The NRP-1 expression level was evaluated using Western blot analysis and showed that, as expected, both tumor and endothelial cells express high levels of NRP-1 [18], whereas no expression was detected in CHO cell lines (Figure?1 .01; *** .001 versus control. migration and tumor cell invasion. These results suggest that NRP-1/PTN interaction provides a novel mechanism for controlling the response of endothelial and tumoral cells to PTN and may explain, at least in part, how PTN contributes to tumor angiogenesis and cancer progression. Introduction The heparin-binding growth factor pleiotrophin (PTN), also known as stacks of confocal images (8 m from the base to the top in 0.5-m steps). Image processing was done using ImageJ software [28]. Residual blurring was removed by spatial deconvolution: Point-Spread Function (PSF) was calculated using the ImageJ plugin PSF Generator [29], and deconvolution properly speaking was carried out using the Richardson-Lucy algorithm used with the Deconvolution Lab ImageJ plugin [30]. The latter two freeware are available from the EPFL (Ecole Polytechnique Fdrale de Lausanne [Biomedical Imaging Group]), Lausanne. Switzerland. ? Fire ? look-up table was used in merged representations to improve visibility of the NRP-1 labeling, without altering linearity of the signal. Migration and Invasion Assays Migration assays were accomplished using a 24-well chemotaxis chamber (Transwell, BD Biosciences). Pore size 8-m polycarbonate filters were JANEX-1 coated with 10 mg/ml of type I collagen (Serva). A total of 1 1 105 cells in EBM-2 medium supplemented with 1% FBS were plated into the upper chamber of Transwell chamber in the presence of PTN alone or with PTN and an antiCNRP-1 blocking antibody, or IgGs control (R&D System) were added to the lower chamber. Cells were allowed to migrate for 6 hours at 37C. Nonmigrated cells were then removed by wiping with a cotton tip, and migrated cells were fixed with absolute ethanol and stained with crystal violet 0.2% (v/v) in ethanol 2%. Migrated cells in three fields of each well (Leitz Aristoplan microscope, ?10 objective) were quantified by image analysis. Briefly, a parameterized extraction of the blue color was followed by a threshold (“Otsu” method [31]) and a subsegmentation (watershed method [32]) to determine the number of cells. In invasion assay, PC3 cells (2 104) were suspended in serum-free medium and seeded onto Matrigel (BD Biosciences)-coated Transwell chamber (20 g/well). Medium with or without PTN supplemented with antiCNRP-1 IgG or nonimmune IgG (R&D system) was introduced into the lower chamber. Invasion was carried out for 14 hours at 37C. Cells were fixed and treated, and the number of invading cells was determined as described above for migration assay. Statistical Analysis Values are reported as means SEM. Statistical significance was determined by the analysis of variance unpaired test using GraphPad Prism 4.0 software. Values of .05 were considered significant. Results NRP-1 and PTN Growth Factor To determine whether PTN interacts with NRP-1, we first checked NRP-1 expression levels in four cell cultures, including those that express NRP-1, such as Rabbit polyclonal to ARL16 PC3 and MDA-MB231, HUVEC, as well as in Chinese hamster ovary (CHO) cells, which do not express JANEX-1 NRP-1. The NRP-1 expression level was evaluated using Western blot analysis and showed that, as expected, both tumor and endothelial cells express high levels of NRP-1 [18], whereas no expression was detected in CHO cell lines (Figure?1 .01; *** .001 versus control. (C) The heparin JANEX-1 binding growth factors FGF-2 or VEGF A165 compete with PTN for binding to NRP-1. PC3 cell lysate was incubated with PTN-GST alone (?) or in the presence of 10-fold excess of indicated growth factors followed by incubation with glutathione-Sepharose beads. Immunoblotting was done as described above. Because NRP-1 binds multiple mitogenic and angiogenic growth factors, such as vascular endothelial growth factor (VEGF A165) and basic fibroblast growth factor (bFGF, FGF2), we carried out a pull-down experiment with GST-PTN1-136 in the presence of VEGF A165 or FGF2. As shown in Figure?2 .05; ** .01 versus control. (C) PTN bound directly to NRP-1. Microtiter plates were coated with rNRP-1 or BSA and incubated with increasing concentrations of PTN (0 to 8 JANEX-1 g/ml). Bound PTN was detected by ELISA-based measurement of streptavidin-HRP staining. The results represent the means of three experiments. To further study whether PTN interacts directly with NRP-1, we used an ELISA-based assay in which rNRP-1 was immobilized. PTN bound specifically and in a dose-dependent manner to immobilized rNRP-1, whereas a BSA control bound very weakly to rNRP-1, corresponding to nonspecific binding (Figure?3 .05; ** .01 versus control..
