MyH9 is essential for the angiogenic function of nucleolin. (A) MyH9-RNAi suppressed the expression of MyH9. After being transfected with control or MyH9 siRNA expression vectors, HMECs were lysed and subjected to immunoblotting analysis with anti-MyH9 Ab, and anti–β-actin as a sample loading control. (B) MyH9-RNAi diminished cell-surface nucleolin. After cotransfection with GFP and MyH9 siRNA expression vectors, HMECs were fixed without being permeabilized to visualize cell-surface nucleolin (red). Arrows indicate GFP-negative cells, and asterisks show GFP-positive cells in which expression of MyH9 was inhibited. Bar, 20 μm. (C) Without being permeabilized, the ML7-treated HMECs and cell nucleolin (green) were stained and the nuclei (blue) were counterstained with DAPI. PBS was added as a control. Bar, 10 μm. (D) The amount of cell-surface nucleolin of HMECs was determined by FACS as described in “Materials and methods.” For the histograms, NC indicates negative control; isotype Ab, isotype Ab control; ST, serum starved; VM, cultured on Matrigel supplemented with 10 ng/mL VEGF; MyH9i, RNAi against MyH9; ML7, treated cells with 5 μM ML7. Suppression of the endogenous MyH9 led to reduction of cell-surface nucleolin. HMECs were subjected to the following treatment: ML7-treated (E); down-regulated MyH9 by MyH9 RNAi (F); overexpressed MyH9ΔN-GFP (G), and then the migration of HMECs was analyzed as described in “Materials and methods.” ***P < .01. (H) MyH9ΔN-GFP was overexpressed in ECV304, and then a cell-invasive assay was performed; cell invasion was significantly inhibited by overexpressed, deficient MyH9. Overexpressed GFP was used as a control. Bar, 50 μm. (I) ECV304 cells were subjected to the following treatment: ML7-treated; down-regulated MyH9 by MyH9 RNAi; overexpressed MyH9ΔN-GFP; and then the tubule formation of each group was analyzed as described in “Materials and methods.” Bar, 100 μm. ***P < .01.