Mapping an Adeno-associated Virus 9-Specific Neutralizing Epitope To Develop Next-Generation Gene Delivery Vectors

• Published in: Journal of virology Vol. 92; no. 20
• Main Authors: Giles, April R, Govindasamy, Lakshmanan, Somanathan, Suryanarayan, Wilson, James M
• Format: Journal Article
• Language: English
• Published: United States American Society for Microbiology 15-10-2018
• Subjects: Animals; Antibodies, Monoclonal - immunology; Antibodies, Monoclonal - isolation & purification; Antibodies, Neutralizing - immunology; Antibodies, Neutralizing - isolation & purification; Capsid - immunology; Cryoelectron Microscopy; Dependovirus - genetics; Dependovirus - immunology; Dependovirus - physiology; DNA Mutational Analysis; Epitope Mapping; Epitopes - immunology; Gene Delivery; Gene Transfer Techniques; Genetic Vectors; Mice, Inbred BALB C; Mice, Inbred C57BL; United States; Viral Tropism; United States; gene therapy; neutralizing antibody; adeno-associated virus; structural biology; vector engineering
• ISSN: 0022-538X; 1098-5514
• DOI: 10.1128/JVI.01011-18

Recent clinical trials have demonstrated the potential of adeno-associated virus (AAV)-based vectors for treating rare diseases. However, significant barriers remain for the translation of these vectors into widely available therapies. In particular, exposure to the AAV capsid can generate an immune response of neutralizing antibodies. One approach to overcome this response is to map the AAV-specific neutralizing epitopes and rationally design an AAV capsid able to evade neutralization. To accomplish this, we isolated a monoclonal antibody against AAV9 following immunization of BALB/c mice and hybridoma screening. This antibody, PAV9.1, is specific for intact AAV9 capsids and has a high neutralizing titer of >1:160,000. We used cryo-electron microscopy to reconstruct PAV9.1 in complex with AAV9. We then mapped its epitope to the 3-fold axis of symmetry on the capsid, specifically to residues 496-NNN-498 and 588-QAQAQT-592. Capsid mutagenesis demonstrated that even a single amino acid substitution within this epitope markedly reduced binding and neutralization by PAV9.1. In addition,
studies showed that mutations in the PAV9.1 epitope conferred a "liver-detargeting" phenotype to the mutant vectors, unlike AAV9, indicating that the residues involved in PAV9.1 interactions are also responsible for AAV9 tropism. However, we observed minimal changes in binding and neutralizing titer when we tested these mutant vectors for evasion of polyclonal sera from mice, macaques, or humans previously exposed to AAV. Taken together, these studies demonstrate the complexity of incorporating mapped neutralizing epitopes and previously identified functional motifs into the design of novel capsids able to evade immune response.
Gene therapy utilizing viral vectors has experienced recent success, culminating in U.S. Food and Drug Administration approval of the first adeno-associated virus vector gene therapy product in the United States: Luxturna for inherited retinal dystrophy. However, application of this approach to other tissues faces significant barriers. One challenge is the immune response to viral infection or vector administration, precluding patients from receiving an initial or readministered dose of vector, respectively. Here, we mapped the epitope of a novel neutralizing antibody generated in response to this viral vector to design a next-generation capsid to evade immune responses. Epitope-based mutations in the capsid interfered with the binding and neutralizing ability of the antibody but not when tested against polyclonal samples from various sources. Our results suggest that targeted mutation of a greater breadth of neutralizing epitopes will be required to evade the repertoire of neutralizing antibodies responsible for blocking viral vector transduction.