Immunological and Biochemical Evaluation of the AAV Capsid to Advance Next-generation Gene Therapy Vector Design

Gene therapy utilizing adeno-associated viral (AAV) vectors has experienced much success in the clinic recently, culminating in the approval of Luxturna, the first AAV gene replacement product approved by the FDA. However, many significant obstacles remain to the translation of AAV vectors into widely available therapeutics, including immune responses to the capsid as well as a general lack of biochemical characterization of the AAV capsid itself. Humoral immunity, generated in response to viral infection or vector delivery, can reduce therapeutic efficacy due to AAV neutralization by neutralizing antibodies (NAbs). To study these interactions and to inform the design of novel capsids able to evade NAbs, we identified the novel and highly neutralizing antibody PAV9.1 from a panel of hybridoma clones and mapped its epitope on AAV9 by cryo-EM. Mutagenesis efforts within this epitope established the minimal changes required to confer evasion of both PAV9.1 binding and neutralization, but this evasion was not found to extend to polyclonal, NAb positive samples from a number of sources, reflecting the complexity of the NAb repertoire against AAV. To enable more complete study of this repertoire in a more therapeutically-relevant fashion, we also established a pipeline for the unbiased cloning of novel ?α-AAV antibodies from single memory B cells isolated from seropositive individuals. Using this approach, we were able to isolate a panel of mAbs found to recognize AAV capsid, validating the use of this method for further study of the polyclonal AAV Ab repertoire. Finally, to come to a better understanding of the biochemical properties of the AAV capsid with the potential to influence critical aspects of vector performance, including cellular transduction and interactions with the immune system, we evaluated AAV by mass spectrometry for the presence of post-translational modifications. We established that there is extensive deamidation of asparagine residues and through genetic deamidation mutagenesis determined that these spontaneous events are a source of potentially undesirable heterogeneity that modulates capsid protein charge, assembly, and in vitro and in vivo transduction. These studies highlight novel findings that contribute to the greater understanding of AAV biology and will inform future efforts to develop next-generation vectors