Post by Lei Zhiyong

Last week I shared our NEO-TOP-EVs preprint — three biogenesis principles for engineering extracellular vesicles (https://lnkd.in/eifE_9Tq). The next question is: could we fix something bigger, like AAV? Today the answer is live on bioRxiv as our follow-up preprint, NEO-AAV. Two headaches have haunted AAV for years. Pre-existing antibodies neutralize the virus and shut down redosing. And primary human T cells, the cells you most want for immunotherapy, barely let AAV in without CD3/CD28 bead activation first. So we built a single fusion protein (PH–ALG2–PKD12) that actively wraps assembled AAV capsids inside extracellular vesicles as they're secreted. And it worked. Cryo-EM shows capsids packed neatly inside the vesicles. Nano-flow cytometry: about 25% AAV-positive particles versus 7% for the passive control. Throw a neutralizing antibody at it and naked AAV6 collapses while NEO-AAV6 keeps right on transducing. Then came the part that surprised us. We added a CD7/CD3/CD28 tri-chimera on the surface, so the particle delivers the activation signal and the cargo at the same time. In primary human PBMCs, around 38% of CD7+ T cells got transduced in a single step. No beads, no pre-activation. As a proof-of-concept we generated anti-CD19 CAR-T cells this way, and they killed Nalm6 with antigen specificity while sparing CD19-negative K562. We're looking forward to sharing the in vivo performance of NEO-AAV soon. Credits go to lead authors Songpu Xie and Qiangbing Yang, who drove this work, and to the whole team(Lisa Vork Jansen, Kaiyong Qu, Bing Yao, Saskia de Jager, Pieter Vader, Geng Yang, Gerard Boink, @Christian Snijders Blok, Pieter Doevendans, 肖俊杰, Raymond Schiffelers, Joost Sluijter across UMC Utrecht and our collaborators. Thanks also to the funders who made it possible! Preprint: https://lnkd.in/g86-uEky #AAV #GeneTherapy #ExtracellularVesicles #CART #Immunotherapy #bioRxiv