AAVs Targeting Human Carbonic Anhydrase IV Enhance Gene Delivery to the Brain

Abstract

Clinically approved gene therapies based on natural adeno-associated virus (AAV) serotypes have restricted applications, particularly in the brain, due to their poor targeting, high dose requirements, and resulting safety concerns. Directed evolution of enhanced AAV capsids in mice or non-human primates (NHPs) has resulted in markedly improved performance in those species, but inter-species differences present a serious challenge for translating these vectors into human therapies. Here, we engineer AAVs to target human carbonic anhydrase IV (CA-IV), a recently identified blood-brain barrier (BBB) transcytosis receptor. Among known transcytosis receptors, CA-IV is notable for its relatively specific expression in brain endothelial cells and the potency of AAVs that target it in mice. CA-IVs AAV binding site, and thus the mouse vectors enhanced brain potency, is not conserved across species, so we employed a two-phase engineering strategy to identify AAVs optimized for human CA-IV-dependent gene delivery to the brain. We first used in vitro receptor-based selection of a vast AAV library to exclude capsids that do not bind human CA-IV, followed by in vivo selection in "humanized" mice expressing human CA-IV in brain endothelial cells. Notably, we find that human CA-IV binding capsid variants that were poorly enriched in the pull-down selection outperform strong binders in vivo. The most promising vector, AAV-hCA4-IV77, engages human CA-IV to achieve 100-fold greater brain transduction than AAV9, with robust neuronal and astrocytic coverage throughout multiple brain regions. These results advance our understanding of receptor-targeted capsid design and support the therapeutic potential of human CA-IV-engaging AAVs.