Adeno-associated virus (AAV) is a promising gene therapy vector that is currently approved for use in humans. However, areas of improvement have been identified that can expand the success and applications of AAV. This thesis addresses several of these areas, with a particular focus on the role of the virus capsid. To provide further insight on the role of the virus capsid in the infection pathway, we characterized a serine/threonine rich region in the N-terminal region of VP1 and VP2 capsid proteins and found this region to play an essential role in virus transduction in multiple AAV serotypes. To improve capsid engineering for targeted vector delivery, we discovered a novel design criterion for the protease-activatable provector platform which significantly reduced off-target transduction without affecting on-target activity. To contribute to a solution for evading or escaping from the immune system, we studied the immune response to the AAV capsid in a panel of immune-deficient mice, leading to the identification of B cell deficiency as being sufficient for AAV re-administration. Furthermore, we showed that several components of the innate immune system were necessary for generating a robust antibody response against the AAV capsid. Altogether, this work contributes to our understanding of AAV behavior and the interactions of the viral capsid with its target cells and the host immune system. This information will advance the development of novel AAVs and AAV-based therapies that have improved efficacy, targeting, and longevity for the treatment of a broad range of diseases.