(597) AAV capsid modification to improve tropism for airway epithelial progenitor cells

Background: A major hurdle for gene therapy in the lung is delivery. Our goal is to achieve life-long correction from a single dose of aerosolized viral vector. As such, efficient delivery of a therapeutic gene or gene editing machinery to airway progenitor cells is critical. In this study, we use adeno-associated virus (AAV)-based vectors as vehicles for stable delivery of genes of interest to airway cells. Basal cells comprise an important progenitor population in the conducting airways and are responsible for replenishing multiple surface epithelial cell types. Thus, this cell population is an important gene therapy target. Here, we modify two capsids with demonstrated tropism for the conducting airways, AAV2.5T and AAV6.2, to further improve basal cell transduction.

Methods: AAV2.5T and AAV6.2 are two capsid types that were generated by directed evolution to achieve improved airway tropism. AAV2.5T is a combination of AAV2 and 5 with a point mutation, A581T. AAV6.2 is the AAV6 capsid with a mutated F129L residue. Both A581 and F129 are present in AAV6.2 and AAV2.5T. We first asked if incorporating reciprocal residues would increase airway tropism in primary surface epithelia and basal cells, termed AAV2.5T_F129L and AAV6.2_A581T. We next asked if modifying surface tyrosine residues would enhance transduction efficiencies as has been reported with the AAV6.2FF capsid (Y445F, Y731F). To this end, we generated a AAV2.5T_FF capsid by modifying Y437F and Y720F residues. Lastly, we generated combinatorial mutations, termed AAV2.5T_FFL and AAV6.2_FFT. We compared transduction efficiencies on the apical surface of well differentiated primary airway epithelia as well as primary basal cells. GFP was quantified by flow cytometry 3 weeks post-transduction. Ratios of cell types (basal, ciliated, and secretory) were confirmed by flow cytometry.

Results: We observed increased transduction efficiencies using the AAV2.5T_F129L and AAV2.5T_FFL capsids compared to AAV2.5T alone. Additionally, we observed increased GFP expression with the AAV6.2FF capsid relative to AAV6.2. As expected, ratios of cell populations transduced remain unchanged with capsid modification. Apical transduction leads to mostly ciliated and secretory cells transduced, but basal cells are also transduced. Overall, AAV2.5T transduces surface epithelia with higher efficiencies than AAV6.2, however AAV6.2 better transduces airway basal cells.

Conclusions: AAV2.5T and AAV6.2 have different mechanisms of entry. AAV2.5T entry is dependent on sialic acid binding, whereas AAV6.2 enters via a heparin binding motif. By incorporating reciprocal residues of each and ablating surface tyrosine residues, we observed increased transduction efficiencies of surface airway epithelia. Correcting airway progenitor cells is an important consideration for achieving life-long therapeutic benefits from cystic fibrosis gene therapy.

Acknowledgements:

References: