bioRxiv Science⌕ Search
Home/Archive/bioRxiv

bioRxiv · 10.64898/2025.12.25.696171

Developing High-Yield and Safe Therapeutic EVs by Ablating Tissue Factor-Mediated Toxicity

Luo, Q.·Li, X.·Yang, M.·Xu, S.·Jiang, T.·Xia, J.·Ding, J.·Qin, X.·Zhao, T.
View original source ↗Export metadata ↓

Abstract

The clinical translation of extracellular vesicles (EVs) as next-generation drug delivery vehicles is currently stalled by two interrelated challenges: the difficulty of manufacturing EVs at an industrial scale ("the yield problem") and the risks of toxicity associated with systemic administration of high doses ("the safety ceiling"). In this comprehensive study, we present a two-pronged genetic engineering strategy to overcome these limitations. First, we discovered that overexpression of glycosylphosphatidylinositol (GPI)-anchored proteins, such as CD55 and CD59, significantly enhances EV production in HEK293T and Expi293F cells. By engineering a truncated CD55 variant (TR3) that retains the GPI-anchor but lacks functional domains, we developed the ExoBoost cell line, which boosts EVs yield by approximately 50-fold without altering vesicle morphology. Second, we observed acute toxicity in mice following intravenous administration of high-dose ExoBoost EVs, with symptoms resembling venous thromboembolism. Through comparative analysis of EVs derived from other cell types, including keratinocyte stem cells (KSCs) and mesenchymal stem cells (MSCs), we identified Tissue Factor (TF/F3) as the critical causative agent of this lethal response. Moreover, we completely abolished this toxicity by generating a F3-knockout cell line (ExoBoost2.0) using CRISPR/Cas9, thereby raising the safe dose to as high as 1.5E12 particles per mouse, as quantified by nano flow-cytometry (nFCM). Finally, using a novel CD46-nanoluciferase reporter system, we demonstrated that high-dose administration of these safe EVs saturates hepatic and splenic clearance pathways, leading to a dramatic increase of systemic EVs accumulation, particularly in hard-to-target tissues such as the brain ([~]10,000-fold) and muscle ([~]1,000-fold). Additionally, repeated high-dose administration of ExoBoost2.0 EVs did not upregulate inflammatory cytokines or production of IgG and IgM. In summary, these findings establish a scalable, safe, and highly biocompatible EVs platform, potentially revolutionizing the drug delivery system in clinical application.

Explore related subjects

bioengineeringPreprint

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Luo, Q., Li, X., Yang, M., Xu, S., Jiang, T., Xia, J., Ding, J., Qin, X., Zhao, T.. 2025-12-25. Developing High-Yield and Safe Therapeutic EVs by Ablating Tissue Factor-Mediated Toxicity. https://doi.org/10.64898/2025.12.25.696171

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

KEEP EXPLORING

Related preprints

bioRxiv2026

RNASeek: A Cross-Phyla Generative Foundation Model for Multipurpose RNA Modeling and Reinforcement Learning-Based Design

RNA plays central roles in regulating information flow and provides a versatile substrate for engineering biological functions. While large language models (LLMs) have transformed natural language processing and protein design, a general framework connecting RNA foundation models to functional sequence design remains limited. Here, we present RNASeek, a 1.6-billion-parameter generative foundation model built on a DeepSeek architecture and trained on a cross-phyla transcriptomic corpus for RNA sequence representation and generation. Natural-language tokens enable flexible conditional prediction and sequence design using a unified backbone. RNASeek captures species-specific transcript features and intron-exon boundaries in a zero-shot setting. We then fine-tune RNASeek to predict ribozyme self-cleavage activity and viral mRNA stability, revealing interpretable sequence features associated with function, including ribozyme loop flexibility and stem stability, as well as AU-rich motifs associated with mRNA stability. We use these functional predictors as reward models and apply Group Relative Policy Optimization (GRPO) to update the generation policy of RNASeek toward sequences with desired properties. GRPO-guided generation produces faster-cleaving ribozymes and stability-enhancing 3' UTRs while satisfying user-specified IUPAC constraints. Experimentally validated RNASeek-generated ribozymes achieve wild-type levels of activity, while RNASeek-generated 3' UTR sequences exceed the performance of the training data and benchmarked AI-generated 3' UTRs. Together, RNASeek establishes a unified pretrain-predict-optimize framework that connects learned RNA function to controllable de novo sequence design and provides a general strategy for engineering regulatory RNAs with desired properties.

bioengineering↗
bioRxiv2026

Joint Vector Flow Mapping and Segmentation: Ill-Posedness,Differentiable Bayesian Inference, and Synthetic Vortex-FlowBenchmarks

Vector flow mapping (VFM) reconstructs left-ventricular (LV) blood velocity from color-Doppler echocardiography by combining the measured beamwise component with physical and regularizing constraints. Analysis of the discrete VFM formulation shows that the inverse problem is intrinsically ill posed: the occurrence of singular modes can be predicted from the geometry of the segmented blood-pool domain, the imposed boundary conditions, and the degree of smoothing. These modes can propagate uncertainty along entire transverse bands of the reconstructed velocity field, yet conventional VFM neither quantifies this uncertainty nor allows for correcting the blood-pool segmentation. We introduce Bayesian VFM (B--VFM), a hierarchical framework that jointly infers radial and transverse velocities, a probabilistic blood-pool mask, their spatially resolved uncertainties, and hyperparameters weighting Doppler and segmentation fidelity, mass conservation, boundary conditions, and smoothness. The discretized posterior admits a closed-form gradient and exact Hessian, enabling computationally efficient, gradient-based MAP estimation, sampling, and direct analysis of ill-posed modes. Posterior inference combines Gibbs sampling of conjugate Gamma-distributed hyperparameters with conditional maximum-a-posteriori estimation and a Laplace approximation for the high-dimensional velocity and mask fields. To accommodate systematic departures from planar mass conservation, B-VFM can learn the covariance of the planar divergence residual from an ensemble of flows and incorporate it as a structured model-discrepancy prior. Independent chains converged reproducibly, while covariance priors learned from flow ensembles illustrated how model discrepancies can be incorporated into the inference. B--VFM was evaluated using Lamb-Chaplygin dipoles under ideal conditions and with Doppler corruption, Doppler voids, and segmentation defects, and using the Hicks-Moffatt family of spherical vortices to assess violations of planar mass conservation. The method produced smooth reconstructions, localized uncertainty near unreliable measurements and regions of model inconsistency, and used flow information to correct segmentation errors. Within the tested vortex family, the data-informed planar divergence prior reduced velocity bias and mask distortion. B--VFM thus provides an uncertainty-aware reconstruction method and a flexible foundation for future VFM formulations incorporating additional priors, observations, and physical models. Future work will evaluate the method using clinical data and more complex three-dimensional benchmark flows.

bioengineering↗
bioRxiv2026

Computational design of a versatile, zero-radius proximity labeling enzyme

The ability to map protein interactomes and organelle proteomes is foundational for achieving a molecular understanding of living cells. Proximity labeling (PL) provides a powerful strategy for this, but existing enzymes and photocatalysts are limited by their spatial resolution, reliance on biotin, and/or in vivo compatibility. Here we report FlexID, an engineered promiscuous ligase that catalyzes the rapid attachment of diverse small-molecule probes to proximal endogenous proteins. Critically, FlexID operates through a zero-radius, direct-contact mechanism, offering superior spatial precision compared to existing PL tools. We engineered FlexID by combining the strengths of sequence- and structure-trained computational models to enhance its catalytic activity and structural stability. Biophysical analysis revealed that specific conformational changes in FlexID improve its ability to recognize diverse target proteins while simultaneously preventing the premature release of the reactive intermediate. We demonstrate FlexID's versatility through in vivo proximity labeling, comprehensive organelle proteome mapping, and a high-throughput, fluorescence-based screen for molecular glues. Our work shows that computational methods can be harnessed to create mechanistically distinct PL enzymes and establishes FlexID as a flexible, high-resolution tool for mapping protein interactions and proteomes in living cells.

bioengineering↗