Search bioRxiv⌕ Search
Home/Archive/bioRxiv

bioRxiv · 10.1101/2025.06.09.658602

High-throughput in vivo screening using barcoded mRNA identifies lipid nanoparticles with extrahepatic tropism for cancer immunotherapy

Hamilton, A. G.·Thatte, A. S.·Xu, J.·Luo, Z.·Safford, H. C.·Swingle, K. L.·Muscat-Rivera, J.·Kegel, M.·Han, X.·Joseph, R. A.·Murray, A. M.·Geisler, H. C.·Whitaker, R. C.·Xue, L.·Spektor, R.·Melamed, J. R.·Weissman, D.·Mitchell, M. J.
View original source ↗Export metadata ↓

Abstract

Interest continues to grow in the use of mRNA vaccines for cancer immunotherapy. While effective for immunization against infectious diseases, current clinical lipid nanoparticle (LNP) formulations used for mRNA delivery suffer from off-target accumulation, poor immune transfection, and reactogenicity, limiting their application to cancer immunotherapy. Development of new mRNA LNPs is severely bottlenecked by the LNP discovery process, which is historically low-throughput due to reliance on low-plexity measurements. Here, we develop a next-generation high-throughput in vivo mRNA LNP screening platform based on barcoded mRNA (b-mRNA). Using this b-mRNA screening platform to simultaneously evaluate 122 LNPs, we identify novel LNP formulations capable of potent hepatic and extrahepatic transfection. We employ novel biochemical characterization techniques to analyze nanoparticle protein corona formation with single-particle resolution and gain insight into the influence of protein adsorption on hepatic and splenic transfection. We evaluate a lead LNP candidate for therapeutic cancer vaccination in a syngeneic mouse model of melanoma and demonstrate a significant reduction in tumor burden and increase in survival compared to a clinical mRNA LNP formulation. Together, our results demonstrate the value of advanced LNP screening and characterization techniques for the development of next-generation mRNA therapeutics and vaccines.

Source connections

Explore related subjects

bioengineeringPreprint

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Hamilton, A. G., Thatte, A. S., Xu, J., Luo, Z., Safford, H. C., Swingle, K. L., Muscat-Rivera, J., Kegel, M., Han, X., Joseph, R. A., Murray, A. M., Geisler, H. C., Whitaker, R. C., Xue, L., Spektor, R., Melamed, J. R., Weissman, D., Mitchell, M. J.. 2025-06-09. High-throughput in vivo screening using barcoded mRNA identifies lipid nanoparticles with extrahepatic tropism for cancer immunotherapy. https://doi.org/10.1101/2025.06.09.658602

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

bioRxiv2026

Ferrybots: Programmable Hydrogel Microbots for Rapid Magnetic Navigation and Modular Function

Untethered microrobots could enable localized delivery, sensing, and intervention in confined biological and engineered environments. Magnetic actuation is especially attractive for such systems because it provides remote, fuel-free control, but practical magnetic microrobots remain difficult to manufacture and functionalize. High magnetic loading can interfere with polymerization and structural integrity, while integrating application-specific payloads directly into the robot body often requires the entire material formulation to be redesigned. Here, we introduce Ferrybots, programmable hydrogel microrobots that physically separate magnetic navigation from functional cargo. An automated droplet-on-demand microfluidic flow-lithography process confines high-concentration ferrofluid cores within photopolymerized poly (ethylene glycol) diacrylate (PEGDA) structures, enabling magnetic loading to be programmed through core number and spacing without uniformly dispersing magnetic particles throughout the hydrogel. The process achieved a minimum fabrication cycle time of 15 s. Ferrybots exhibited magnetic-dose-dependent translation and rotation, branched-channel steering, out-of-plane obstacle traversal, and rapid translation, with two-core robots following an external magnet at speeds up to the 9000 mm/min limit of the actuation system. Differences in magnetic loading also enabled sequential selective navigation of distinct Ferrybot populations. Using PEGDA 575 as a common interface material, independently formulated hydrogel cargo modules could subsequently be fused to preformed Ferrybots while preserving navigation. Separately, cargo modules enabled composition-dependent molecular release and oxygen-responsive sensing through 1-mm-thick skin-mimicking optical phantoms representing a broad range of skin types. By decoupling locomotion from function, Ferrybots provide a common magnetic vehicle that can be combined with interchangeable delivery, sensing, or responsive-material modules, offering a general strategy for rapidly adapting microrobots to different applications.

bioengineering↗
bioRxiv2026

Circular-Arc Modal Migration for Ultrasound Imaging--Part I: Finite-Arc Spectral Theory and Stolt-Pullback Factorization

Circular-arc ultrasound imaging requires a spectral construction that accommodates radius-dependent propagation and the spatially varying phase of offset-source transmission. This first of two articles develops Circular-Arc Modal Migration (CAMM), a direct Stolt-type imaging framework in angular modal coordinates. Starting from a first-Born acoustic model, angular decomposition and local radial Stolt mappings convert channel spectra into fields for coherent image synthesis. The offset transmit phase is represented by a phase-screen factorization through reusable spectral factors and spatially weighted fields. A scalar-pullback approximation simplifies this correction to reading one reference image at a geometrically fitted radial coordinate. Both realizations share the same modal construction. Error controls address local radial synthesis, transmit-path expansion, scalar approximation, and phase-screen truncation, with direct integration verifying the separated numerical evaluation. Same-RF point-target comparisons show that scalar pullback closely preserves the principal peak positions and widths of phase-screen imaging in the compact semicircular configuration, while phase screens better control surrounding responses in the shallow convex examples. The retained modal spectrum, frequency band, image radius, and source displacement guide this accuracy--computation choice. CAMM thereby extends direct spectral image formation to circular arcs and supplies the theoretical and numerical basis for the image-quality and performance studies in Part~II.

bioengineering↗
bioRxiv2026

Identification and isolation of migratory subpopulations of cancer cells using microfluidics with fluid walls

Microfluidic assays have great potential for investigating cancer cell migration, a key process underlying metastasis which remains a major clinical challenge. However, adoption by biologists is limited due to complex fabrication and operation methods combined with difficulty in retrieving cells of interest. We jet-print dumbbell-shaped microfluidic environments with fluid walls in standard Petri dishes in seconds using standard growth media and immiscible oil. Prostate cancer cells and chemoattractant (bovine serum) are introduced into each end using local transient differences in Laplace pressure to enable spatial confinement. Automated live-cell imaging monitors chemotaxis towards serum using standard microscopes, as the confining fluid walls are optically transparent. Distinct subpopulations of migratory cells exhibit different morphologies and motilities, assessed using a custom machine-learning segmentation model. Then, we leverage the reconfigurability of fluid walls to isolate and retrieve rapidly-migrating single cells. This demonstrates that fluid-walled microfluidics is a versatile and accessible approach for investigating cancer cell migration at single-cell resolution, offering new opportunities to examine cellular heterogeneity in metastasis.

bioengineering↗