Genetic modification of adeno-associated viral vector type 2 capsid enhances gene transfer efficiency in polarized human airway epithelial cells

Abstract

Cystic fibrosis (CF) is a common genetic disease characterized by defects in the expression of the CF transmembrane conductance regulator (CFTR) gene. Gene therapy offers better hope for the treatment of CF. Adeno-associated viral (AAV) vectors are capable of stable expression with low immunogenicity. Despite their potential in CF gene therapy, gene transfer efficiency by AAV is limited because of pathophysiological barriers in these patients. Although a few AAV serotypes have shown better transduction compared with the AAV2-based vectors, gene transfer efficiency in human airway epithelium has still not reached therapeutic levels. To engineer better AAV vectors for enhanced gene delivery in human airway epithelium, we developed and characterized mutant AAV vectors by genetic capsid modification, modeling the well-characterized AAV2 serotype. We genetically incorporated putative high-affinity peptide ligands to human airway epithelium on the GH loop region of AAV2 capsid protein. Six independent mutant AAV were constructed, containing peptide ligands previously reported to bind with high affinity for known and unknown receptors on human airway epithelial cells. The vectors were tested on nonairway cells and nonpolarized and polarized human airway epithelial cells for enhanced infectivity. One of the mutant vectors, with the peptide sequence THALWHT, not only showed the highest transduction in undifferentiated human airway epithelial cells but also indicated significant transduction in polarized cells. Interestingly, this modified vector was also able to infect cells independently of the heparan sulfate proteoglycan receptor. Incorporation of this ligand on other AAV serotypes, which have shown improved gene transfer efficiency in the human airway epithelium, may enhance the application of AAV vectors in CF gene therapy.

FIG. 1.

Western blot analysis of 293 cell lysates containing wild-type (wt) and mutant rAAV capsids and transgene expression in 293 cells with rAAV-luc encapsidated in wild-type or mutant capsids. (A) Equal volumes of cell lysates containing wild-type and representative mutant capsids, after transfection of the respective plasmids along with the plasmid pXX6, containing adenovirus helper functions, were separated by 10% SDS–PAGE and analyzed by Western blotting with AAV capsid antibody B1 followed by horseradish peroxidase-conjugated anti-mouse secondary antibody. The molecular masses of VP-1, VP-2, and VP-3 are 87, 73, and 62 kDa, respectively. (B) HEK-293 cells were either mock-transduced or transduced with 100 MOI of rAAV-luciferase with wild-type or mutant capsid for 2 hr at 37°C. After infections, free viral particles were removed by washing with phosphate-buffered saline (PBS) and the cells were incubated for 48 hr. Luciferase activity was determine from cell lysates and expressed as relative light units (RLU), normalized to the protein content of each cell lysate. *p < 0.01 between wild type and m2.

FIG. 2.

Luciferase activity in Calu-3 cells transduced with rAAV-luc containing wild-type and mutant capsids, and the effect of soluble heparin on transduction of wild-type and m2 rAAV-luc in Calu-3 cells. (A) Calu-3 cells, grown as a monolayer culture, were mock-transduced or transduced with 100 MOI of rAAV-luciferase with wild-type or the indicated mutant capsids for 2 hr at 37°C. After infection, free virus was removed by washing with PBS and the cells were incubated for 48 hr. Luciferase activity was determined from cell lysates and expressed as relative light units, normalized to the protein content of each cell lysate. *p < 0.01 between wild-type and m2. (B) Both virus and cells were preincubated with heparin (500 μg/ml) at 37°C for 1 hr, after which infection was performed for 2 hr, also in the presence of soluble heparin. After infection, cells were washed with PBS and grown in complete medium for 48 hr. The luciferase assay was done, using cell lysates from individual transductions, and values were normalized to the protein content of each lysate. *p < 0.0015.

FIG. 3.

Transduction of rAAV-luc in wild-type or mutant capsids in polarized human airway epithelium. Calu-3 cells were grown in Transwell filters for 7 days to allow polarization. The cells were either mock-transduced or transduced with rAAV-luc encapsidated in wild-type or mutant capsids, as indicated previously, for 2 hr at 37°C. After infection, free viral particles were removed by washing with PBS and cells were grown for an additional 48 hr. Luciferase activity was measured from cell lysates and expressed as relative light units, normalized to the protein content of each lysate. *p < 0.0091 between m2 and the other vector types.

FIG. 4.

Transduction efficiency of rAAV with mutant capsids in CFBE cells. CFBE cells were mock-transduced or transduced with rAAV-luc encapsidated in wild-type or the indicated mutant (m) capsids for 2 hr at 37°C. Free viral particles were removed by washing with PBS. The cells were harvested 48 hr later and luciferase activity was determined from the cell lysates. Luciferase activity from individual samples was normalized to the protein content of each lysate and expressed as relative light units. *p < 0.03 between m2 and the other types.