Back

A minimally modified AAV9 variant (AAV9fyhr) for enhanced CNS gene delivery by intra-CSF administration.

Garza, I. T.; Ramesh, V.; Hu, Y.; Rioux, M.; Takar, M.; Higgs, H.; Lykken, E. A.; Adams, R.; Kunach, P.; Kieffer, D.; Trygg, C. B.; Fisher-Perkins, J. M.; Kumar, A.; Joachimiak, L. A.; Bailey, R. M.; Xing, C.; Braun, S. E. A.; Gray, S. J.

2026-10-09 bioengineering
10.64898/2026.10.07.756773 bioRxiv
Show abstract

Adeno-associated virus (AAV) mediated gene therapies continue to show promise for the treatment of genetic diseases. However, delivery and safety still represent significant prohibitive barriers restricting translation of these delivery modalities to humans. We describe a novel AAV9 variant, AAV9fyhr or FYHR, identified from screening a controlled randomization point mutation AAV9 library in mice and nonhuman primates followed by directed evolution of in vivo CNS-derived libraries screened in human iPSC differentiated cortical neurons. Capsid variants of interest were identified by coupling long read sequencing and computational structure modeling. Structure-function analysis of our lead candidate reveal that two targeted amino acid changes are sufficient to result in beneficial pleiotropic effects across viral stability, cell surface attachment, and gene delivery. We developed a novel multiplexed barcoding reporter system, enabling a time and cost-effective option to gather spatially resolved single cell capsid validation data in mice and nonhuman primates. When administered intrathecally, AAV9fyhr exhibits 2-to-6-fold higher gene delivery across several tissues in the CNS while minimally altering the tropism to peripheral tissues in comparison to AAV9. These findings demonstrate the potential of whole capsid engineering for the rational development of translational next generation AAV capsids.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.