105. Interactions between Modified AAV Vectors and Alternative Cell Surface Receptors Define Intercellular Trafficking Pathways and Tranduction in Human Vascular Endothelial Cells

Abstract

Top of pageAbstract Adeno-associated virus (AAV) vectors transduce a wide variety of cell types. However, cells of the vascular system remain refractive to AAV-mediated gene transduction. AAV1-based vectors have shown the most promise in the vasculature. To improve upon AAV1's vascular tropism, peptides targeted to receptors expressed on the vasculature were incorporated into the AAV1 capsid. These included peptides targeted to vascular endothelial growth factor receptor 2 (VEGFR2), Tie2, and the 4C-RGD peptide targeted to integrin receptors. Near wild type titers were obtained from all of the peptidemodified vectors. Modified vectors were used to infect four different human endothelial cell (EC) populations (human umbilical vein EC, HUVEC; human coronary artery EC; HCAEC; human saphenous vein EC, HSaVEC; and human coronary microvascular EC, HCMVEC) and human HeLa cells. At least one modified AAV1 vector targeted to each of these receptors significantly increased transduction of all EC lines (average 10 to 60-fold). Specificity was assessed using either soluble peptides or anti-receptor antibody. To determine the mechanism of increased transduction, endothelial or HeLa cell lines were transduced with modified vectors in the presence of inhibitors of endosomal acidification or proteasome activity, adenovirus, or after treatment with neuraminidase to remove cell surface sialic acid. Inhibiting endosomal acidification greatly reduced transduction of both cell types with all vectors tested, suggesting that both targeted and un-modified AAV1 vectors enter cells via endocytosis and require some degree of endosomal maturation for effective gene transfer. Removing sialic acid from the cell surface drastically reduced transduction with all vectors except the RGDmodified vector. This suggested that while the VEGFR2 and Tie2 targeted vectors were significantly more efficient than unmodified vectors at transducing EC, enhancement was due to novel co-receptor usage and/or altered intracellular trafficking rather than redirected cellular attachment. These results further suggest different requirements for receptor/co-receptor usage for different targeted AAV vectors. Co-infection with adenovirus showed variable effects depending on the targeting ligand. When self-complementary vectors were used to infect EC, adenovirus greatly increased transduction of unmodified AAV1 vectors but only moderately increased transduction of targeted vectors. Proteasome inhibitors had no effect on transduction mediated by the RGD-modified vector, and only slightly increased transduction of the VEGFDR2 and Tie2 targeted vectors (1.5 to 3-fold), whereas proteasome inhibitors increased transduction of unmodified AAV1 vectors over 14-fold on HUVEC. Taken together, these studies suggest that intracellular trafficking pathways are defined by initial vector cellular interactions and that transduction of endothelial cells can be increased by either targeting a new receptor/cellular entry pathway, or by altering post-attachment cellular entry/trafficking pathways.