Figure 6

NHERF1 is required for chemotactic migration of cancer cells. (a) Schematic domain structures of GFP-labeled NHERF1 constructs including the wild type (WT), C-terminally truncated form (ΔCT), and C-terminal fragment (CT) (left panel). Expression levels of GFP-NHERF1 constructs were determined using anti-GFP antibody (right panel) as indicated by the arrows. (b) Inhibition of pseudopodia formation by overexpression of NHERF1 deletion mutants. OVCAR-3 cells were transfected with GFP-NHERF1 constructs, and migration assays were performed 2 days after transfection, as described above. The migratory cells fixed on the membrane were probed with anti-cpERM antibody. Pseudopodial localizations of GFP-NHERF1 and cpERM proteins were imaged under confocal microscopy with constant settings. These results are from a single experiment that was representative of three experiments performed on independent preparations. Scale bar, 20 μm. (c) siRNA-mediated silencing of NHERF1 expression in various cell lines derived from gynecological cancers. OVCAR-3 (ascites-derived ovarian cancer cell line), SK-OV-3 (ascites-derived ovarian cancer cell line) and MDA-MB-231 cells (breast carcinoma cell line) were tested in this study. Each cell line was transfected with either siRNA duplex specific for NHERF1 or control siRNA. NHERF1 levels in cell lysates were determined by western blotting with anti-NHERF1 antibody. (d) Effect of NHERF1 depletion on LPA-induced migration of OVCAR-3 cells. Three days after siRNA transfection, OVCAR-3 cells were mounted onto a collagen-coated porous membrane (pore size, 8 μm) of a modified Boyden chamber and allowed to migrate to the lower side for 3.5 h in the presence or absence of 1 μM LPA in the lower chamber. Migratory cells in the lower membrane were visualized by Hoechst 33342 staining of nuclei. These images from OVCAR-3 cells represent three independent experiments. (e) Effect of NHERF1 depletion on the migration of gynecological cancer cell lines. OVCAR-3, SK-OV-3 and MDA-MB-231 cells were tested in this study. Each cell line was transfected with siRNA duplexes as described in (c). After chemotactic migration assays, the membranes were observed by fluorescence microscopy, and images were captured in five randomly chosen high-power fields (HPF, × 20 objective). The numbers of migratory cells per membrane were directly counted and analyzed statistically in each cell line. Each bar represents the mean±s.e.m. from three independent experiments. All data represent mean ±s.e.m. *P<0.05 and NS, non-significant.