Effects of rK5 on endothelial cells. (A) The effect of rK5 on VEGF-induced cell proliferation. BrdU incorporation was used as an index of proliferation. Each data point represents a mean of 3 independent experiments using triplicate cultures. HUVECs (□), human foreskin fibroblast (■), OVCAR-5 (●), and OVCAR-3 (▵). Purified rAAV-GFP CM (3 μg/mL)–treated HUVECs (▴) was used as negative control. Inset: anti-His Tag immunoblot analysis of purified recombinant AAV-K5 (rK5) in fractions eluted by 100 mM imidazole (lanes 2-5) and CM before purification (lane 6). A major band of approximately 10 kDa was detected. Lane 1 shows CM of nontransduced HEK-293 cells; lane 7 shows endostatin with a C-terminal poly-His–tagged (positive control); and lane 8 shows purified rAAV-GFP CM. (B) The effect of rK5 on endothelial cell migration in Boyden chamber assay. The migration of endothelial cells was determined by a method previously described.¹⁸  Endothelial cells were prelabeled with 5.0 μM 5⁶-CFDA and induced to migrate toward the bottom chamber containing 40 ng/mL VEGF. The number of cells migrated to the bottom side of the membrane was counted using a PixCell II LCM system at 40× magnification. Data represent values from 3 independent experiments using 3 wells per concentration. M199 medium (5% FBS; ▵) was used to determine basal level of migration. P125A endostatin (■) was used as a positive control. rK5 induced concentration-dependent inhibition of cell migration (□). (C) The effect of rK5 on angiogenesis was evaluated by using VEGF-stimulated tube formation in a Matrigel assay.¹⁸  Representative images of endothelial cell tube formation are shown. (Ci) Basal level of tube formation. (Cii-iv) Tube formation induced by 40 ng/mL VEGF. (Ciii) rK5. (Civ) P125A endostatin (300 ng/mL). Bright-field images were recorded at 40× magnification (top row) and processed for analysis (Cv-viii) as described in “Materials and methods” (bottom row). (D) Morphometric analysis of tube formation. Ends (□), branch points (■), and tube length (▵) are shown. Values represent data from 3 independent experiments using 3 wells per sample. VEGF-induced tube formation was considered as 100% tube formation. Values are shown as means ± SE. (E) rK5 induced endothelial cell apoptosis under nutrient-rich culture conditions. HUVECs were cultured in the presence of 40 ng/mL VEGF. Cells were treated with 1.5 μg/mL rK5 for 24 hours and then labeled with annexin V FLUOS and DAPI. Images were recorded at 40× magnification and processed by the Image Processing Toolkit and Adobe Photoshop. Percentage of apoptotic cells was determined by the ratio of annexin V⁺ cells over the total number of nuclei per field. Purified rAAV-GFP CM was used as a negative control at a similar concentration. Values represent data from 2 independent experiments using 3 wells per sample. Error bars denote SE (*P < .05). (F) HUVECs were stimulated with VEGF and treated with different concentrations of rK5. In one set of cultures, rK5 was withdrawn after 24 hours of exposure (■) and replaced with fresh media. The second set of cultures was continuously exposed to rK5 (□). Cells from both treatment groups were labeled with TUNEL concurrently with DAPI after 72 hours of treatment. Percentage of apoptotic HUVECs was determined by the ratio of TUNEL⁺ cells over the total number of nuclei observed per field. Purified rAAV-GFP CM (3 μg/mL; ▴) and 0.02% H₂O₂ (♦) were used as a negative and positive control, respectively. Values represent data from 3 independent experiments using triplicate wells per concentration. Values are shown as means ± SE.