Objective
The primary objective of this study is to sustain long-term exposure to multiple beta cell antigens, to halt autoimmunity in T1D. A beta cell antigen is a fragment of a beta cell (e.g., of insulin) that, when presented under inflammatory conditions, triggers harmful T cells (autoreactive T cell) to attack the beta cells. To achieve this, I will evaluate two types of tolerogenic cells (beneficial cells that can inhibit harmful T cells): HSPC-derived APCs (Aim 1) and FRCs (Aim 2). HSPCs are cells found in the bone marrow and are responsible for producing cells of the blood and immune system. FRCs are cells that reside in, and give shape to, the lymph nodes (small bean-shaped structures that are part of the body's immune system where T cells encounter antigens).
HSPCs and FRCs cells will be isolated from the bone marrow or the lymph nodes of IDEAS mice, respectively. The IDEAS mouse is a new type of mouse that we developed in the lab; as the name suggests, expression of multiple beta cell antigens can be induced in specific cell types from IDEAS mice. Thus, HSPCs and FRCs will be induced to express multiple beta cell antigens. In Aim 1, HSPCs from IDEAS mice will be directly injected into NOD mice and will traffic to the bone marrow. From there, they will give rise to various antigen-expressing immune cells that will distribute broadly throughout the body. Unlike treatments for blood diseases such as leukemia, which require high survival rate of injected HSPCs and heavy pretreatment, only a small fraction of HSPC-derived APCs expressing beta cell antigens is needed here to interact with harmful T cells. As a result, minimal pretreatment (or possibly none at all) will be necessary to support the level of HSPC engraftment needed to produce those APCs. HSPCs possess self-renewal capabilities, continuously producing new immune cells that can engage harmful T cell. In Aim 2, FRCs from IDEAS mice will be implanted into NOD mice either in their original form by transplanting whole lymph nodes, or by harvesting them from IDEAS lymph nodes and seeding them onto scaffolds. These scaffolds are made of biomaterials (e.g., gelatin) and are needed to enhance FRC long-term viability and function. In this localized approach, I will test different implantation sites.
For both HSPCs and FRCs, I will first determine their ability to interact with harmful T cells, by monitoring specific T cell activation markers and assessing the efficacy of prolonged T cell activation, which may lead to their exhaustion or conversion into regulatory T cells (beneficial type of T cell that does not destroy beta cells and can instead inhibit other harmful T cells). Secondly, I will evaluate whether these two approaches can prevent diabetes in NOD mice, a model that mirrors various stages of T1D progression based on the age and glycemia of the mouse. Testing will encompass early stages characterized by minimal beta cell destruction to advanced stages marked by active beta cell destruction and high glucose levels. Therefore, I will evaluate various disease stages, to determine whether interactions of antigen-expressing tolerogenic cells with harmful T cells at different stages can impact the efficacy of these therapeutic strategies. In later disease stages, combinations of either HSPCs or FRCs with anti-mouse CD3 (the murine equivalent of Teplizumab) will be explored, to potentially extent treatment applicability to a broader patient population.
Overall, the project aims to provide proof-of-concept of the benefit of persistent antigen presentation in conferring protection in the NOD mouse model of T1D; if effective, these strategies might be translated to patients at risk of or with new-onset T1D.
Background Rationale
Antigen-specific immunotherapies aim to selectively inhibit harmful T cells that destroy pancreatic beta cells, which is crucial for a definitive cure of T1D. However, clinical trials using antigen-specific drugs have not shown significant effects. One reason is that these trials typically used a single antigen, corresponding to a fraction of harmful T cells, while multiple antigens are targeted in the disease process, as indicated by the presence of multiple autoantibodies (e.g., anti-GAD65, anti-insulin, anti-ZnT8) in the blood of patients. We have previously demonstrated the importance of using multiple beta cell antigens for antigen-specific immunotherapy; combining two antigens prevented the disease in the NOD mouse model, while individual antigens were ineffective. Additionally, our and others’ data highlight the necessity of long-term antigen exposure to selectively block harmful T cells. Short-term targeting does not durably diminish their ability to kill beta cells; therefore, frequent administration of antigens appears to be required. An alternative is to use tolerogenic cells that engraft into the human body and continuously present antigens to harmful T cells, thereby exhausting or desensitizing them. This approach ensures that harmful T cells consistently encounter beta cell antigens outside the pancreas and its draining lymph nodes. The location and context of T cells encounter with antigens determines whether they will be triggered to destroy their target (beta cells in T1D) or be desensitized and lose their pathogenic abilities. In T1D, ongoing destruction in the pancreas generate an inflammatory environment (including in draining lymph nodes) that favor the activation of harmful T cells, whereas encountering antigens in other non-inflamed sites is expected to produce opposite results.
In my proposal, antigen presentation to T cells will be either disseminated throughout the body or localized to a specific area away from the pancreas and its draining lymph nodes. I will use HSPCs for the disseminated strategy, and FRCs for the localized strategy; both can be engineered to express beta cell antigens. These cells are ideal candidates for being used as tolerogenic cells; HSPCs can develop into diverse cell types that present antigens without the signals that trigger harmful T cells (because of the local environment), while FRCs naturally express molecules that dampen T cell responses. In humans, both HSPCs and FRCs can be collected from a patient with T1D with a minimally invasive procedure. After engineering, these cells can be infused (HSPCs) or implanted (FRCs) back into the patient in a single treatment. This autologous approach is crucial because the patient’s body will not reject the infused or implanted cells. Continuous antigen presentation by tolerogenic cells might represent a novel strategy to inhibit harmful T cells that cause T1D selectively and persistently.
Description of Project
Type 1 Diabetes (T1D) is a condition in which the body’s immune system progressively destroys insulin-producing beta cells. This destruction is primarily caused by T cells attacking the beta cells. As a result, patients with T1D must inject insulin to survive. While insulin remains the primary treatment, new therapies aimed at blocking T cells are being developed. One such therapy is Teplizumab, which was recently approved by the Food and Drug Administration to delay the onset of T1D. It is important to note that not all T cells are harmful; only a small percentage of them (autoreactive T cells) are responsible for beta cell destruction. However, Teplizumab and similar immunotherapies indiscriminately target both harmful T cells and other, harmless, T cells; this poses a significant health risk due to side effects, especially if reapplication is necessary.
In this project, I propose to selectively target the harmful T cells, through long-term exposure of beta cell antigens in a manner that cause the T cells that are specific to become tolerant rather than triggered, and also possibly acquire the ability to block other harmful T cells. A beta cell antigen is a fragment of a beta cell protein (for example, a fragment of insulin), which, when presented under inflammatory conditions, triggers harmful T cells to attack the beta cells. Through continuous stimulation of the harmful T cells, mostly under non inflammatory conditions, I aim to render them exhausted and ineffective. To achieve this, I will use tolerogenic cells—beneficial cells that can specifically interact with and inhibit harmful T cells—engineered to express selected beta cell antigens.
I will test two types of cells as tolerogenic cells: hematopoietic stem and progenitor cells (HSPCs) and fibroblastic reticular cells (FRCs). Before testing in humans, I will assess the efficacy of HSPCs and FRCs in the Non-Obese Diabetic (NOD) mouse, a mouse model that closely resembles human T1D. For this, I will isolate HSPCs and FRCs from a new type of mouse that we developed in the lab, called IDEAS (Inducible Diabetogenic Endogenous AntigenS). As the name suggests, cells from IDEAS mice can express multiple beta cell antigens. I will induce expression of these antigens in IDEAS-derived HSPCs and FRCs; then, I will inject HSPCs (Aim 1) or implant FRCs (Aim 2) into NOD mice, to evaluate the ability of HSPC-derived antigen presenting cells (APCs) and FRCs to selectively inhibit the harmful T cells and prevent the development of T1D. HSPCs and FRCs that persistently express multiple beta cell antigens could thus serve as part of novel therapies for treating early-stage T1D, during which harmful T cells are actively destroying the beta cells. In the future, they could also be used in combination with beta cell transplants for treating established T1D, to avoid the rejection of the transplanted beta cells.
Anticipated Outcome
The overall goal of the proposed approach is to achieve long-term presentation of multiple antigens relevant to T1D using therapeutic methods. We anticipate that both HSPCs (used in the disseminated approach) and FRCs (employed in the localized approach) will persist long-term in the body while expressing these antigens. We expect a low but sufficient number of injected HSPCs to survive in recipient mice even with no or minimal pretreatment (for example, without irradiation) and persist indefinitely. FRCs are also expected to survive long-term, but this approach involves their surgical implantation with formation of new blood vessels (revascularization) to connect them to the circulation and to the immune system. While they should be able to revascularize and endure long-term, they may survive for a more limited period of time.
Both HSPC-derived APCs and FRCs will interact with harmful T cell (that cause autoimmune diabetes), rendering them ineffective, and eventually desensitizing them through sustained engagement. Under some conditions, these interactions are expected to convert some harmful T cells into regulatory T cells, further inhibiting other harmful T cells. Both strategies—disseminated and localized—will achieve these goals by ensuring that harmful T cells consistently encounter beta cell antigens outside the pancreas and its draining lymph nodes. The effects of different implantation site on FRCs’ engagement with harmful T cells will be assessed by implanting them in various body locations.
We anticipate that these therapeutic strategies will effectively halt the progression of autoimmune diabetes in NOD mice, especially when administered in early disease stages (at 5 weeks old, resembling early T1D stages). Protection is also expected in later stages as well, though these approaches may take time to neutralize the harmful T cells, resulting in diminished effectiveness if many beta cells have already been destroyed. Therefore, the use of anti-CD3, a drug that rapidly target all T cells, in combination with antigen-expressing tolerogenic cells (HSPCs or FRCs), is expected to provide a respite in the autoimmune response while these tolerogenic cells fully engraft, likely resulting in enhanced protection.
HSPCs are expected to offer greater protection in early disease stages due to their crucial role in the thymus (the thymus is the organ where T cells are generated and educated). HSPC-derived APCs can migrate to the thymus and efficiently present antigens to harmful T cells, leading to their direct elimination. However, the contribution of the thymus diminishes with age in both mice and humans, being most active during childhood. Hence, transplanting HSPCs engineered to express multiple disease-relevant antigens may primarily benefit children at risk of T1D, such as those with multiple autoantibodies or siblings of patients with T1D.
These advancements aim to develop therapies that can effectively influence the immune activation and subsequent beta cell destruction seen in T1D in a targeted and sustainable manner.
Relevance to T1D
In 2024, people with T1D still depend on daily insulin injections and strict blood glucose monitoring. Technological advancements like insulin pumps, continuous glucose monitoring and closed-loop systems have improved T1D management. However, these technologies require a deep understanding of the disease, constant oversight and can be lethal if not used correctly. This is particularly challenging for children, as T1D is one of the most common chronic diseases in childhood. Currently, T1D is a lifelong disease with no known cure. The scientific community is getting closer to understanding the causes of T1D. Many genes have been identified as contributing to the disease, with the human leukocyte antigen being particularly significant. Most of these genes involve an incorrect interplay between beta cells (which produce insulin), antigen-presenting cells, and T cells (which attack and destroy the beta cells after recognizing antigens in the presence of additional activating signals). There are several subtypes of T cells, most of which are essential for fighting infections or tumor cells; only a few of these are responsible for destroying beta cells. An ideal therapy of T1D would selectively stop the T cells that cause the disease. Therapies aiming to do this are called antigen-specific immunotherapies. Beta cell antigens can be delivered into the body using various technologies (e.g., DNA, mRNA, peptide, protein), similar to vaccines against infectious diseases. However, increasing evidence indicates that the continuous presentation of beta cell antigens in other (non-inflamed) environments is key to blocking harmful T cells. Without this persistence, the T cells could become activated, upon re-exposure as it happens with vaccines for infectious diseases.
In my proposal, I aim to achieve a continuous presence of beta cell antigens by using two different types of cells, HSPCs and FRCs. Success in stopping the disease using this strategy in the NOD mouse model (the most common and robust model of human T1D) would offer two potential cell therapies for human clinical trials. Additionally, it will pave the way for other strategies aimed at achieving persistent antigen delivery in the human body, such as long-term biomaterials that release antigens. These strategies could be used to prevent early-stage T1D (when beta cells are not yet fully destroyed) and could also be combined with insulin-producing beta cell transplantation to treat established T1D (since new beta cells would still be rejected if harmful T cells are not blocked). Despite recent technological advancements in T1D management, biological cures represent the only solution in the long run. Selectively blocking harmful T cells appears to be an extremely promising and safer approach.