AAV capsid functions, immune evasion and neuronal targeting in mice and NHP

DESCRIPTION (provided by applicant): Adeno-associated virus (AAV) is the most promising in vivo viral gene delivery vector currently available. However, there still remain
various issues to be resolved, including the high prevalence of pre-existing anti- AAV neutralizing antibodies (NtAbs) in humans, efficacy-limiting host immune responses against viral proteins, and promiscuous viral tropism. In addition, there are species-specific differences in AAV-mediated immune responses and tropism, which often make it difficult to predict clinical outcomes from small animal studies and underscore the importance of nonhuman primate (NHP) studies. As for gene therapy for the central nervous system (CNS) diseases, the inability of AAV to efficiently cross the blood-brain barrier (BBB) poses an additional obstacle to be overcome. Therefore, to ensure a greater success for gene therapy, there is an urgent need to thoroughly address these issues. The ultimate goal of this project is to acquire a large dataset of AAV capsid amino acid sequence-phenotype relationships in cultured cells, mice and NHPs~ utilize the data to understand how the multifaceted AAV capsid phenotypes are determined in each different context~ establish means to overcome the current limitations~ and create NtAb escape AAV vectors that specifically target CNS neurons via the intravenous (IV) route in mice and NHPs. To achieve this goal, we have devised a novel next generation sequencing-based approach, termed AAV Barcode-Seq, which allows us to investigate an array of viral capsid phenotypes of hundreds of different AAV species in a high-throughput manner using only a small number of replicates. We will fully utilize our unique ability to conduct AAV research using this contemporary technology in order to achieve four specific aims. In Aim 1, we will draw high-resolution functional maps of the liver, heart, muscle and CNS-tropic robust serotype capsids (AAV8 and 9) and determine functional roles of each amino acid in manifesting a spectrum of phenotypes including cell surface binding, transduction, tropism and clearance. This analysis will allow us to identify amino acids important for CNS targeting. In Aim 2, we will map epitopes of anti-AAV NtAbs using a novel type of peptide libraries expressed on viral capsids in a native 3-D structure, and establish a means to create NtAb escape mutants. In Aim 3, we will investigate how AAV crosses the BBB using in vitro and in vivo models, and establish a means to create AAV mutants with increased BBB penetrability. In Aim 4, by combining the experimental outcomes of Aims 1-3 and utilizing a novel knowledge-based directed evolution approach, we will create novel NtAb escape AAV capsids that specifically and efficiently target neurons throughout the mouse and NHP brain by IV injection. Successful completion of this proposed project will 1) yield an abundance of insightful data on viral capsid amino acid sequence-host interactions that further our understanding of the AAV capsid biology~ 2) yield novel neuron- specific NtAb escape AAV vectors that can cross the BBB efficiently and readily be used in clinically relevant animal models~ and 3) provide valuable tools
and data resources for the entire gene therapy community.