The researchers’ design choice was deliberate. Rather than using electroporation, a common laboratory method that pushes nucleic acids through cell membranes with electric pulses, they encapsulated synthetic anti-CD19 CAR mRNA in LNPs. Their head-to-head comparison was striking: electroporation delivered mRNA into roughly 78 percent of T cells but slashed viability from 97 percent to 49 percent, whereas LNP transfection preserved nearly all cell viability while producing even stronger protein expression relative to the mRNA dose used. The CAR construct itself was a second-generation design based on the FMC63 antibody fragment, coupled to CD8 hinge and transmembrane domains and the 4-1BB and CD3-zeta signaling modules, with a T2A self-cleaving peptide allowing co-expression of an enhanced green fluorescent protein (EGFP) reporter so that translation could be tracked inside each cell.

One of the study’s most intriguing mechanistic findings concerns why activated T cells take up these nanoparticles so efficiently. When the researchers activated freshly isolated human T cells with anti-CD3 and anti-CD28 antibodies plus interleukin-2, they observed upregulation not only of the activation marker CD25 but also of the low-density lipoprotein receptor (LDLR), the same receptor that LNPs exploit when they enter liver cells. Flow cytometry revealed that the T cell population split into two distinct groups after transfection: larger, highly activated cells expressed far more EGFP and CD25 than their smaller, less activated counterparts. This correlation between cell size, activation state, and transfection efficiency suggests that the very process of preparing T cells for CAR engineering also primes their molecular machinery for LNP uptake, a detail that could inform how future non-viral manufacturing protocols are designed.

Macrophages proved to be even more avid recipients of the mRNA than T cells. Quantitative PCR showed that human primary macrophages internalized about 13,693 parts per million of CAR mRNA relative to total RNA, dwarfing the uptake by other cell types tested, including HeLa, fibroblast, and murine cell lines. Flow cytometry confirmed that over 56 percent of transfected macrophages displayed the CAR protein on their surface, and confocal microscopy verified active translation of the EGFP reporter. The expression persisted for 24 to 96 hours after transfection, with the most robust signal lasting up to 48 hours. Because mRNA never enters the nucleus, this transient expression window is a safety feature as much as a limitation: the engineered receptors fade away rather than persisting indefinitely, which could reduce risks such as uncontrolled cell expansion or off-tumor toxicity.

When the team pitted the two engineered cell types against CD19-positive NALM6 leukemia cells, the differences in their effector mechanisms became clear. CAR-T cells were dramatically potent, killing 86 percent of cancer cells at a 10:1 effector-to-target ratio and still eliminating 82 percent at a mere 1:1 ratio, with detectable effects even at 1:20. CAR-macrophages, by contrast, worked through phagocytosis rather than direct cytotoxicity. Viability staining of cancer cells remaining in the supernatant showed that macrophages were not primarily killing their targets from a distance; they were swallowing them whole. Live-cell imaging captured this process in remarkable detail: after roughly 15 hours of co-culture, CAR-macrophages began actively engulfing leukemia cells, and by 24 hours some individual macrophages had consumed multiple cancer cells simultaneously, while control macrophages ignored cancer cells even in direct contact.

The mRNA transfection itself reshaped macrophage behavior in ways that complicate the picture. Gene expression analysis showed that CAR mRNA-LNP treatment upregulated the pro-inflammatory M1 marker TNF while suppressing the M2 marker CCL22, but also increased TGFB, an anti-inflammatory mediator, indicating a mixed polarization state rather than a clean M1 shift. More surprisingly, CAR transfection sharply reduced macrophage motility, from 12.55 nanometers per second in controls to 5.58 nanometers per second after CAR mRNA treatment, and the cells adopted a less elongated, more activated morphology. Since macrophage migration is essential for infiltrating and remodeling tumors, the authors suggest that co-encapsulating motility-restoring RNAs with CAR mRNA in the same nanoparticle could be a rational next step.

Another layer of the study targeted the so-called myeloid checkpoint, the CD47-SIRPA axis that cancer cells exploit to broadcast a don’t-eat-me signal. The researchers found that CAR mRNA transfection, and even control EGFP mRNA, paradoxically upregulated SIRPA, the receptor on macrophages that recognizes CD47 and suppresses phagocytosis. To counteract this, they delivered SIRPA-targeting siRNA either separately or co-encapsulated with CAR mRNA in a single LNP formulation. The combination effectively neutralized the SIRPA upregulation and potentiated phagocytosis beyond what either component achieved alone. Blocking CD47 on the cancer cells with antibodies also modestly increased uptake, and the team showed that apoptotic cancer cells, which express lower levels of CD47, were readily consumed, confirming that manipulating this checkpoint is a viable strategy for boosting macrophage-mediated cancer clearance.

The centerpiece experiment brought everything together in a triple co-culture of CAR-T cells, CAR-macrophages, and fluorescently labeled NALM6 leukemia cells. Live-cell imaging over 24 hours revealed a striking behavioral shift: in control cultures the three cell types remained largely segregated, but after CAR transfection the culture became a web of cell-cell contacts, with macrophages dragging cancer cells and T cells clustering around both. Quantitatively, only 13.87 percent of cancer cells survived in the dual-CAR cultures after 24 hours, compared with 60.26 percent in controls. Crucially, when the team tested 36 different combinations of CAR-T and CAR-macrophage numbers across three donors and analyzed the results with the Loewe additivity model in SynergyFinder, the average synergy score was 10.013, exceeding the accepted threshold of 10, with a peak score of 21.05. This was true synergy, not simple additivity, likely driven by reciprocal cytokine signaling: activated T cells secrete interferon-gamma, which pushes macrophages toward an M1 state, while activated macrophages release TNF, which promotes tumor cell apoptosis and supports T cell function.

The implications extend beyond the laboratory bench. The only CAR-macrophage therapy to reach clinical testing, CT-0508, an anti-HER2 candidate developed with an adenoviral vector and tested in combination with pembrolizumab, has already shown tumor accumulation and expansion of CD8-positive T cells in early trials, lending real-world plausibility to the synergy the Aachen team observed in vitro. An mRNA-LNP approach could make such therapies faster, cheaper, and free of the risks associated with viral genome integration, and the transient expression profile might allow staggered dosing or combination regimens that are difficult with permanently engineered cells. The authors are careful to note the limitations of their work: co-cultures lack the full complexity of the tumor microenvironment, systemic circulation, complement proteins, and the diverse immune populations that shape responses in patients. Still, their central message is compelling and quantitatively grounded. T cells and macrophages engineered with the same CAR mRNA kill leukemia cells through fundamentally different mechanisms, direct cytotoxicity versus phagocytosis, and when deployed together they achieve something neither can accomplish alone. If future in vivo studies confirm that this synergy permits lower cell doses and a wider therapeutic window, the humble lipid nanoparticle, already famous for its role in vaccines, may become the manufacturing backbone of a new generation of dual-cell cancer immunotherapies.

Subject of Research: CAR mRNA lipid nanoparticle engineering of human T cells and macrophages for synergistic clearance of CD19-positive leukemia cells

Article Title: Distinct effector functions and synergy of CAR mRNA-engineered T cells and macrophages in the clearance of CD19 + leukemia cells

Article References: Liao, L., Lin, C., Kuzmanović, A., Wang, N., Jans, A., Penners, C., Ohl, K., Johnen, S., Geczy, R., Rabel, M., Clarke, S., Balgi, A., Liedtke, C., & Bartneck, M. (2026). Distinct effector functions and synergy of CAR mRNA-engineered T cells and macrophages in the clearance of CD19+ leukemia cells. Journal of Advanced Research, 88, 1119-1136. https://doi.org/10.1016/j.jare.2026.01.049

Image Credits: AI Generated

DOI: 10.1016/j.jare.2026.01.049

Keywords: CAR-T cells, CAR-macrophages, lipid nanoparticles, mRNA therapy, CD19, leukemia, phagocytosis, SIRPA, CD47, immunotherapy, synergy, non-viral gene delivery