Objective
A key challenge in identifying promising therapeutics in Type 1 diabetes is the limitations of our experimental platforms. Conventional 2D cell cultures used for FDA drug screens are poor predictors, as they lack biological complexity and often use immortalized cell lines that exhibit abnormal behavior. Animal models are also limited, as they lack human drug targets and do not recreate human disease. Our proposal seeks to create a more predictive and insightful benchtop screening platform that incorporates human cells important to T1D into a 3D pancreatic niche.
The enabling technology of the project is a three-dimensional (3D) hydrogel that mimics the natural environment of the pancreas. Inside this matrix, stem cell-derived beta cells will be placed. These beta cells can also be genetically modified to reduce their immune reactivity. To the system, cytotoxic T cells (CTLs) that carry T cell receptors tailored to recognize and attack beta cells, like the real ones found in people with T1D, will be added. In this 3D environment, we can observe in real time how CTLs move, latch onto a beta cell, and trigger its death. By providing an authentic, long-term culture that can be imaged continuously, we will generate precise, quantitative data at every step of the attack: how quickly CTLs arrive, how they navigate the matrix, how long they pause on a β-cell, and how many cells die.
Leveraging this platform, we will conduct two aims. One aim will be to screen a panel of therapeutics—some already approved, some experimental—to determine which can prevent CTL recruitment or stop them from killing beta cells without harming the cells themselves. The second aim is to alter beta cells to reduce immune recognition genetically. We will then measure how these changes affect CTL attack. We will also examine how local beta-cell inflammation alters these measurements. These experiments will set the stage for a streamlined, reproducible method to test future T1D drugs and ultimately for improving treatment strategies.
Background Rationale
Type 1 diabetes is an autoimmune condition where the immune system mistakenly hunts and destroys beta cells. The key players in this self-attack are cytotoxic CD8⁺ T cells (CTLs) that recognize beta cell–specific antigens and release lethal molecules. While human pathological samples have documented CTL infiltration in the pancreas, these snapshots miss the dynamics of how CTLs rush in, reach a target, and kill it.
Recent advances in stem cell differentiation and genome editing now allow us to generate batches of uniform beta cells from human pluripotent stem cells. Placing these elegant cells in conventional in vitro 2D plastic dishes reduces their physiological relevance. They also lack the 3D environment, spatial distribution, and biochemical signals that direct real immune traffic. Moreover, they are not durable; beta cell function drops within days, and the artificial geometry distorts cell-cell contact and migration. The transplantation of human cells into animal models increases physiological relevance, but assessment of these transplants is limited and suffers from species differences that weaken their relevance to human patients.
Description of Project
Type 1 diabetes (T1D) is an autoimmune disease in which cytotoxic T cells (CTLs) hunt and destroy pancreatic beta cells. Existing drug screening methods, 2D immortalized cell cultures, or animal studies, poorly predict clinical efficacy. This is likely due to their poor cellular relevance and to a culture system that fails to capture the 3D architecture and environment of the human pancreas. To address these limitations, we created a pancreatic extracellular matrix hydrogel, called the Native Islet Immune Cell Hydrogel Environment (NIICHE). Into this 3D platform, we embed stem cell-derived beta cells (sBCs) and CTLs engineered to recognize human beta cell antigens. The hydrogel is translucent, stable, and rich in native collagen and cytokine-binding proteins, allowing both cell types to move and interact in a more physiologically relevant manner.
Because the matrix preserves spatial organization, we can use confocal microscopy to film the same field of view for days, creating a real-time movie of immune cells recruiting, engaging, and killing their beta cell target. From these movies, hundreds of cellular trajectories are automatically recorded and analyzed by custom software that reduces the data to quantitative metrics, including CTL recruitment, accumulation, migratory phenotype, and β-cell death rates. These readouts are non-invasive and high-throughput, while providing a level of spatial and temporal resolution never before achieved in vitro.
The goals of this proposal are to leverage this novel NIICHE system as a testbed to screen pharmaceutical agents, as well as investigate the impact of genomic editing and the local environment on immune cell recruitment and killing. Aim 1 goals will evaluate NIICHE’s utility as a drug screening platform, with a focus on future FDA translation. Aim 2 will assess the impact of engineering beta cells to be less immunogenic on immune cell recruitment and killing, while also examining the effects of local inflammation. By capturing how genetic changes or environmental cues alter immune-beta cell interactions, the NIICHE will inform the rational design of new therapies aimed at making beta cells “invisible” to immune cell attack.
This work is impactful as it seeks to establish a novel alternative method (NAM) with high human relevance, scalability, high content, and low cost. Readouts are compatible with FDA’s reduced animal model goals, positioning NIICHE for eventual acceptance as an approved preclinical screening platform.
In short, NIICHE turns the pancreas into a living, transparent laboratory where the interactions between CTLs and beta cells can be observed, modulated, and measured in a way that mimics human physiology. The insights and drug-response data produced will pave the way for therapies that protect beta cells and dampen immune responses, bringing us closer to a durable cure for type 1 diabetes.
Anticipated Outcome
Within a year, we seek to deliver a validated, human-based 3D platform—NIICHE (Native Islet-Immune Cell Hydrogel Environment). Specifically: 1) A comprehensive, quantitative data set detailing CTL recruitment, migration classification, and killing for a range of pharmacological agents and genetic β-cell modifications. Each scenario will generate hundreds of cell tracks per field of view, providing robust statistical confidence that can be translated into preclinical decision-making. 2) A drug response portfolio for 5-10 clinically relevant molecules that will confirm whether suppression of CTL activity or protection of beta cells is achievable in a physiologic context. Cross-validation using conventional assays will provide further context for the benefits of this more automated and temporal method for screening agents. This portfolio will aid in a future FDA ISTAND submission. 3) Mechanistic insights into β-cell immunogenicity. By comparing HLAII knockout, PD L1 overexpressing, or CD155 displaying β cells with and without inflammatory stimulation, we will quantify the extent to which antigen presentation and inflammatory signaling alter CTL recruitment and killing. These results will clarify whether β-cell intrinsic alterations can dampen autoimmunity without sacrificing insulin secretion. Altogether, these outcomes will deliver a robust, scalable tool for the T1D community and lay the foundation for a novel alternative model (NAM) for drug screening and beta cell protection studies.
Relevance to T1D
Type 1 diabetes arises when the immune system mounts a targeted attack on the body’s own insulin-producing β cells. Although the involvement of CTLs is established, the details of how these immune cells find, bind to, and eliminate β-cells remain poorly understood. The lack of physiologically relevant human models also hampers drug development.
NIICHE seeks to address this unmet need by combining two key advances: 1) human-sourced T1D relevance cells, including stem cell-derived beta and T1D-antigen-specific CTLs; and 2) a 3D hydrogel that mimics pancreatic extracellular matrix, allowing immune cells to move, sense, and act as they would in real tissue. In effect, NIICHE becomes a “mini pancreas” providing temporal insight into T1D. This system should support the efficient identification of promising therapeutics or beta-cell-modifying approaches that would have a substantial impact on T1D.