mRNA delivery system activates T cells for in-body CAR T generation

Scientists have developed an mRNA delivery system that can generate functional chimeric antigen receptor (CAR) T cells inside the body without requiring antibody targeting or external immune stimulation. The polymer–lipid nanoparticles, described in Nature Materials, preferentially deliver genetic cargo to T cells in the spleen after systemic administration and simultaneously activate those cells. In animal models, the platform enabled the in situ production of CAR T cells directed against fibroblast activation protein, or FAP, a marker associated with pathological fibroblasts in cancer and fibrotic disease. The engineered cells eliminated these disease-associated fibroblasts while producing minimal detectable off-target effects.

CAR T cell therapy has transformed the treatment of several blood cancers, but its conventional manufacturing process remains complex, expensive and difficult to scale. Patient T cells must generally be collected, genetically modified outside the body, expanded under controlled conditions and reinfused. This individualized workflow can take weeks and may be inaccessible to patients whose immune systems are already severely compromised. In vivo engineering could bypass much of this infrastructure by delivering nucleic acids directly to immune cells inside the patient. However, T cells are among the most difficult cell types to transfect with lipid nanoparticles, the delivery systems most widely explored for mRNA therapeutics.

The challenge arises from both biological and physical barriers. T cells circulate through blood and lymphoid tissues, are present at relatively low abundance in many organs and possess cellular membranes that are not easily penetrated by conventional nanoparticles. Many experimental systems have therefore relied on antibodies or other ligands attached to the nanoparticle surface to recognize T cell receptors. Such targeting molecules can improve selectivity, but they add manufacturing complexity, increase the possibility of immune reactions and may limit the adaptability of the platform to different therapeutic targets. The new system is designed to achieve T cell delivery through the intrinsic properties of the carrier rather than through an external targeting ligand.

The delivery vehicle, called an ERTLNP, is based on oligoethylenimine-derived lipids modified with p-toluenesulfonyl arginine, abbreviated RT. Oligoethylenimine provides a polymeric component capable of interacting with negatively charged mRNA, while the lipid architecture helps the resulting particles enter cells and release their cargo. The RT modification incorporates an arginine-containing chemical group that changes the particle’s interactions with biological membranes and immune cells. According to the study, this formulation preferentially accumulated in the spleen following intravenous administration, a significant finding because the spleen is a major site of immune-cell organization and contains substantial populations of T cells.

The nanoparticles carried messenger RNA rather than DNA or a viral vector. Once released into the cytoplasm, mRNA can be translated directly into protein without needing to enter the nucleus or integrate into the genome. This makes the approach inherently transient: the engineered receptor is produced for a limited period as the mRNA is naturally degraded. Transient expression may be advantageous when the therapeutic objective is to eliminate a temporary or disease-associated cell population, because it can reduce the risks associated with permanent genetic modification. At the same time, effective treatment requires enough delivery, translation and cellular expansion to produce a meaningful therapeutic response.

A central feature of the ERTLNP platform is that it did more than transport mRNA. The particles intrinsically activated T cells in the absence of an added stimulatory antibody or a separate immune-activating reagent. This activation produced robust mRNA expression and encouraged T cell proliferation, allowing the same carrier to combine delivery and cellular conditioning. The researchers linked this response to engagement of the phosphoinositide 3-kinase, protein kinase B and mechanistic target of rapamycin pathway, commonly known as the PI3K/AKT/mTOR axis. This signalling network regulates cell growth, nutrient use, protein synthesis and energy metabolism, and is particularly important when T cells transition from a resting state into active proliferation.

The metabolic component may help explain why the system generated more than a short-lived burst of receptor expression. Activated T cells must reprogramme their metabolism to support rapid division and the production of effector molecules. Through PI3K/AKT/mTOR signalling, ERTLNP treatment was reported to promote this metabolic transition while restraining features associated with T cell exhaustion. Exhaustion is a dysfunctional state that can develop during prolonged antigen exposure and is characterized by reduced proliferation and impaired immune activity. By supporting expansion while limiting exhaustion-related changes, the nanoparticles may create a cellular environment in which mRNA-engineered T cells remain therapeutically effective for longer than would be expected from delivery alone.

To test the system’s therapeutic potential, the researchers loaded ERTLNPs with mRNA encoding a CAR against fibroblast activation protein. FAP is expressed by activated fibroblasts that contribute to the structure and progression of several tumors and fibrotic disorders. These fibroblasts can help create a protective tissue environment around cancer cells, while in fibrosis they participate in the excessive deposition and remodeling of extracellular matrix. Rather than directing CAR T cells against the malignant cells themselves, the strategy targets the supportive or pathological stromal cells that help sustain disease. Following systemic administration, the nanoparticles generated FAP-directed CAR T cells inside the animals, and those cells effectively removed pathological fibroblasts in models of cancer and fibrosis.

The reported activity is notable because it was achieved without the antibody conjugation commonly used to direct nanoparticles to T cells. Ligand-free delivery could simplify production and make it easier to adapt the platform to different CAR sequences or other therapeutic mRNAs. The study also reported minimal off-target effects, suggesting that the combination of tissue distribution, carrier chemistry and transient mRNA expression may provide a degree of biological control. Nevertheless, the findings remain preclinical. Before clinical translation, researchers will need to determine how the particles behave in humans, how long CAR expression and T cell activation persist, whether repeated dosing is safe and how strongly the formulation affects other immune or non-immune cell populations.

The platform also raises broader questions about the relationship between delivery and immune-cell programming. Traditional nanoparticle design often treats cell activation as an undesirable inflammatory side effect, whereas this work incorporates activation into the therapeutic mechanism. That strategy could be useful when the target cell requires a coordinated sequence of uptake, gene expression, metabolic adaptation and proliferation. However, precisely controlling that sequence will be essential, since excessive PI3K/AKT/mTOR activity or unintended immune activation could produce toxicity. The balance between sufficient T cell expansion and systemic inflammation will likely be a major consideration in future studies.

By combining ligand-free mRNA delivery with intrinsic T cell stimulation, the ERTLNP system offers a potential alternative to the conventional ex vivo CAR T manufacturing pipeline. Its use against FAP-positive fibroblasts also illustrates how in vivo CAR T generation might be applied beyond blood cancers, including solid tumors and fibrotic diseases in which pathological stromal cells are important drivers. If the delivery and safety profile can be reproduced in human studies, this type of metabolically reprogramming nanoparticle could provide a more scalable way to produce therapeutic immune cells directly inside the body. The work positions nanoparticle chemistry not merely as a method for transporting genetic instructions, but as an active component of cell engineering and immune therapy.