Objective
Not everyone who lives with Type 1 diabetes has a similar journey to disease onset and management. Children diagnosed in their earliest years tend to experience greater challenges in disease management than those diagnosed in adulthood. Our previous studies of the pancreas in people of different ages has provided some insights into why this may be. We have observed that the autoimmune processes targeting the insulin-producing beta cells are very aggressive in individuals with early onset such that, by the time they experience clinical symptoms they have few residual beta cells left. By contrast, those who are diagnosed later in life, retain a significant number of residual beta cells at onset, and these can be preserved for many years post-diagnosis. It is well-established that individuals who retain some beta cells can better manage their blood sugar levels and have a reduced risk of long-term complications. We do not understand why the autoimmune processes differ between individuals diagnosed at different ages, and this proposal plans to address this. To understand what might be different, we first need to have a detailed understanding of the normal pancreas and how it changes during early life into adulthood. Our preliminary findings have demonstrated that during the period where the first signs of type 1 diabetes development can be detected in the pancreas, the gland is undergoing significant changes in architecture. We hypothesise that the maturity of the pancreas at the time the disease process starts will determine the clinical course of the disease. This proposal seeks to characterise architectural changes in the developing pancreas and determine why and how beta cells might be targeted at different stages of life.
Background Rationale
Surprisingly, despite many years of study, there is still a large gap in our knowledge about how the pancreas normally develops in early life. This is primarily due to the limited availability of organs from young donors. To begin to address this gap, we have combined archival and contemporary pancreas bioresources, leveraging previously stained tissue sections from 100’s donors, comprising thousands of slides from people of different ages, disease statuses and ethnicities. We have developed AI-assisted pipelines to accurately recognise architectural pancreas features, including hormone-producing cells, and have quantified key parameters across entire tissue sections. We have found that when pancreatic endocrine objects are enumerated (from single cells to large islets), an unexpectedly large proportion consists of single cells/ small clusters of cells. This proportion is highest among the youngest donors (>70%), and it decreases with age. These endocrine objects have been largely overlooked in earlier histological studies, and they are not included in studies of isolated islets since, due to their small size, they are lost during the isolation process.
Our studies have provided insights into how the pancreas changes during early normal development, when there is a dramatic increase in the size of the organ. We have demonstrated that, with increasing age, there is a progressive shift towards larger endocrine objects (islets), suggesting that single endocrine cells and small endocrine clusters are key components of the anatomical mechanisms used to generate larger islets during development. We have further shown that the single cells and small endocrine cell clusters are predominantly made up of beta cells. To our surprise, when examining the pancreas of individuals with Type 1 diabetes we noted that single beta cells and small clusters of beta cells are virtually absent, and we have confirmed this in multiple cohorts. By contrast, the residual insulin-containing islets that persist in type 1 diabetes are relatively large and appear to be protected preferentially. Our data are consistent with the hypothesis that, in type 1 diabetes, autoreactive immune cells preferentially target single and small beta cell clusters. Since these are more prevalent in the early years of life, we reason that larger islets may form less frequently in individuals progressing to type 1 diabetes, and that this effect would be more profound the earlier that autoimmunity appears. To verify this, we have demonstrated a reduction in the density of larger islets in donors with type 1 diabetes, particularly those with younger onset and short disease duration.
Collectively, we have identified a significant gap in our knowledge regarding small endocrine objects with the pancreas and how these may change during early life in health and disease. This proposal aims to address this gap in knowledge using several approaches that will generate transcriptomic, proteomic and functional data on these small endocrine objects. Spatial transcriptomics will allow us to capture the positional context of transcriptional activity (gene expression) within intact tissue at single-cell resolution. From this, we can determine if genes expressed in single beta cells differ from those expressed in beta cells within larger islets. This will then be validated at the protein level. We will further explore if there are differences in the expression of key targets of the autoimmune processes between small clusters and larger islets and if this changes during development. Finally, using living pancreas slices, we will assess the function of beta cells in small objects and compare this with that of beta cells in larger islets.
Description of Project
Insulin-producing beta-cells are present in the pancreas where they associate with other hormone-producing cells to form defined clusters, known as islets of Langerhans. Recent advances in 3D visualisation of the whole human pancreas have revealed a surprise by suggesting that many of the beta cells reside outside the islets and occur as single cells or form very small cell clusters. For technical reasons, these single beta cells and the very small clusters have eluded previous study. Our preliminary data have replicated these unexpected findings using 2D analysis of pancreas tissue sections from >300 individuals using AI-assisted pipelines. We have also demonstrated that in very early life, the majority of beta cells reside in the small endocrine object clusters initially but that, as individuals age, they are found mainly in larger endocrine objects (islets). Strikingly, we have also shown that the small clusters of beta cells are virtually absent in individuals with Type 1 diabetes and that the total number of larger islets is dramatically reduced, particularly in individuals diagnosed with diabetes at a young age. We have previously revealed differences in the underlying autoimmune processes that cause beta-cell loss in individuals diagnosed with diabetes at different ages and we now hypothesize that such differences are driven by the structural organisation of the pancreas prevailing when the disease process initiates. In particular, we propose that in the youngest subjects (whose beta cells are present mostly in small endocrine objects) these are readily targeted during the autoimmune response, to halt islet development and minimise the formation of larger islets.
This proposal will study beta cells with different pancreatic localisation using advanced spatial transcriptomic methodologies. The techniques will allow us to compare beta cell gene expression profiles between cells occurring singly or in small clusters versus those localised in larger islets. We will explore how differences in gene expression change during early-life, as the cells present in small clusters develop to form larger islets. This will provide insights into how islets are formed normally and why certain beta-cells might be more susceptible to autoimmune-mediated attack. Differentially expressed genes will be studied at the protein level to confirm and validate the findings. The presence of key protein targets of the autoimmune processes other than insulin (e.g. ZnT8, IA2 and GAD65) will also be studied during early-life to establish if altered expression of these antigens could explain early differences seen in the autoimmune responses in individuals at-risk of type 1 diabetes. Finally, in collaboration with colleagues at the City of Hope, we will use living human pancreas slices to explore whether the secretory dynamics of beta cells occurring singly or in small clusters differs from those localised with other endocrine cells in larger islets. This will reveal whether all beta cells respond to changes in blood sugar in a similar way or if some beta-cells play an alternative role within the wider development and function of the pancreas.
Collectively, these studies will address a large gap in our understanding concerning a previously overlooked population of beta cells that are lost early in the disease process in individuals with diabetes. Developing our understanding of how and why these cells are targeted by the autoimmune processes will suggest how we may be able to protect them in people at-risk of diabetes. It could also inform the optimal timing of immunotherapeutic interventions and provide knowledge to support islet replacement strategies.
Anticipated Outcome
This proposal will fill a gap in our knowledge surrounding the human pancreas's single/ small clusters of beta cells. It is likely to provide key insights into how these objects can contribute to the formation of larger islets throughout normal development and why they are preferentially targeted by autoimmune-mediated processes in type 1 diabetes. The emerging datasets will be a key resource for the field.
Relevance to T1D
The findings of these studies are likely to inform a range of future treatment approaches for Type 1 diabetes. For example, knowing that there is specific targeting of the single/ small clusters of beta cells by the autoimmune attack would support alternative approaches dependent on the stage of disease:
Individuals at risk of Type 1 diabetes: Using early (antigen-specific) immunotherapeutic interventions in young individuals at high risk of developing type 1 diabetes could prevent the loss of single beta cells/ clusters, allowing the formation of larger, better-protected islets, which could delay or prevent the onset of disease. This would also emphasise the need for general population screening programmes to identify people at-risk who would benefit from early interventions.
Individuals with existing Type 1 diabetes: Our data suggest that, particularly in individuals diagnosed early in life, fewer islets overall are generated. It is, therefore, impossible to protect or recover cells that are already destroyed at the onset of the disease. This highlights the importance of future islet replacement strategies.
In individuals diagnosed later in life, despite losing the single/small clusters of beta cells, there is frequently significant residual beta cell mass retained within the larger islets. Strategies to protect this remaining mass will be key. This could involve reducing beta cell stress using agents such as verapamil or lowering BMI.
Endogenous beta cell regeneration therapies: As these therapies are likely to generate single/ small clusters of beta cells, which may be particularly susceptible to autoimmune targeting, combining these with targeted immunotherapies to protect the newly generated cells will be critical.
Importantly, these findings may alter the way we think about the development of type 1 diabetes since it has traditionally been conceived as an illness of beta-cell loss (which it obviously is), but now we might also see it as a disease of beta-cell lack (i.e. not just loss).