Session 5: Islet - Immune interactions

Pancreatic islets contain resident myeloid cells that include macrophages and dendritic cells, with additional monocyte recruitment following autoimmune infiltration. These CD11c-expressing myeloid populations have diverse roles in modulating the T cell response in the islets over the course of islet infiltration and destruction. Data will be presented showing that islet CD11c+ cells are critical gatekeepers for lymphocyte entry into the pancreatic islets. Additionally, a subset of CD11c+ cells in islets with mild insulitis serve as antigen presenting cells that drive restimulation and pathogenesis of autoreactive T cells. However, as insulitis progresses, infiltrating CD11c+ cells drive a tolerance checkpoint that functions in the islets. While the functions of CD11c+ cells in the islets are varied, the mechanisms by which they modulate the autoreactive T cell response in the islets may be novel targets for disrupting the ongoing autoimmune response.

Type 1 diabetes (T1D) occurs following a breakdown in immunological tolerance resulting in the eventual loss of functional beta cell mass. The natural history of disease includes a high degree of variability in clinical presentation between subjects with T1D. Recent studies in organ donors with T1D highlight a growing appreciation for the variable loss of insulin production within individual islets from those with established T1D. These findings support a model wherein a loss of immune tolerance is regionalized and under checkpoint control.
The factors impacting disease heterogeneity likely include genetic, stochastic, and environmental contributors. Data will be presented from a large cross-sectional cohort of peripheral blood samples assessing the impact of genetic susceptibility alleles on the frequency and phenotype of innate and adaptive immune subsets. We will highlight the impact of age and CMV serostatus on immune phenotypes thought to play an important role in disease pathogenesis. As a model candidate gene, we will present data on the CD226 costimulatory axis and discuss emerging mechanistic data from the NOD on its role in T cell activation. Collectively, the data presented will address a number of sources of disease heterogeneity within the immune system and include integrated studies of peripheral blood, disease-associated tissues, and animal models for functional testing.  This integrated systems approach should help to identify disease endotypes that could benefit from rational targeting of immune checkpoints to restore immune regulation in T1D.  

Type 1 diabetes is an autoimmune condition in which the islets of Langerhans become infiltrated by lymphocytes which then mediate beta cell dysfunction and loss. This model has been developed mainly from studies in rodents but it is less clear whether equivalent processes occur in humans. To address this, we have studied human pancreas samples recovered at autopsy to characterise the process of islet inflammation and beta cell demise in people who died soon after the onset of type 1 diabetes. This has revealed the existence of two distinct patterns of insulitis which vary according to the absolute numbers and relative proportions of influent immune cells and in relation to the age at onset of disease. Among children diagnosed in the earliest years of life (<7y) the autoimmune attack is relatively aggressive and leads to early and extensive loss of beta cells.  In those diagnosed in their teenage years (>12y) islet inflammation is less intense and associated with the maintenance of a greater proportion of insulin-containing islets (typically ~40%) at onset. Detailed analysis of the distribution of insulin and proinsulin in the residual insulin-containing islets of children diagnosed <7y reveals that the two are highly co-localised, suggesting a marked defect in prohormone processing. Among people diagnosed >12y, this defect is also evident but only in a small subset of insulin-containing islets. Importantly, the pattern of proinsulin processing correlates closely with the defined immune phenotype. This was seen most clearly when studying children diagnosed in the intermediate age range (7-12y) where the proportion of islets displaying aberrant proinsulin processing correlated firmly with the islet immune phenotype. These observations imply that type 1 diabetes in children comprises two distinct endotypes which are readily distinguished by pancreatic immunohistology. Understanding these differences will be crucial to the design of immunotherapies intended to minimise beta cell loss.