Strategies for improving adeno-associated viral infection of airway epithelial cells

Cystic fibrosis (CF) is a lethal autosomal recessive genetic disease caused by mutations in a single gene, the cystic fibrosis transmembrane conductance regulator (CFTR). CF affects multiple organ systems, but the major cause of morbidity and mortality is due to disease in the lungs. In theory, using gene therapy to deliver a correct copy of CFTR to the cells of the airway epithelium could result in a lifelong cure. Adeno-associated virus (AAV) is a single stranded DNA virus that is a promising candidate vector for gene therapy of multiple diseases, and numerous clinical trials are currently underway. Despite recent clinical successes, several challenges still impede wider application of AAV gene therapy to numerous diseases, including CF, as AAV-mediated gene transfer to the airways remains below the level needed for therapeutic efficacy for CF. We hypothesized that the low transduction efficiency of AAV in the airways could be overcome by using directed evolution of AAV in organotypic human and pig airway models, and in vivo in the lungs of pigs to select novel AAV capsid variants with improved infectious properties. We discovered a highly infectious, novel AAV that was a chimera of AAV2 and AAV5 with one point mutation (A581T) which we called AAV2.5T. We found that AAV2.5T mediated gene transfer significantly better than its parental serotypes, and corrected the chloride transport defect in CF human airway epithelial cultures. We determined that AAV2.5T developed increased binding to the apical surface of human airway epithelial cells, and that it has evolved to utilize specific 2,3N-linked sialic acid residues on the cell surface that mediate rapid internalization and subsequent infection. Thus, sialic acid serves as not just an attachment factor but is also required for AAV2.5T internalization, possibly representing an important rate-limiting step for other viruses that use sialic acids. Additionally, we utilized directed evolution in vivo in the lungs of pigs to select a novel AAV capsid that is identical to AAV2 except for five point mutations, which we called AAV2H22. We found that AAV2H22 mediated gene transfer to pig airway epithelial cultures significantly better than AAV2, and that it had evolved altered receptor binding. We also found that directed evolution in vitro in human and pig airway epithelial cultures results in the selection of distinct viruses for the two species, and that maintaining different selection stringencies results in the recovery of different AAV variants. Finally, we utilized Hoechst 33342, a DNA binding compound which was previously found to increase AAV transduction in cell lines, to increase AAV-mediated gene expression in primary human airway epithelia. We determined that the mechanism of this effect was due to activation of the CMV promoter. The findings from this research have significant implications for our understanding of AAV biology and for pulmonary gene therapy.