Scientists at the University of Chicago have built a lipid nanoparticle system that delivers mRNA directly to insulin-producing beta cells in the pancreas, triggering expression of a protective protein called PD-L1 in both mouse beta cells and human islet cells transplanted into mice. The work, published in Cell Reports Medicine, represents one of several recent advances in pancreas-targeted nanoparticle delivery that could eventually give doctors a tool to stop type 1 diabetes before it destroys the cells that make insulin.
Why the Pancreas Has Been So Hard to Reach
Most lipid nanoparticles injected into the bloodstream end up trapped in the liver or spleen. That sequestration problem has been a core barrier for researchers trying to send mRNA therapies to other organs, according to a Nature commentary by experts reviewing the latest pancreas-targeting work. The liver is efficient at grabbing circulating particles, which is useful for hepatitis vaccines but useless for a disease centered in the pancreas. For type 1 diabetes, the immune system mistakenly attacks beta cells in pancreatic islets, and any therapy meant to protect those cells needs to actually arrive there in meaningful concentrations.
The University of Chicago team, working through the Pritzker School of Molecular Engineering, took a different route. Rather than trying to dodge the liver through the bloodstream, their nanoparticles were designed to deliver PD-L1 mRNA specifically to beta cells. PD-L1 is a checkpoint protein that tells immune cells to stand down. By getting beta cells to produce it on their own surfaces, the therapy essentially engineers the target cells to defend themselves against autoimmune attack, rather than suppressing the entire immune system. In mouse models, the approach reduced destructive immune infiltration in islets and preserved insulin production, an early sign that localized immune modulation could be feasible without the broad immunosuppression that raises infection and cancer risks.
Size-Shifting Particles and Macrophage Shuttles
The UChicago study sits alongside two other recent advances in getting nanoparticles to the pancreas, each using a different physical trick. A team led by Lei and colleagues developed a platform called AH-LNP, described in a Nature study showing that the particles enlarge after absorbing proteins in the body, which allows them to accumulate in the pancreas through capsule filtration and receptor-mediated uptake. That work demonstrated in vivo mRNA delivery to pancreatic tissue, including genome editing using Cas9 mRNA paired with guide RNA, a result that shows the particles can carry complex genetic cargo to a traditionally difficult target while still achieving functional gene modification in situ.
A separate approach, published in Science Advances, skips the bloodstream entirely. Researchers administered nanoparticles through intraperitoneal injection and found that peritoneal macrophages captured the particles and then transferred their mRNA cargo to islet cells via extracellular vesicles. This macrophage-mediated relay produced pancreas-specific protein expression, effectively using the body’s own immune cells as delivery vehicles to shuttle genetic instructions to the islets. The two strategies, capsule filtration and macrophage shuttling, represent fundamentally different engineering philosophies, but both solve the same problem: bypassing the liver to reach the pancreas. No head-to-head comparison of the two methods exists yet, so which approach proves more efficient, scalable, or practical in larger animals remains an open question for future preclinical work.
Tolerogenic Vaccines Cut Diabetes Rates in Mice
Targeting the pancreas is only half the challenge. The other half is teaching the immune system to stop attacking beta cells once the particles arrive. A separate line of research has used mRNA nanoparticles not just as delivery vehicles but as tolerogenic vaccines, particles designed to retrain the immune response rather than simply block it. One such construct, described in a study in the Journal of Controlled Release, encoded tandem glutamic acid decarboxylase 65 (GAD65) epitopes and cholera toxin B subunit (CTB-GADII). Delivered intramuscularly to non-obese diabetic (NOD) mice, the standard animal model for type 1 diabetes, the vaccine reduced disease incidence and delayed onset while improving glucose tolerance and pancreatic morphology. The study also reported effects on autoantibodies including GADA and IAA, along with shifts in cytokine profiles and markers associated with immune tolerance.
A related concept called T1D-tolLNP pushed results even further. This formulation co-encapsulated tolerogenic immunomodulators with mRNA encoding multiple type 1 diabetes autoantigens, aiming to induce a broad, antigen-specific regulatory response. In NOD mice, the platform reduced diabetes incidence from 67% to 13%, with a P value of 0.002, suggesting a robust protective effect. Treated animals showed increased regulatory T cells (Tregs), including PD1-positive Tregs, and blunted inflammatory cytokines, consistent with a shift toward immune tolerance rather than generalized suppression. That 54-percentage-point drop in disease incidence is striking for a single intervention, though the data comes from a conference abstract and has not yet been replicated in larger, longer-duration studies or tested in models that better mirror the genetic and environmental diversity seen in human patients.
Lessons From Autoimmune and Vaccine mRNA Research
These diabetes-focused efforts build on a broader body of work showing that mRNA platforms can be tuned toward immune tolerance instead of stimulation. In a recent preclinical study of autoimmune myocarditis, researchers used lipid nanoparticles encoding cardiac self-antigens and found that the therapy expanded regulatory T cells and reduced tissue damage in mice. The myocarditis model differs from type 1 diabetes, but the principle is similar: carefully designed antigen-encoding mRNA, delivered with the right formulation and dosing schedule, can nudge the immune system away from destructive autoimmunity and toward controlled recognition of self. These findings support the idea that pancreas-directed mRNA vaccines might one day prevent or slow beta-cell loss if given early enough in the disease process.
At the same time, the COVID-19 experience has highlighted both the power and the limits of systemic mRNA vaccination. Work on next-generation SARS-CoV-2 formulations has shown that changing lipid composition, route of administration, and dosing can dramatically alter where particles go and how long they express their payload. One such study reported that an updated mRNA vaccine improved protection in animal models by refining antigen design and delivery parameters. Although that research focuses on infectious disease, it underscores how iterative optimization of nanoparticle chemistry can translate into better performance in vivo. For type 1 diabetes, similar fine-tuning will likely be needed to balance effective delivery to pancreatic islets with acceptable reactogenicity and manufacturing complexity.
What Stands Between Mice and Medicine
The gap between mouse success and human therapy is wide, and these studies sit squarely in it. All the results described here come from rodent models or, in the case of the University of Chicago work, human islet cells transplanted into mice. No clinical trials in humans have been announced for any of these pancreas-targeted mRNA platforms, and regulators will expect extensive toxicology and biodistribution data in larger animals before first-in-human dosing. Long-term safety data for repeated nanoparticle dosing does not yet exist in primates, let alone people, and questions remain about how chronic exposure to synthetic lipids and modified nucleosides might affect organs such as the liver, spleen, and lymph nodes over years. While NOD mice are the gold standard for preclinical type 1 diabetes research, the disease in humans involves a more complex interplay of genetic susceptibility, environmental triggers, and immune regulation, which mouse models only partially capture.
The most compelling aspect of these parallel efforts is the way they converge on a two-part strategy: get genetic instructions into the pancreas efficiently, and then use those instructions to teach the immune system to stand down without compromising its ability to fight infection and cancer. Size-shifting particles, macrophage shuttles, and beta cell-targeted PD-L1 expression all address the delivery problem from different angles, while tolerogenic vaccines and multi-antigen formulations tackle the underlying autoimmunity. Turning these ideas into medicine will require solving manufacturing challenges, demonstrating durable benefit in diverse preclinical models, and carefully staging human trials in people at high risk of type 1 diabetes but not yet insulin dependent. If those hurdles can be cleared, the same toolbox now used to make pandemic vaccines could eventually be repurposed to preserve the body’s own insulin factories before they are lost for good.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
