Engineering Adeno-Associated Virus for Receptor-Mediated Gene Delivery

Abstract

Adeno-associated virus (AAV) is a useful gene delivery tool for clinical and basic research applications. Since its discovery nearly 60 years ago, it has become a popular vector because of its small size and low immunogenicity. However, natural tissue tropisms are limited and frequently not useful. Past work modifying the capsid has been limited by structural constraints and a lack of modularity. I aimed to address this problem by creating a modular retargeting system in which the AAV capsid can be rapidly retargeted to any given antigen without newly mutating capsid proteins. I characterized AAV-antibody composites produced by incorporating a small DNA binding domain into one of the AAV capsid proteins and using chemical conjugation to attach the paired DNA sequence to an antibody. I demonstrated that these antibody-AAV conjugates are capable of infecting cells via the antibody-antigen interaction in immortalized and primary cells. Additionally, I created six capsid variants incorporating small targeting scaffolds into each AAV capsid protein. These variants will enable future researchers to select variants from scaffold libraries in vivo, making more specific, better targeting moieties. Using these retargeted vectors, I worked with colleagues to target prostate tumors in vivo and to deliver new payloads with therapeutic potential. A collaborator had recently developed an antibody against a prostate tumor stromal marker, fibroblast activation protein alpha (FAP). By conjugating this antibody to my modified virus, we were able to deliver a fluorescent marker specifically to the tumor. Next, we plan to deliver therapeutic payloads to treat the tumor with minimal off-target effects. Additionally, I assisted in designing and testing a version of prime editor suitable for delivery by AAV. While the efficiency of this tool is still relatively low, it represents an important starting point for adaption of prime editor for use in vivo. In future work, we hope to use antibody-targeted AAV to deliver this and other editing reagents in murine disease models. Finally, to further investigate the landscape of the AAV capsid, we prepared a domain insertion library using a protocol recently developed by the lab. The results from this study will provide valuable information about the plasticity of the AAV capsid that can be used by future researchers to create new variants with added functionality.