transfer apoptotic signals and in the orchestration of inflammatory responses and inflammasome activation. The levels of ER-mitochondria overlap and number of ER-mitochondrial associated membranes (MAM) will be quantified using an in situ proximity ligation assay. We plan to compare the degree of RE-mitochondria interactions in inflamed vs intact islets, in beta vs non beta cells, in patients with T1D vs T2D. Expression of specific mitochondrial proteins involved in Ca2+ uptake and in the balance of mitochondria fission and fusion processes will be analyzed. Since excessive production of ROS and release of oxidized mitochondrial DNA following mitochondrial stress can be sensed by the inflammasome, levels of NLRP3 expression will be assessed both in mononuclear cells contained in inflamed islets and eventually in the beta cells. This could be an important link between innate immunity in response to viral infections and autoimmunity to beta cells. Since miRNAs can also regulate mitochondria activity, redox state and inflammatory pathways, semi-quantitative analysis of specific miRNAs will be performed by fluorescent micro-RNA in situ hybridization (FISH) technique with digoxigenin-labeled probes. Altogether, this project will give a comprehensive analysis on the role of mitochondria in the beta cell defects during the early phases of diabetes.

Qibin Zhang & Mark Atkinson The University of North Carolina Greensboro Human pancreatic tissue proteome of type 1 diabetes

The goal of this project is to comprehensively profile the proteins expressed in human pancreas within both the islets and their surrounding environments, and to obtain relative expression levels of pancreatic proteins from different organ donor groups using mass spectrometry based bottom-up and top-down proteomics technologies. Snap-frozen pancreatic tissue sections from organ donor groups with  1) autoantibody negative, no diabetes; 2) autoantibody positive, no diabetes; 3) autoantibody positive and T1D; and 4) autoantibody negative and T2D will be analyzed on both the peptide and intact protein levels for comprehensive identification and quantification of proteins and protein isoforms that differentially expressed among these subject groups. To complement with our study on pancreatic tissue sections, islets and beta cells will be isolated from pancreatic tissues using laser capture microdissection for their comprehensive proteomic characterization and comparison as well. The significantly changed proteins and the specific pathways and networks they are involved in may serve as candidate biomarkers for diagnosis and aid in elucidation of the pathogenesis of T1D.


Peter Butler, Alexandra Butler & Mark Atkinson University of California, Los Angeles Pancreatic duct glands in type 1 diabetes

The initial overall question to be tested by the collaborative of the Peter Butler laboratory at UCLA and the Mark Atkinson laboratory in Florida was to explore the possibility that the recently described pancreatic duct gland (PDG) compartment of the pancreas might serve as the elusive pancreatic stem cell niche, potentially offering an avenue to promote beta cell regeneration. Both laboratories had previously noted the presence of pancreatic beta cells in long standing type 1 diabetes mellitus (T1DM) and had been interested in the possibility that these were derived from beta cell regeneration rather than simply reflecting beta cell survivors.

The initial studies focused on quantification of the PDG compartment in T1DM with comparison to type 2 diabetes mellitus (T2DM) and non-diabetic controls. Since the primary interest in the PDGs was as a potential source of endocrine cells, we also examined the nPOD immunostained sections for insulin and glucagon to evaluate if the PDGs did indeed contain insulin or glucagon staining cells. During this process it was noted that there some of the nPOD sections had markedly expanded glucagon staining compared to normal controls. Further examination of the possible explanation of this led to the observation that the increased glucagon staining was most prominent in the group of individuals with T2DM treated with GLP-1 based therapy prior to their demise.

Following communication between the collaborating laboratories in Florida and Los Angeles, it became apparent that we had a responsibility to further delineate the unexpected differences between the pancreases of individuals with T2DM treated by GLP-1 based therapy and other therapies. The need for this was further heightened by the finding by the University of Florida group of neuroendocrine tumors expressing glucagon in GLP-1 treated T2DM patients. Further studies by an experienced pancreatic pathologist at UCLA raised concern that there was increased pancreatic dysplasia in the pancreases, and data from nPOD revealed increased overall pancreas replication (by Ki67) and expanded pancreas size. A question arose as to how best to make these observations known. It was decided that it would be preferable to include all the findings that seemed relevant to these observations in a peer-reviewed paper as well as to inform the domestic regulatory agency in charge of drug safety (i.e., the Food and Drug Administration – FDA). After months of rigorous peer review, the work saw publication in May of 2013 in the journal Diabetes.  In addition, following notification, the FDA posted a Drug Safety Communication in March 2013 noting that a group studying pancreatic pathology had suggested an increased risk of pancreatitis and pre-cancerous cellular changes in patients with type 2 diabetes treated with incretin mimetics.; however, the FDA could not reach any new conclusions about safety risks of incretin mimetic drugs. A vigorous response was seen from interested pharmaceutical interests including broad based subpoenas to multiple individuals at UCLA.  The current work in relation to the nPOD studies is now focused largely on repeating various aspects of the study (e.g., glucagon immunostaining originally preformed by nPOD at the University of Florida now being repeated at UCLA) to establish to what extent the criticism of the studies is valid (in which case the authors will acknowledge this) or if not, then respond with additional findings.  Since the initial FDA statement, they have subsequently published a follow up finding together with EMA (European Medicines Agency) in February 2014 in the New England Journal of Medicine. These agencies concluded that, based on current data, a causal association between incretin-based drugs and pancreatitis or pancreatic cancer could not be drawn.

In this regard the University of Florida group is now preparing a publication for peer review supporting the notion that the pancreas mass is indeed increased in individuals treated with GLP-1 based therapy. Moreover, the expansion is most apparent in the head of the pancreas, the region most vulnerable to pancreatic dysplasia.

The ongoing UCLA based studies to re-evaluate the alpha cell hyperplasia in GLP-1 treated individuals previously reported from the nPOD immune staining, has agreed with the conclusion of the prior analysis, that there is alpha cell hyperplasia in the incretin treated cases.

Once the issues of criticism in regards the prior study have been addressed the plan is to revert to the originally intended focus, specifically on the PDG compartment as a potential source of endocrine cell formation in humans.

Ben Giepmans University Medical Center Groningen Funded by: JDRF and STW Nanotomy of human islets of Langerhans in type 1 diabetes

Microscopy is of paramount importance in diabetes research: Paul Langerhans was the first to describe (1869) Islets based on a light microscopic (LM) evaluation of the pancreas. (Immuno-)EM-analysis of the Islets of Langerhans is an outstanding way to discriminate beta-cells from other endocrine cells: The crystalline appearance of insulin granules makes beta-cells readily distinguishable from other cells, allowing to assess total insulin content. In addition, analysis of environmental factors hypothesized to caused diabetes, e.g. viral infection, can be directly visualised. Ultrastructural analysis by EM reveals beta-cell condition, ER stress, abnormalities in mitochondria, fibrosis, as well as apoptosis, necrosis, macro- and micro-autophagy. EM can be used to follow the etiology of T1DM, the fate of transplanted Islets in both human as well as animal models and to assess the success of cellular trans differentiation to create insulin-producing cells. Moreover, EM allows other cell-types to be recognized in the Islet that affect the functioning of beta-cells, including immune cells attacking beta-cells, duct cells, endothelial cells and nerve cells. Our proposed approach enables identification of morphological and macromolecular characteristics in different cell-types within the Islets in a quantitative and high-throughput manner. A single dataset could resolve most of the features raised above. So why is EM-analysis not standard included? Traditionally, EM data-interpretation is mainly done by the publishing scientist(s). EM-examination is hampered by the laborious procedure that requires technical expertise and expensive equipment. In addition, data-sharing is limited, the field of view of the area inspected at high magnification is traditionally very small, the data is noisy making it difficult to objectively quantify and segment the greyscale EM-images. Without the knowledge of how Islets of Langerhans look like, it will remain cumbersome to establish the cause and solution to diabetes. Implementation of computational science has now revolutionized EM, paving the path for routine EM-analysis of complete Islets, as we have pioneered.

We are creating and analyzing high-resolution EM-maps of cross-sections of Islets of Langerhans in health and diabetes, which we term Nanotomy of Islets. The datasets allow to determine many aspects, including leukocyte infiltration, insulin content and cellular stress. Importantly, small particles such as viruses will also be detectable. The data first will be shared with nPOD co-investigators and at a later stage will be made publically available. (adopted by nPOD staff)

Jake Kushner Baylor College of Medicine Do de-granulated ß-cells persist in type 1 diabetes?

The goal of this proposal is to define the lineage mechanism of ß-cell persistence in type 1 diabetes (T1DM). A few ß-cells often persist in T1DM pancreata. But, the lineage mechanism of ß-cell persistence in T1DM remains poorly understood. Our overarching hypothesis is that ß-cell persistence in T1DM is due to ongoing ß-cell regeneration. Alternatively, ß-cell persistence in T1DM could be due regional differences in islet autoimmunity, or because ß-cells acquire a degranulated state that protects them from autoimmune mediated destruction. These hypotheses are readily testable with NPOD samples. However, by traditional techniques the analysis would require a huge number of slides and a many years to analyze. Our approach combines multi-labeling (5 antigens can be simultaneously detected) and high-throughput imaging technologies. This strategy will allow us to carry out this analysis using an absolute minimum of precious NPOD slides in a precise manner.

Christopher Rhodes University of Chicago “Empty Beta Cells” in type 1 diabetes

During the pathogenesis of Diabetes Mellitus, whether it be type 1 or 2 diabetes, islet cell heterogeneity becomes apparent. This is most commonly seen as a marked variability between the insulin content of some ß-cells versus others. Indeed, some ß-cells have such low insulin content that they appear to be empty. As such, in using insulin as a marker to estimate ß-cell numbers (as commonly done), one could be under-estimating the actual numbers of ß-cells presented since ‘empty ß-cells’ would not be accounted for. The concept of ‘empty ß-cells’ in the pancreas of diabetic patients could alter current thinking on ß-cell growth and regeneration. And, of course, if ß-cells are present in significant numbers, then it may become an issue of suitably refilling these cells with insulin rather than replacing/regenerating them.

We are interested in further examining the basis of ß-cell heterogeneity and ‘empty ß-cells’. Is this symptomatic of ß-cell dedifferentiation, a degranulated ‘hard working’ ß-cell and/or a stressed ß-cell approaching its end. We have used the transcription factor MafA as an alternative specific marker of adult pancreatic ß-cells, where a MafA+/Insulin- cell would indicate an ‘empty ß-cell’. But MafA is also susceptible to oxidation under oxidative stress conditions (e.g. hyperglycemia) and thus we are interested in examining other markers of ß-cell specificity to complement this analysis. We are also interested in examining adaptive plasticity of ß-cells and what mechanism guides the expansion of the rough endoplasmic reticulum, Golgi apparatus and trans-Golgi network reflective of upregulated insulin production to meet the increased demand in diabetes. Finally, we suspect that oxidative stress is a prime contributor to ß-cell destruction in the pathogenesis of diabetes and want to examine if an ‘empty ß-cell’ marks susceptibility to such stress and/or ß-cell very close to an apoptotic state.

Srinath Sanda & Rebecca Hull University of California, San Francisco Funded by: NIH IL-1beta expression in cystic fibrosis diabetes

Cystic fibrosis related diabetes is a major morbidity for patients with cystic fibrosis and negatively impacts mortality.  The mechanistic basis for this form of diabetes is not well understood.  In this project we are analyzing the hypothesis that islet inflammation leads to islet amyloid deposition and beta cell failure in patients with CFRD.  Samples from the nPOD repository are being stained for both amyloid and interleukin 1 beta.   In the future we will look at additional inflammatory mediators in the islets.

Lydia Sorokin & Eva Korpos University of Muenster, Germany Peri-capsular basement membrane degradation during leukocyte penetration Into the pancreatic islet during development of human type 1 diabetes and tertiary lymphoid organs in human T1D
Several steps are crucial for induction of type 1 diabetes (T1D); the first is extravasation of CD4+ T lymphocytes from blood vessels into the pancreatic tissue, the second is penetration of the peri-islet basement membrane (BM) surrounding the β-islets, and third the β-cell destruction which leads to appearance of disease symptoms.
BMs act to separate tissue compartments and represent barriers to the movement of both soluble molecules and cells. Hence, cells penetrating such protein barriers must employ specialized mechanisms (1,2). The main question addressed in our nPOD project is what is the mechanism used by leukocytes to penetrate the peri-islet BM and, thereby, reach the insulin producing  β-cells during the  development of T1D, in the non-obese diabetic (NOD) mouse model and in recently diagnosed T1D patients.
We recently published a comprehensive analysis of the extracellular matrix (ECM) composition of peri-islet capsules, composed of the peri-islet BM and subjacent interstitial matrix, in development of T1D in NOD mice and in human T1D (3), for which nPOD samples were crucial. Our data demonstrated global loss of peri-islet BM components only at sites of leukocyte infiltration into the islet in both mouse and man. Stereological analyses reveal a correlation between incidence of insulitis and the number of islets showing loss of peri-islet BM versus islets with intact BMs, suggesting that leukocyte penetration of the peri-islet BM is a critical step in disease development. We identified cathepsin S, W, and C activity at sites of leukocyte penetration of the peri-islet BM in association with a macrophage subpopulation in NOD mice and human T1D (3). Interestingly, the peri-islet BM is reconstituted once inflammation subsides, indicating that the peri-islet BM-producing cells are not lost due to the inflammation, which has important ramifications to islet transplantation studies (3).
We are now investigating how the cathepsins are involved in loss of the peri-islet BM: Studies are underway to detect and image cathepsin activity in mouse NOD and nPOD samples using cathepsin activity-based probes, which could lead to the development of novel diagnostic tools and therapeutic targets. In addition, we are examining the BM of the pancreas from different stages of human T1D and islet transplantation, with the aim of identifying ECM molecules that could promote islet survival.  
1) Korpos, E. et al, 2010. Cell Tissue Res. 339:47-57.
2) Wu, C. et al, 2009. Nat Med. 15:519-27.
3) Korpos et al, 2013. Diabetes 62:531-42.


Investigation of tertiary lymphoid organ formation in nPOD samples with ‘’typical insulitis’’

Lymphoid neogenesis is associated with chronic inflammatory diseases such as type 1 diabetes (T1D) and leads to formation of tertiary lymphoid organs (TLO). TLOs, like secondary lymphoid organs (lymph nodes and the spleen), are compartmentalized into T- and B-cell zones and contain reticular fiber networks composed of reticular fibroblast cells (RFCs) and reticular fibers (RFs), the latter of which are unique extracellular matrix (ECM) structures that have been characterized in our lab (Sixt et al, 2005; Song et al, 2013). The RFs of lymph nodes constitute the structural backbone of the organ, but also support the migration of T- and B-cells (Bajénoff et al, 2008) and conduct fluid transport of small molecules and soluble antigens (<70kDa), (Sixt et, 2005). Although TLOs have been described in many chronically inflamed tissues, their characterization from the ECM point of view is limited and their contribution to autoimmune diseases such as diabetes is not clear.
Using whole mount stainings and confocal microscopy we have characterized the cellular and ECM composition of RFs in TLOs of the inflamed pancreas of non-obese diabetic (NOD) mice, where the autoimmune destruction of insulin producing β-cells leads to replacement of pancreatic islets by TLOs. Our data suggest a role for the RFs in the conduction of chemokines from the infiltrating front to the high endothelial venules (HEVs) to propagate immune cell recruitment to the inflamed islet as well as antigen presentation to dendritic cells bound to the RFs, thereby, accelerating and promoting the inflammatory response. To be able to stem this acceleration step by interfering with TLO formation represents a potential mode of at least slowing T1D progression. However, it is first necessary to define whether the structure of the TLO is the same in mouse and in human T1D pancreatic samples with “typical” insulitis.
To investigate whether the presence of TLO in nPOD samples is correlated with insulitis we will perform immunofluorescence staining using antibodies specific for
1) different ECM molecules to identify the structure of the RFs;
2) HEVs, immune cell compartments and reticular fibroblasts to define the interconnection between the infiltrating front and the HEVs;  and
3)  auto-antigen (insulin) and chemokines to define whether the RFs also have a conduit function.
1) Sixt M, et al, 2005. Immunity, 22(1):19-29.
2) Bajénoff M, et al, 2008.  J Immunol, 181(6):3947–3954.
3) Song J, et al, 2013. Proc Natl Acad Sci, 110(31):E2915-24.
Ranjeny Thomas & Katharine Irvine University of Queensland Funded by: JDRF Pathological changes associated with chronic RelB activation in T1D

Dendritic cells (DCs) play a crucial role in establishing and maintaining the balance between immunity to pathogens and tolerance to self. Whether DCs evoke T cell activation or tolerance in response to antigen presentation is determined by the physiological context in which they differentiate and mature. DC differentiation and inflammatory signalling is abnormal in type 1 diabetes (T1D), but the relationship between this abnormality and disease pathogenesis requires clarification. We have shown constitutive activation and defective inflammatory responses in critical sub-units of the NF-kappaB transcription factor family, known as RelB and p65, which are required for myeloid DC differentiation, in T1D monocytes and DCs. Abnormal RelB activation may constitute a novel T1D risk phenotype. Our data indicate that peripheral blood cell RelB activation and associated inflammatory gene expression is increased in autoantibody-negative first-degree relatives of T1D patients. In autoantibody-positive and autoantibody-negative subjects, serum inflammatory cytokine expression is heterogeneous. We hypothesize that premature RelB activation and pro-inflammatory cytokine production during DC differentiation compromises the development of immunological tolerance and contributes to T1D pathogenesis. In this project we wish to investigate whether RelB activation and inflammatory cytokine expression in peripheral blood is associated with distinct pancreatic pathology in T1D and at-risk subjects. This information could enable us to stratify pancreatic pathology in living T1D patients and at-risk subjects, based on their peripheral blood RelB and inflammatory biomarker status. Moreover strategies for early therapeutic intervention to prevent or treat T1D may differ depending on pancreatic pathology.

Thomas Wight, Marika Bogdani & Susan Perigo Benaroya Research Institute at Virginia Mason Funded by: JDRF Extracellular matrix involvement in type 1 diabetes pancreatic islet destruction

The overall purpose of our research is to understand how changes in extracellular matrix affect insulitis initiation and progression and beta cell destruction in diabetes.

Our research focuses on a specific extracellular matrix molecule, hyaluronan, a long chain glycosaminoglycan distributed widely throughout different tissues. Hyaluronan is highly abundant in inflamed tissues and its synthesis is responsible for many of the physiologic changes associated with inflammation, including edema, vascular permeability changes, and leukocyte egress and activation at sites of injury, as well as the maturation of dendritic cells, antigen presentation, and the function and number of regulatory T cells. Hyaluronan exerts these effects through interactions and formation of macromolecular complexes with hyaluronan-binding molecules called hyaladherins, such as inter-α-inhibitor, versican, and tumor necrosis factor-stimulated gene-6.

We have made the novel observations that hyaluronan and hyaladherins increase in pancreatic islets of younger donors and accumulate in regions of lymphocytic infiltrates in T1D, and that both the amount and distribution of these specific extracellular matrix molecules vary with time since diabetes onset. We have also shown that hyaluronan and hyaladherins are deposited in the follicular germinal centers and T-cell areas, regions of immune cell activation, in both human pancreatic lymph nodes and spleen in type 1 diabetes.

We are investigating (1) the regulation of hyaluronan accumulation in pancreatic islets and lymphoid tissue in T1D, (2) the functional significance of hyaluronan – hyaladherin complexes in initiation and progression of insulitis, (3) the effect of altered hyaluronan synthesis and hyaluronan interactions with hyaladherins on diabetes development, (4) the role of hyaluronan – hyaladherin matrices in modulation of immune cell interactions and their adhesive and migratory properties, and in promoting T-cell activation and proliferation.

Our novel findings were obtained from study of human samples provided by nPOD. Analysis of those tissues also forms the basis of the work in progress on changes in hyaluronan and hyaladherin matrices in human tissues in diabetes.

Type 1 Diabetes Etiology & Environment

William Bachovchin Tufts University Funded by: NIH Fibroblast Activation Protein in Type 1 Diabetes

Fibroblast activation protein is an extracellular membrane bound and soluble protease that is commonly used as a marker of activated fibroblasts. Expression of FAP in healthy adult tissues is limited, but up-regulation is observed in a number of inflammatory and tissue remodeling processes, including wound healing, liver fibrosis and epithelial cancer, where it is thought to participate in remodeling of the extracellular matrix via its protease activity.

We hypothesized that FAP would also be up-regulated in the pancreatic remodeling and inflammatory process associated with type I diabetes. FAP has the potential to participate in this disease process in a number of ways. First, it has been suggested that immune cells cannot infiltrate pancreatic islets unless the membranous capsule is destabilized in some way. As a protease with collagenase activity, FAP may participate in degradation of the basement membrane allowing immune cell infiltration into the islets. Second, FAP may cleave hormones involved in glucose metabolism. The FAP KO mouse has been reported to have improved glucose tolerance, particularly after high fat feeding, suggesting as yet undiscovered substrates may exist for FAP. Preliminary work in our lab has identified FAP cleavage of a number of metabolically active hormones in vitro and FAP consensus sequences exist in many soluble factors including the hormone betatrophin. Thus FAP may have substrates of consequence related to beta cell regeneration or glucose control.  

The goal of this project is to quantify FAP activity in the pancreas and serum of type 1 and type 2 diabetics compared to controls. This study will also be, to our knowledge, the first to examine FAP activity in the human pancreas.     

David Carpenter, Miquel Porta & Sarah Howard University of Albany Persistent organic pollutants in the pancreatic tissue of people with and without diabetes

There is now strong evidence that persistent organic pollutants (POPs) such as polychlorinated biphenyls (PCBs) are major risk factors for type 2 diabetes.  While the mechanisms involved are not known with any certainty, the associations between development of type 2 diabetes and blood concentrations of POPs are stronger than those for almost any other factor, including obesity.  Some studies have suggested that obesity is in fact not a risk factor for type 2 diabetes, and that it is rather all of the POPs that are in the animal fats that are often consumed in excess that is really the risk factor. 

However almost nothing is known about any possible involvement of PCBs and other persistent organics as factors contributing to type 1 diabetes.  Type 1 diabetes is a quite different disease than type 2, but they do have some things in common.  Our proposal is to begin investigation of the possible role of POPs in type 1 diabetes by obtaining measurement of concentrations of PCBs in the pancreas of individuals with type 1 diabetes as compared to controls and those with type 2 diabetes.  PCBs will be used as an indicator POP.  All POPs are fat soluble and are found almost exclusively in the fat component of human tissues.  The pancreas, like every other organ, has some fat cells, and there are fats in the membranes of the beta cells that normally produce insulin.  One possibility is that there is excessive accumulation of POPs including PCBs in the pancreas in causes of type 1 diabetes.  This will be tested by direct determination of the PCB concentrations in tissues obtained through the nPOD project.  

Knut Dahl-Jorgensen & Lars Krogvold Oslo University Hospital & University of Oslo Funded by: European Union's Seventh Framework Programme Diabetes Virus Detection Study (DiViD)

The Diabetes Virus Detection Study (DiViD) is the first to examine fresh pancreatic tissue at the diagnosis of type 1 diabetes (T1D) for the presence of viruses. Surgical minimal pancreatic tail resection was performed in general anesthesia 3-9 weeks after onset of type 1 diabetes in six adult patients (age 24-35 years). Pancreatic tissues were snap-frozen within three minutes after dissection, live tissues collected for functional studies, and tissues fixated with formalin and glutaraldehyde and stored. Duodenal biopsies were collected by gastroscopy. Serum, PBMCs and serum/plasma were also collected. All patients had HLA-risk for T1D and at least one positive autoantibody. All six cases fulfilled the criteria for insulitis (5-58% of the insulin-containing islets in the six patients were affected by ≥15 T cells). We demonstrated a low-grade, persistent enterovirus infection in the pancreatic islets of Langerhans. Both viral genome and viral capsid protein was detected and confirmed in different laboratories. We also observed a reserve reservoir of betacells in the islets that improved insulin secretion in a non-diabetogenic environment in vitro. This reservoir may restore the need for insulin if the disease process could be halted at the time of diagnosis. Recent gene expression studies of laser-captured pancreatic islets from the same DiViD patients show significant enrichment of gene ontologies for viral and reproduction/infectious cycle, and also significant mitochondrial dysfunction and increased cellular endoplasmatic reticulum (ER) stress. Pancreas samples from the nPOD program, recovered from organ donors, will be used as a control group together with the data generated from the biopsy of the patients from the DiViD study.

Gun Frisk & Olle Korsgren Uppsala University, Sweden Funded by: EU and Diabetesförbundet Different expression of extracellular CAR in islets in pancreatic sections and isolated islets

The etiology of the type 1 diabetes (T1D) is unclear but it has been shown that genetic factors influence the pathogenesis. In addition, several studies have shown that environmental factors likely contribute to the disease development. One such factor is virus infections, particularly the Coxsackie B viruses (CBVs), are among the main candidates and there has been numerous of studies showing association between these gut viruses and T1D. To be able to infect a certain cell it has to express an appropriate receptor and the main receptor for CVBs is the tight junction protein Coxsackie-Adenovirus-Receptor (CAR). The expression of the CAR protein has been shown to be increased during healing and during inflammation. The primary aim of this study is to determine the expression of CAR in pancreas especially on b-cell, and to what extent this expression is affected by the disease progress in T1D. The expression on human isolated islets, infected and un-infected cultured for prolonged periods will be studied also be studied. CAR expression in pancreas will also be studied in human pancreatic tissue from recent onset T1D donors, long standing T1D donors, islet autoantibody positive donors, T2D donors and controls to see if this protein “normally” is expressed in pancreas, especially on the b-cells. Studies are also on-going to see if the expression is affected by the auto- immunity in the pre diabetic and T1D islets. Preliminary results indicate that the expression level of CAR is significantly higher in islets from diabetes related donors compared to controls.  It is also 10-100 fold higher expression in the endocrine tissue from pre diabetic and T1D donors compared to the exocrine tissue from the same donors.

Dale Greiner & Rita Bortell University of Massachusetts Medical School Investigate the molecular link between increased risk of diabetes among patients with psychiatric disorders

Compared to the general population, studies show 2-3 times increased risk of developing metabolic syndrome in drug naïve patients with severe mental illnesses like schizophrenia or bipolar disorder. Impaired glucose metabolism in non-obese never-treated patients and in first degree relatives of schizophrenic patients, suggests a possible genetic association between diabetes and schizophrenia. We are proposing to study the molecular cause of increased metabolic risk among patients with psychological disorders.

Heikki Hyoty University of Tampere, Finland Funded by: JDRF Virus detection in pancreas and other tissues

Enterovirus infections cause pancreatitis and damage endocrine cells in the pancreatic islets. Epidemiological studies have shown an association between enterovirus infections and type 1 diabetes, and viral structures have been detected in the pancreas of diabetic patients. Our recent epidemiological studies have suggested that among all different enterovirus types, the group of coxsackie B viruses is linked to the initiation of beta-cell autoimmunity and development of type 1 diabetes.  However, it is not known if this association is causal, and further studies are needed to address this important question. Current evidence suggest that enteroviruses can establish a slowly replicating persisting infection in the pancreas. In such case, the virus should be detectable by highly sensitive assays in the pancreas of diabetic and prediabetic patients. Our aim is to develop such assays, validate them in collaboration with other nPOD-V investigators and screen nPOD tissues for enteroviruses using the most optimal test panel.

Our primary aims include the development of different histological methods, as well as PCR-based techniques, for enterovirus detection in various samples. These techniques are utilized in detecting enteroviruses in various tissue samples – such as pancreas, duodenum and spleen – of T1D, pre-diabetic and non-diabetic donors, and further in comparing the prevalence of EV infections between different donor groups, as well as in different tissues. We are applying three main methods to accomplish our goals:

  1. Immunohistochemistry (IHC) with a wide panel of both commercial and in-house antibodies suitable for FFPE and frozen samples.
  2. Standardized in situ hybridization (ISH) assay suitable for FFPE and frozen samples. We are using novel probes for both general and serotype-specific detection of enteroviruses.
  3. Highly sensitive RT-PCR method for the detection of viral RNA and for the sequencing of viral genome.

Our studies have shown that these markers of enterovirus infection are more prevalent in type 1 diabetic and pre-diabetic subjects than in non-diabetic controls, and that the presence of enterovirus in pancreas correlates with that in other tissues.

Richard Lloyd & Joseph Petrosino Baylor College of Medicine Funded by: Baylor Diabetes Research Center Virome and microbiome in T1D onset

T1D is recognized to result from both genetic and environmental factors.  Chief among environmental factors that are strongly linked to T1D are Type B enteroviruses (HEV-B), yet the association of these viral triggers is not proven and many questions exist. Which of the 60 HEV-B serotypes trigger T1D, what type of infections do they cause in the pancreas, and how can they trigger autoimmunity are just a few. We are working within the nPOD-V, the virus working group, to attempt to amplify and isolate infectious viruses from nPOD samples, and to characterize the type of infections that occur in pancreatic cells.   Upon amplification, samples are subjected to next generation deep sequencing to identify all viruses and microbes present in the samples in an unbiased manner.  Samples are also examined using very sensitive qRT-PCR assays that detect single copies of enterovirus genomes.  This work will provide the most complete picture of what infectious agents may be harbored in the pancreas of T1D patients, both before and after disease onset.

Kathrin Maedler & Federico Paroni University of Bremen Funded by: ERC Detection of coxsackievirus infection in β-cells in T1D by short fluorescently labeled oligonucleotide probes

Type 1 diabetes (T1D) results from a complex interplay between genetic polymorphisms, immune system and environment. Viruses, mainly enteroviruses such as members of the Coxsackie B virus (CVB) family, have been suspected since long time to trigger diabetes and epidemiological studies support a causative role of viral infections during diabetes onset. Nonetheless, the presence of the enterovirus in diabetic tissue has not been fully confirmed. Here we present an adapted method to target single RNA molecules with short fluorescently labeled oligonucleotide probes in situ, that anneal to common regions of the coxsackievirus family. We successfully detect viral RNA in both cell culture and formalin fixed tissue sections bypassing the potential limitation of fragmented RNA. With the use of small contiguous RNA probes we were able to detect lower levels of strain specific infection compared to classical immunostaining. Specificity of detection was further confirmed without the well-known cross-reactivity with cellular proteins shown by antibodies. Furthermore the use of RNA probes allowed the discrimination between single RNA molecules and virus replication loci. Our data show a novel method to analyze human samples for the presence of enteroviral RNA and will be ideally suited to detect enterovirus infection with a very high sensitivity in paraffin embedded or frozen pancreatic sections.

Mary Markiewicz University of Kansas Medical Center Funded by: NIH/NIAID, American Diabetes Association, & Helmsley Charitable Trust George S. Eisenbarth nPOD Award for Team Science NKG2D ligand expression in type 1 diabetes development

Recent evidence derived from animal models of type 1 diabetes indicates that interaction of immune cells expressing the protein known as NKG2D with cells expressing other proteins, termed NKG2D ligands, plays an essential role in initiating diabetes development. In an effort to define more precisely the role these proteins play in type 1 diabetes progression, we developed novel mouse models. Based on results generated with these experimental models, we believe that expression of NKG2D ligands expressed on infiltrating immune cells within pancreatic islets is critical to the development of type 1 diabetes. Here we propose to confirm there is NKG2D ligand expression in the pancreas of type 1 diabetic donors.

Grant Morahan, Patrick Concannon & Stephen Rich University of Western Australia Funded by: National Health and Medical Research Council of Australia and DRF Genetics of nPOD

We have found that people with type 1 diabetes fall into one of six subtypes that can be defined on the basis of the genetic variants they have. These subtypes differ with respect to various clinical features, including immunological traits and risk of diabetic complications.

In this proposal, we plan to use genetic information from the nPOD donors to define which T1D subtype they belonged to. This will allow analyses of findings from nPOD researchers to see if any of the features they are testing differ between the subtypes. This could be very important information for understanding T1D and future treatment of people with T1D.

In addition, we will set up a user interface that will allow nPOD researchers to test whether any of the information they have found from nPOD samples vary according to any of the genetic variants that have been tested.

Noel Morgan & Sarah Richardson University of Exeter Medical School, UK Funded by: EU-FP7 Enteroviral infection as a causative factor in type 1 diabetes

Considerable evidence has accumulated implying that enteroviral infection is associated with the development of type 1 diabetes but the precise nature of this relationship remains unclear. Many studies have revealed that enteroviral RNA is detectable in patient serum at, or before, the onset of disease but it is unclear whether viral infection also occurs commonly in other organs and tissues at this time. We are investigating this possibility using pancreas (and other tissue) samples recovered from patients with recent-onset type 1 diabetes which are made available from within the nPOD collection. We study these samples in parallel with a larger collection available to us within the UK.

Our initial aims are to evaluate whether enteroviral infection can be detected at the level of the pancreas in patients with type 1 diabetes and to compare the prevalence with relevant age-matched controls. To achieve this, we are employing immunological approaches using antisera raised against enteroviral capsid proteins to detect infected cells and to provide unambiguous identification of these cells. Such studies are coupled to parallel experiments designed to define whether a “viral footprint” can also be found which would indicate that infected cells mount an active anti-viral response. Among the candidate molecules comprising such a “footprint” we are examining relevant pathogen recognition receptors, pro-and anti-apoptotic proteins, interferon response genes and downstream targets of interferon signalling.

A further goal is to establish whether the infection of islet cells by enteroviruses follows an atypical course in type 1 diabetes. In particular, we are developing methods to differentiate between more classical acute infections (which culminate in cell lysis) and those which persist for prolonged periods without causing the demise of the cells. We hypothesise that the development of a persistent infection within islet beta-cells might be critical to the initiation of islet cell autoimmunity and we are examining the molecular events that could underpin the development of such a response. 

Hervé Perron & Virginie Vidal Geneuro, Switzerland Evaluation of HERV-W Envelope antigen expression in Pancreas and serum from patients with Type 1 Diabetes.

Endogenous retroviruses are known to represent 8% of the Human genome.  HERV-W family retains elements expressing an envelope protein (Env), which activates a pro-inflammatory and autoimmune cascade through interaction with Toll-Like receptor 4 (TLR4). This Env protein was evidenced in brain lesions, sera and circulating mononuclear cells of patients with Multiple Sclerosis (MS)1. Thus, all the background science from the past two decades that has brought independent confirmations and reproductions of an involvement of HERV-W in MS 2 convinced us to further study its association with other “autoimmune diseases”. If no association was found when testing series of patients with rheumatoid arthritis3 or systemic lupus1,  about one-third of sera from patients with T1D revealed positive for the HERV-W Envelope antigen. We therefore further explored whether this immunopathogenic endogenous protein could be expressed in pancreas from T1D versus controls, which revealed positive in about 40% of first pilot series, and are now preparing to extend these analyses to larger number of serum, PBMC and pancreas samples. In parallel, we are developing an HERV-W-Env induce mouse model of diabetes for pre-clinical studies with an anti-Env neutralizing humanized antibody.

1.             Perron H, Germi R, Bernard C, et al. Human endogenous retrovirus type W envelope expression in blood and brain cells provides new insights into multiple sclerosis disease. Mult Scler. 2012; 18: 1721-36.
2.             Dolei A and Perron H. The multiple sclerosis-associated retrovirus and its HERV-W endogenous family: a biological interface between virology, genetics, and immunology in human physiology and disease. J Neurovirol. 2009; 15: 4-13.
3.             Gaudin P, Ijaz S, Tuke PW, et al. Infrequency of detection of particle-associated MSRV/HERV-W RNA in the synovial fluid of patients with rheumatoid arthritis. Rheumatology (Oxford). 2000; 39: 950-4.
Lorenzo Piemonti & Ilaria Capua San Raffaele Scientific Institute, Italy Funded by: Italian Ministry of Health Role of influenza viruses in the etiopathogenesis of diabetes

The rapid worldwide incidence increase suggests a major role for environmental factors in the aetiology of type 1 diabetes (T1D). According to cross-sectional and prospective studies on T1D patients and/or prediabetic individuals, virus infections may be one of these. Influenza viruses are known to cause severe pancreatic damage in birds and in some mammals, and we have preliminary strong evidence that these virus growth in primary pancreatic islet (human and mouse) cultures “in vitro”. The objective of our project is to demonstrate that influenza viruses may play a role in the etiopathogenesis of T1D. Due to the interdisciplinary nature of this study, two different Research Institutes will be involved: Istituto Zooprofilattico Sperimentale delle Venezie (I Capua, PI) for the influenza virus expertise and the San Raffaele Diabetes Research Institute (L Piemonti, CoPI) for the diabetes expertise.

The overall goal of this research is to demonstrate that influenza virus infection may be involved in the etiopathogenesis of diabetes.
Specifically we will aim to:
1)      Test “in vitro” the ability of influenza viruses to infect, replicate, and damage beta cell, pancreatic islet and endocrine cell
2)      Test “in vivo” (mouse) the ability of influenza viruses to induce beta cell damage and/or alterations of glucose metabolism
3)      Look for direct and/or  indirect  evidences of correlation between influenza viruses infection and autoimmunity/diabetes onset in humans
For the aim 3 we are testing the presence of influenza virus in pancreas at disease onset in T1D patients. For this purpose tissue samples are asked to the Network for Pancreatic Donors with Diabetes (nPOD). In situ hybridization will be performed to visualize viral RNA localized within cells using the Quantigene ViewRNA technique based on branched DNA signal amplification technology. Moreover the presence of influenza virus will be eventually confirmed by digital PCR and immunohystiochemistry
Alberto Pugliese Diabetes Research Institute, University of Miami The nPOD-Virus Group
nPOD has now assembled a “cloud” of investigators with diverse expertise, who are interested in collaborative studies, and is ready to take the nPOD research model to a higher level: the nPOD working groups. These groups are intended to collectively tackle key questions in diabetes research.
The nPOD virus working group (nPOD-V) represents a self-assembled a collaborative effort with the goal of investigating the role of viruses in type 1 diabetes through the study of nPOD samples. An innovative concept implemented by the group is “real-time” data sharing, to help coordinate studies and inform strategy adjustments. This approach should accelerate the rate of discovery and maximize the potential for robust results. The premise underlying this collaboration, based on the nPOD research model and the concept of data sharing and collaboration promoted by the nPOD leadership, has been fully embraced by the nPOD-V investigators; we believe that this in itself represent a major step-forward and innovation in the study of complex questions in human disease.
Many studies have linked enterovirus infection to islet autoimmunity and diabetes. Yet, many questions remain about which virus serotypes are linked to T1D, what type of infection they cause, and how this may contribute to the autoimmune process leading to diabetes. Earlier studies from individual laboratories suffered from limitations with samples and reagents, and did not have access to the latest methodological advances. This collaborative effort recognizes those limitations, and addresses them with an integrated and multidisciplinary approach that includes many innovative and powerful technologies that have not been used before for studying viruses in type 1 diabetes. The availability of shared tissues from the same patients and their coordinated analysis provide unprecedented opportunity that investigations can be exhaustive and the most informative.
The precise identification of enterovirus serotypes associated with T1D and a better molecular characterization of the virus host interaction and relation to pancreas pathology could lead to the identification of novel, T1D-specific, biomarkers of infection and potentially novel therapeutic targets. If a virus plays a role in islet autoimmunity, a vaccine or drugs that target viral responses could perhaps become important therapeutic avenues for preventing the triggering of autoimmunity or its progression, reducing patient burden and potential exposure to immunosuppressive or immunomodulatory treatments which can be associated with significant side effects and long term concerns, especially when children are concerned.
Nora Sarvetnick University of Nebraska Medical Center Funded by: NIH/ Helmsley Charitable Trust George S. Eisenbarth nPOD Award for Team Science Pancreatic antibacterial responses in T1D

The etiology of autoimmune diabetes is not known. However, the disease is accompanied by varying amounts of immune activation and inflammation. Although genetics clearly plays a role in determining who gets the disease, we do not know the instigating events. There has been a lot of investigation into the potential role of virus infections in triggering disease. However, there are no direct associations known between virus and disease in humans.

Recently, there has been increased interest in issues such as intestinal permeability and the microbiome in diabetes patients. One theory is that the leaky gut in T1D patients allows the penetration of bacteria and/or their products into the body, causing immune stimulation. When such products arrive into the pancreas after draining from the ducts, they cause changes in gene expression and inflammation. One set of factors that would get induced is bactericidal proteins, whose expression or subcellular localization is affected by exposure to bacterial products.

We have studied the expression of bactericidal proteins in circulation in new onset T1D patients. These studies indicated that there were coordinately high levels of expression in circulation in T1D patients less that five years after disease onset. These results suggest the upregulation of anti-bacterial responses in T1D patients. In this project, we will determine whether bactericidal factors are upregulated in the pancreas in T1D patients.

This project is important because it will tell us whether the disease may be mediated by the release of bacteria or their products into the pancreas, which would cause immune activation and inflammation. Several different bactericidal response elements will be investigated, each reflecting exposure to different bactericidal products including DNA, cell membrane constituents, and pigment molecules, an each providing an indication of the type of exposure the pancreas has encountered.

John Todd, Vincent Plagnol & Herbert Virgin, IV University of Cambridge, UK Funded by: JDRF Viral sequence discovery by high throughput DNA sequencing of the nPOD diabetic pancreatic samples

Several lines of evidence, in particular genetic association data of antiviral response genes with T1D risk, point to a role of viral infections in T1D etiology. However, multiple questions remain about the causal mechanisms and the type of viruses implicated. To contribute to these research questions, we are using high throughput sequencing technologies to find traces of viral RNA in the nPOD pancreatic organ donor T1D cases. Our strategy is based on deep RNA sequencing combined with sequence enrichment for known viral sequences to maximize the sensitivity of viral sequence detection. We then use bioinformatics tools to separate human and viral reads, characterise these reads and eventually provide a description of the virome of nPOD samples.

Yaron Tomer, Mihaela Stefan & Dirk Homann Icahn School of Medicine at Mount Sinai Role of environmental chemicals in the development of type 1 diabetes

There is a well-documented rise in the incidence of type 1 diabetes (T1D) in industrialized countries suggesting a role for environmental chemical exposures. Alarmingly, the increased incidence is seen mostly in children under age 5. However, research on environmental chemical exposures and T1D is very limited and there is an urgent need for research to fill this gap in our knowledge. Therefore, the goal of this proposal is to identify environmental pollutants that play a role in the increased incidence of T1D. We hypothesize that in the first years of life in T1D genetically susceptible children certain pollutants trigger damages specific to beta-cells making them more susceptible to the subsequent autoimmune attack. Of particular importance is the possibility that chemicals can directly affect beta cell physiology and thus may trigger initial insults precipitating the activation of self-reactive T-cells. Therefore, we have adapted a novel multi-chemical tissue mapping technology to study chemical quantitative distributions within the pancreatic tissues from T1D patients and healthy controls. Quantitative assessment of distribution of toxic and essential metals can reveal adverse (and protective) effects that occur at the cellular levels and may be overlooked in studies measuring average concentrations in homogenized pancreatic tissue. Laser ablation-based quantitative imaging provides a powerful tool to study the role of environmental toxicants and essential elements in T1D.

Antonio Toniolo, Andreina Baj & Roberto Accolla University of Insubria Medical School, Italy Funded by: CARIPLO Foundation Detection of enteroviruses in lymphoid tissue of donors with T1D of short duration and attempts to identify the infected cell type(s)

If type 1 diabetes (T1D) is caused/triggered by a viral infection in select genetic backgrounds – and Enteroviruses (EVs) are felt as major culprits –  investigators are expecting to find EVs into pancreatic Langerhans islets and also in association with spleen, lymph nodes, peripheral blood leukocytes. Infection of lymphoid cells, in fact, is common in systemic infections. Using an original method that couples virus amplification in susceptible cells with RNA amplification assays capable of detecting persistent and mutated EV types, we showed that different EV species are present in blood leukocytes of two thirds of children/adolescents at the clinical onset of T1D. Initial study of living cells from nPOD cases (T1D, non-diabetics) demonstrated that EV genomes and infectivity are present also in spleen and lymph nodes of T1D cases (1 to 30 years post-diagnosis). Thus, EVs appear to produce chronic silent infections in diabetes as in the post-polio syndrome (that is being studied by our group). If the viral genome is present, then virus proteins should also be detectable. Initial work done with colleagues looking at the same nPOD samples by ISH, immunohistochemistry, proteomics showed a strong concordance between genome detection and protein expression. So far, we have been successful in sequencing short stretches of RNA encoding conserved EV enzymes. We failed, however, in detecting genes encoding EV capsid proteins. This may be due to the pronounced variability of these proteins, especially in chronic infections. An additional aim of our studies is to determine which leukocyte subsets are carrying EVs. Preliminary results suggest that T cell subsets may be infected. Evidence in this direction, would push the idea that virus-induced immune disturbances might contribute to autoimmunity in T1D. Proof that chronic EV infections are indeed associated with T1D has not been reached, however collaborative studies of the same nPOD cases using many different approaches will certainly reveal if the virus hypothesis is a viable one. Should EVs be implied, major implications for human health will ensue (antivirals, novel preventive measures, etc.).

Steven Tracy University of Nebraska Enteroviral infection and T1D

Despite convincing evidence linking human enteroviruses (HEV) with type 1 diabetes (T1D) onset, it is not known whether  certain HEV are the ones which are most commonly associated with T1D onset or if T1D onset is associated with any HEV serotype.

Archived Projects

Vitaly Ablamunits, Kevan Herold & Jasmin Lebastchi Yale University Funded by: Yale DERC Comparison of total pancreatic insulin content and peripheral markers of beta cell function and desctruction

In both Type 1 and Type 2 diabetes, there is a gradual loss of pacreatic beta cells with time. The degree of beta cell destruction may be indirectly measured by the appearance of islet-specific components in circulation. For instance, insulin mRNA is increased in circulation in T1D patients before clinical loss of function of transplanted islets. We have recently demonstrated, that hypomethylated insulin gene characteristic for beta cells leaks into circulation during beta cell death and can predict hyperglycemia in both STZ-induced and spontaneous (NOD) mouse models. In pre-diabetic NOD mice, hypomethylated proinsulin DNA in circulation negatively correlated with total pancreatic insulin content measured in acid-ethanol extracts. However, in a natural history of prediabetes, the destruction may be slow at the very beginning due to low numbers of effector T cells, or be slow immediately before clinical precipitation of diabetes, because most of the target beta cell mass has been lost already. Thus the same level of destruction may have different prognostic value. The only accurate way to interpret the level of destruction would be to know residual beta cell mass. For that, in clinical setting, surrogate markets have been traditionally used, such as C-peptide, or, more recently, beta cell mass have been visualized by positron emission tomography. However, in humans, this data has never been directly correlated to total pancreatic insulin content. We hypothesize that circulating C-peptide (amylin, proinsulin) levels strongly correlate with total pancreatic insulin content, and that a ratio between the biomarker of beta cell death and C-peptide determines prognosis.

Martha Campbell-Thompson, Dirk Homann & Richard Benninger University of Florida Funded by: JDRF Mapping the histopathological landscape of human T1D: a pilot study (CLARITY)

The overall goal of this project is to extend studies examining human beta cells and immune cells in type 1 diabetes (T1D) to a new 3D fluorescent microscopy format for 500um or greater sized samples. These studies will adapt new tissue imaging methods (CLARITY) developed by  Dr. K. Deisseroth (Nat Methods 2013). The specific aims are: 1. Adapt tissue transparency methods for pancreas sample preparation and multiple immunofluorescence staining. 2. Create robust imaging and data collection methods to produce 3D islet topographical maps. 3. Examine islets in situ from donors with prediabetes and established diabetes for evidence of β‐cell regeneration and death. 4. Characterize immune cell subtypes infiltrating islets and their expression of cytokines and chemokines.

Raphael Clynes Columbia University Funded by: NIH, JDRF, and NIDDK TCR signal transduction in diabetogenic T cells

Tyrosine kinase inhibitors (TKis) effectively limit autoimmunity in animal models and are under clinical study in RA, MS and psoriasis. Recent studies in NOD mice have demonstrated proof-of-principal for promiscuous PTK blockade, using inhibitors initially developed to inhibit oncogenic pathways in cancer. Our recent work has shown that the selective Syk inhibitor R788 limits diabetes progression in NOD mice. Spleen tyrosine kinase (Syk) is most widely known as the PTK downstream of the B cell antigen receptor (BCR) and the IgG receptor (FcR) in myeloid cells and Syk PTKis have demonstrated clinical activity in RA and ITP. R788-treated NOD mice had decreased autoantibody levels presumably the result of BCR-inhibition and decreased T cell priming likely due to syk inhibition in DCs. A direct effect of Syk inhibtion on T cells was not reported, since syk expression in T cells was not expected. However, aberrant Syk expression may occur in autoimmune T cells. Indeed in our preliminary studies, discussed below, Syk is seen expressed in autoaggressive CD8 T cells infiltrating the skin and in activated murine and human T cells ex vivo.

Targeting of T cell activation via small molecule interference with protein tyrosine kinases is a promising approach to autoimmunity and could be directed at TCR signaling or co-stimulatory pathways. In T1D, these efforts will require determination of which signal transduction targets are specifically expressed in diabetogenic T cells in the islet and pancreatic lymph node. Autoaggressive T cells responding to sustained autoantigen drive may express signal transducing molecules not ordinarily expressed in resting or transiently active normal T cells thus providing attractive therapeutic targets. Our preliminary data shows that the Syk kinase is expressed in persistently activated T cells in vitro and in autoimmune T cells in vivo.

Dieter Egli, Ruddy Leibel Columbia University Beta cell defects in cystic fibrosis related diabetes

The aim of this study is to determine how CFTR mutations affect beta cell performance, and whether
their effect is direct (cell-autonomous), or indirect, by affecting other cell types/physiological
processes with secondary consequences for beta cell function and viability. It is likely that any
mechanistic insights obtained from the study of this specific etiology of diabetes will be applicable to
other types as well.

Roberto Gianani University of Colorado, Denver Characterization of type 1 diabetes subsets

Type 1 diabetes is a clinically and pathologically heterogeneous disease. A large subset of individuals with Type 1 diabetes has an autoimmune form of the disease characterized by subtotal loss of beta cells and positivity for islet autoantibodies as well an increase prevalence of DR3 and DR4. We have recently characterized a pathological pattern of beta cell loss in Type 1 diabetes that does not appear to be associated with autoimmunity. We hypothesized that this represent a novel disease characterized by “degenerative” loss of beta cells possibly due to increased apoptosis (affecting also alpha and delta cells), pathogenetically distinct from the beta cell loss seen in Type 2 diabetes.

The Specific Aims Are:
1) Morphometric characterization of Pattern B type 1 diabetes in relation to the beta cell, alpha cell and delta cell relative volume as compared to controls and type 2 subjects.
2) Investigation of surviving expression at the protein and RNA level (initially these studies will be limited to survivin,
3) Comparison of apoptosis rate between different groups
4) Investigation of amiloidosis in the different groups
5) Collaborative efforts with Dr. Sechi Laboratory.
Paul Harris, Masanori Ichise & Matthew Freeby Columbia University Funded by: NIDDK and JDRF Determination of specific and non-specific binding of 18F-FP-DTBZ in whole pancreas homogenates obtained from controls and patients with longstanding type 1 diabetes

In the past we identified Vesicular Monoamine Transporter Type 2 (VMAT2), as a biomarker of beta cell mass that is quantifiable in vivo by Positron Emission Tomography (PET). PET is a tomographic imaging technique providing accurate non-invasive dynamic measurements of regional PET tracer uptake and clearance. These measurements are used to calculate VMAT2 density in the pancreas as a surrogate measure of beta cell mass.  Immunohistochemical staining of both control and long term T1D pancreata show excellent correlation between VMAT2 and insulin staining.  When PET and the radiopharmaceutical 18F-FP-DTBZ are used to non-invasively measure VMAT2 in vivo, we observe that healthy subjects have a significantly greater pancreatic VMAT2 signal than long term T1D patients with documented inability to secret normal amounts of insulin.  However there is a significant residual VMAT2 signal in the pancreata of the T1D patients.

The goals of our project are to understand the nature of this background signal.  There are three not mutually exclusive explanations of the background signal observed in pancreata believe to be beta cell ablated due T1D; 1) The radioligand 18F-FP-DTBZ is binding specifically to VMAT2 or some other receptors, not present in beta cells (specific and displaceable off-target binding).  Specific off target binding is being measured in a clinical PET study currently in progress. We have also quantified, using immunohistochemistry, the expression of VMAT2 laying outside the beta cell subset in human pancreata (e.g. PP cells).  The specific off target bind does not seem to account for the background,  2) 18F-FP-DTBZ may also bind non-specifically and non-displaceably to pancreas tissue (off target non-specific binding).  The in vivo non-specific binding is currently being measured in clinical PET study in progress.  Our in vitro estimates of non-specific binding using cadaveric pancreas tissue supplied by NPOD suggest that a significant portion of the in vivo background signal is due to non-specific binding.  However, these in vitro methods may overestimate in vivo binding of the tracer and must be interpreted with caution, 3) The last formal possibility, that our measurements of VMAT2 in the long term T1D pancreas indeed are on target and specific, representing residual VMAT2 possibly associated with beta linage cells is now under consideration.  Such a possibility seems to conflict with preexisting immunohistochemistry data, but immunohistochemistry is not the most sensitive technique available for detecting low levels of expressed proteins.  The next phase of our project will be to study long term diabetes pancreata for VMAT2 gene expression using in situ hybridization and RT-PCR techniques.  While the pancreas is a difficult organ to obtain RNA suitable for such studies, the generally excellent quality of NPOD tissues will improve our chances for success. 

Akihisa Imagawa & Toshiaki Hanafusa Osaka University, Japan Funded by: Gift Fund Histological differences between Japanese and Western type 1 diabetes

(1) To clarify the infiltration of mononuclear cells in exocrine pancreas of Western type 1A diabetes and compare the histological findings to those of Japanese patients

(2) To clarify the expressions of the molecules related to innate immunity or viral infection in Western type 1A diabetes and compare the hitological findings to those of Japanese patients

George King & Susan Bonner-Weir Harvard University Funded by: JDRF Joslin Medalist Study
History of the Joslin Medalist Program
Joslin first began awarding medals to people with diabetes in 1948 with a 25-year Victory Medal. Believing that proper self-management was the key to minimizing long-term complications, the program was the vision of Elliott P. Joslin, M.D. and served as an incentive for those committed to good, though challenging, diabetes care. In the early 1950s the name was changed to the Blue Ribbon, and as more and more people lived long healthy lives with diabetes it finally became the 25-year Certificate that is awarded today. In 1970, Joslin expanded the program and began awarding a 50-year bronze medal. And Joslin presented the first 75-year medal in 1996.
50-Year Medalist STudy
Joslin’s 50-year medalists are invited to participate in a special study, examining outcomes of long-term diabetes. The study attempts to understand what factors contribute to the longevity of individuals who have received this honor. Currently, over 400 medalists are interested in participating in the study. To date, over 300 of these have completed an extensive questionnaire about their life with diabetes. Data from this questionnaire suggest that the risk of kidney, eye, and nerve problems is different after 50 years with type 1 diabetes than the risk among all individuals with the disease.
A second study is currently being undertaken, which examines factors in the blood and DNA that may help in modifying the risk for complications and survival. Participation in this study is open to all individuals residing in the United States who have received the Joslin 50-Year Medal. In coming months, updates for this study and findings from the previous study will be posted on a separate Web site. 
Rudolph Leibel & Streamson Chua Columbia University Funded by: NIDDK Identification of a gene regulating pancreatic beta cell replication

We cloned a genetic modifier of T2D in obese mice. The gene, Ildr2, encodes a molecule that plays a role in both lipid metabolism and cellular responses to stress.  Animals with lower levels of expression of this gene has reduced number of insulin producing cells in the pancreas, and, ultimately went on to develop mild diabetes.  How the apparent functions of Ildr2 account for these effects is currently under study by us using genetically manipulated mice.

Rudolph Leibel, Wakae, Nao Columbia University Funded by: NIH E-cadherin mediates developmental effects on the proliferation and the function of β-cells in the islet of Langerhans
In rodents and humans, the rate of beta cell proliferation declines rapidly after birth; formation of the islets of Langerhans begins perinatally and continues after birth.  
We tested whether increased levels of E-cadherin during islet formation mediate the decline in beta cell proliferation rate by contributing to a reduction of molecules that mediate cell replication. In vitro, a cultured beta cell line formed into clusters resembling an islet  displayed increased E-cadherin but decreased nuclear β-catenin and cyclin D2 (cell division molecules) , and reduced rates of cell proliferation, compared with monolayers. 
Reduction of E-cadherin increased cell proliferation and levels of cyclins D1 and D2. 
After birth, beta cells showed increased levels of E-cadherin, but decreased levels of D-cyclin, whereas islets of mice in which the E-cadherin gene had been knocked out showed increased levels of D-cyclins and nuclear β-catenin, as well as increased beta cell proliferation. These islets were significantly larger than those of control mice . Interestingly, these changes correlated with reduced insulin response to ambient glucose, both in vitro and in vivo.
The findings support our hypothesis by indicating an important role of E-cadherin in the control of beta cell mass and function.
Douglas Melton Harvard University Funded by: NIH Generation of tools to distinguish human pancreatic cell populations

Our work aims to understand the defects were deficiencies in the pancreatic beta cells of diabetics. We use a number of molecular methods, including but not limited to single cell RNA sequencing, to characterize the molecular phenotypes of islet cells.

Jerry Nepom Benaroya Research Institute at Virginia Mason Funded by: NIAID Novel recombinant antibodies directed to diabetes-associated autoreactive T cell epitopes

The trimolecular complex composed of autoreactive T-cell receptor, MHC class II, and an autoantigenic peptide plays a central role in the activation of pathogenic islet-specific CD4+ T cells in type 1 diabetes (T1D). We isolated and characterized novel antibodies against autoreactive T-cell epitopes associated with T1D. Our antibodies mimic the specificity of the T cell receptor (TCR), while binding MHC class II/peptide complexes in an autoantigen peptide-specific, MHC-restricted manner. The isolated TCR-like antibodies were directed against the minimal T cell epitope GAD-555-567 in the context of the HLA-DR4-diabetic-associated molecule. A representative high-affinity TCR-like antibody clone (G3H8) enabled the detection of intra- and extra-cellular DR4/GAD-555–567 complexes in APC. I561M single mutation at the central position (P5) of the GAD-555-567 peptide abolished the binding of G3H8 to the DR4/GAD complex, demonstrating its high fine TCR-like specificity. The G3H8 TCR-like antibody significantly inhibited GAD-555-567 specific, DR4-restricted T cell response in vitro and in vivo in HLA-DR4 transgenic mice.

Our findings constitute a proof of concept for the utility of TCR-like antibodies as antigen-specific immunomodulation agents for regulating pathogenic T-cells and suggest that TCR-like antibodies targeting autoreactive MHC class II epitopes are valuable research tools, which enable studies related to antigen presentation, as well as novel therapeutic agents, that may be used to modulate autoimmune disorders such as T1D.

We have demonstrated that G3H8 can be used to detect and visualize GAD presentation by APC. The TCR-like fine specificity and requirement for both GAD-555–567 peptide and the HLA-DR4 molecule for G3H8 recognition were verified. Moreover, we have identified the P5 position in the center of the peptide to be essential for G3H8 binding of the DR4/GAD complex. This position was found to play a critical role in T cell recognition [24], and thus the overlap in P5 dependency between the TCR and our G3H8 TCRL antibody implies that our antibody has fine specificity and TCR-like characteristics. Once the fine specificity of G3H8 was determined, we used it to detect naturally processed GAD epitopes. G3H8 stained specific MHC class II
complexes derived from processing of the whole GAD antigen by GAD-transfected Priess cells. The lack of extracellular staining by G3H8 suggests that most of the DR4/GAD complexes in these cells were located inside the cells, whereas the number of complexes presented on the cell surface was below the detection threshold of G3H8. Our failure to stain nPOD pancreata is likely due to this same issue, and in addition we had background issues that were never resolved. We concluded that use of this novel reagent needs a substantial amount of the epitope on the APC for detection (only seen when peptide was added to cells or after an immunization in mice).

Bart Roep Leiden University, Netherlands Detection of islet autoreactive CD8 T-cells in insulitis versus periphery

To determine the specificities and frequencies of islet autoreactive CD8 T-cells in pancreas draining lymph nodes, control lymph nodes, spleen and/or peripheral blood from diabetic and non-diabetic organ donors, we used the Diab-Q-kit, a nanotechnology that allows the direct detection of circulating autoreactive CD8 T-cells against an array of islet-epitopes simultaneously, reducing the needs of patient resources. The detection of these T-cells is based on the T-cell receptor (TCR) specificity for its peptide-HLA ligand. We produced HLA-monomers for an array of islet epitopes presented in the context of HLA -A2, -A3 and -B7. This allowed expanding the number of patients that could be analyzed considerably. Moreover, to characterize the memory phenotype of the autoreactive T-cells, we validated and upgraded the Diab-Q-kit through the inclusion of cell surface markers to distinguish naïve, effector memory and central memory T-cells. From the nPOD repository we selected a total of 29 cases to be analysed for this study from which we could obtain cell suspensions from blood and pancreas draining lymph nodes, control lymph nodes and/ or spleen.

Upon thawing of the samples we unexpectedly observed that far fewer viable cells could be recovered from the majority of the samples than the number reported to be frozen on the vial. For some cases, we also counted the number of dead cells in each vial. We found that the percentage of dead cells was very high in some cases, but not all. As an internal control, we always included samples from healthy controls collected and cryopreserved in our laboratory, which showed very high viability after thawing (>90%), indicating that the thawing procedure was correctly performed.

Similar issues on viability and recovery have since been reported by other partners in nPOD. Importantly, the quality control to detect any issues was critical to identify these problems and potential causes. For instance, issues with viability and recovery were determined immediately after thawing, while FACS analyses performed hours after thawing of the samples would have suggested excellent viability outcome, even in cases that QC earlier on had identified poor recovery and viability.

The consequences of low viability and recovery may differ depending on the type of analysis per project (functional assays vs. mRNA) and incentive (cloning T-cells may still be possible, but may not be representative of the total tissue sample). Such issues have been evaluated in depth and at length between the central facility and participants of nPOD. Indeed, the nPOD central facility was indispensable in identifying important correlations between recovery, viability and donor characteristics that may differ between tissue source of the leukocytes and that need not be independent co-variates ( PBMC: demographics, cause of death [anoxia vs > CV/stroke], disease [T2D/Ab+>T1D], donor age [o>y]; splenocytes: donor age [y>o], BMI [low>hi], transmission time [short>long]),  underscoring that the thawing procedure was not the primary parameter of poor outcome and implying that intrinsic features of the biosamples determines quality and reward.

Despite lower cell recovery than expected, we nevertheless proceeded performing the Diab-Q-kit analysis on all samples, to take maximal knowledge, benefit and reward from the precious nPOD material. Unfortunately, for more than half of the samples no reliable data could be obtained, due to the low number of CD8 T-cells present in each sample. This precluded reliable comparisons in frequencies and specificities of islet-autoreactive CD8 T-cells in the individual organs. The results from this study have been discussed at length with JDRF nPOD, and it has helped the nPOD consortium to decide on the feasibility of projects where viable cell suspensions are needed for analysis and functional assays. Measures have been undertaken to adjust allocations of samples on basis of this important new insight.

Åke Sjöholm Södertälje Hospital, Södertälje, Sweden Ljungan virus in T1D

It was earlier reported the finding that the incidence of type 1 diabetes correlates with small rodent abundance in Sweden. This and other findings initiated an effort to isolate novel infectious agents from small rodents in an attempt to find an etiological agent for type 1 diabetes in humans. As a result of these efforts, a new picornavirus (Ljungan virus [LV]) was isolated from bank voles trapped in Sweden. Moreover, it has been shown that more than one-third of wild-caught bank voles and their offspring develop diabetes when kept under laboratory conditions. This was accompanied by changes in pancreatic islet structure, islet cells becoming immunohistochemically positive for LV and elevated levels of islet autoantibodies. The finding of LV in the pancreatic islets of the diabetic bank voles raised the general question whether or not LV can participate in the development of diabetes in rodents as well as in humans.

In pilot experiments using human pancreatic specimens from nPOD, we have found faint immunostaining for LV in two autoantibody-positive patients using a rabbit anti Ljungan virus protein 1 (LV VP1) antibody. Other type 1 diabetic patients, as well as non-diabetic controls, stained negative for LV in the pancreas. Due to the weak intensity of the immunostaining, results are inconclusive and not further pursued at this time.

Jo Spencer King's College London Funded by: JDRF Phenotype, specificity, diversity and clonal spread of B cells in T1DM

In this preliminary screening of tissues, we propose to characterize B cells in the pancreas of cadavers that are autoantibody positive. Pancreas of non-autoantibody positive cadavers will be studied as controls. We propose to use immunohistochemistry to analyse cell surface marker expression, together with an analysis of the microantomical distribution of B cells and their relationship to other cells. We will attempt to locate specific antibody producing cells. We will be particularly interested in a population of large B cells with dendritic processes that often connect with T cells. These cells, which are normally located in thymus and in T cell zones of organized lymphoid tissues may be antigen presenting B cells.

Gerald Taborsky, Jr. University of Washington Funded by: NIH Glucagon secretion and islet neuropathy

To confirm and publish our preliminary observations (Diabetes 55 (supplement 1):64-OR, 2006) suggesting that humans with Type 1 Diabetes Mellitus have a marked and islet-selected loss of their sympathetic nerves, similar to that we have observed in animal models of autoimmune diabetes. If true such a defect may contribute to the impaired glucagon response to induced hypoglycemia seen early in Type 1 diabetes.

Thomas Waldmann & Jing Chen NCI/NIH Funded by: Metabolism Branch, NCI, NIH, DHHS The role of IL-15 and IL-15Ra in the pathogenesis of type 1 diabetes

Interleukin-15 (IL-15) is a pro-inflammatory cytokine that promotes the activation and maintenance of natural killer (NK) and CD8 (+) T-effector memory (T-EM) cells. In patients with type 1 diabetes, elevated serum levels of IL-15 have been reported. Using an assay developed in the lab, we demonstrated that the serum sIL-15Rα levels were significantly higher in patients with T1D compared with those of normal controls. To investigate whether islet overexpression of IL-15 and IL-15Rα could play a role in the pathogenesis of T1D, we generated double transgenic mice with β cell specific expression of both IL-15 and IL-15Rα under a rat insulin promoter. The mice developed hyperglycemia, marked mononuclear cell infiltration, β cell destruction and anti-insulin autoantibodies that mimic the early events of human T1D. Inhibiting IL-15 signaling with anti-IL2/IL15Rβ (anti-CD122), which blocks IL-15 transpresentation, or with the Janus kinase 2/3 inhibitor tofacitinib, reversed the diabetes in the transgenic mice. This suggested a causative role of IL-15 and IL-15Ra  in the induction of insulin dependent diabetes in mice.

To further investigate whether IL-15 and IL-15Ra play a role in the pathogenesis of human type 1 diabetes, we investigated the expression of IL-15 and IL-15Ra in the islets of patients with type 1 diabetes, anti-insulin autoantibodies and normal donors.  Immunohistochemistry demonstrated IL-15 expression in the islets in all T1D patients examined whereas IL-15 was detected only in the pancreatic islets in one out of four normal donors and two out of five autoantibody positive donors. Interestingly, in two cases where IL-15 was positive, we also detected islet expression of MxA, a GTPase induced by type 1 IFN (α/β) that has been reported to interfere with virus multiplication and spread. Furthermore, IL-15 and IL-15Rα mRNA levels were also higher in the islets from T1D patients compared to that from autoantibody positive donors or normal donors. The islets were microdissected based on insulin positivity using laser capture microscopy (LCM). Taken together, our data suggest that disordered IL-15 and IL-15Rα may be involved in the pathogenesis of human T1D.

Clive Wasserfall University of Florida Humoral immunity in type 1 diabetes

Our project entitled humoral immunity investigates several serological aspects of Type 1 Diabetes (T1D). Our group also serves as the ELISA autoantibody core wherein we oversee the screening laboratories and together with the University of Colorado maintain quality assurance and control for this program. However we also conduct research into additional aspects of autoantibodies from nPOD such as Islet cell autoantibody assays on autologous tissue. From this the idea of looking for complement deposition in pancreas tissue of autoantibody positive donors as well as those with T1D arose. We published the findings in Diabetes Care and in sum we found that C4d deposition was mainly associated with blood vessels and the extracellular matrix that surrounds blood vessels an the pancreatic ducts. This was most pronounced in T1D pancreas tissues and was interpreted as evidence of ongoing inflammation of the pancreas in this disease.