Suppression of angiogenesis by E2-2. (A) Photographs for en bloc resection, including Matrigel plugs with adjacent subcutaneous tissues. Samples were visualized using a stereomicroscope (S8APO; Leica). Images were acquired with EC3 (Leica) and processed with the LAS EZ (Leica) and Adobe Photoshop 7.0.1 software (Adobe). (B) PECAM-1 staining of frozen sections. Matrigels are located to the right side of broken lines, whereas there are mouse subcutaneous tissues to the left side of broken lines. PECAM1-positive areas are shown as white. After samples were mounted with Fluorescent Mounting Medium (Dako Denmark), they were visualized using an immunofluorescence microscope (Axiovert 200M; Carl Zeiss) with a 63×/1.4 oil objective lenses (Carl Zeiss). Images were acquired with AxioCam MRm 60-C1 (Carl Zeiss) and processed with the AxioVision Rel 4.4 (Carl Zeiss) and Adobe Photoshop 7.0.1 software (Adobe). (C) Relative PECAM-1-positive area in Matrigel plugs. Five fields were randomly chosen in each condition from panel B, and PECAM-1-positive areas were scored using the imaging software. PECAM-1-positive areas were normalized with the areas of the field. Each relative PECAM-1-positive area was calculated by the comparison of the score in Matrigel plugs, including GFP. (D) Effect of E2-2 on the expression of mRNAs implicated in EC activation. mRNAs involved in EC activation were analyzed by RT-PCR. HUVECs or CPAEs were infected with adenoviruses expressing E2-2 or GFP as a negative control. Gene transcript names are indicated to the left of the figure. mRNAs for β-actin and GAPDH were included as internal controls. The expression of Myc-E2-2 was evaluated in total cell lysates using an anti-Myc 9E10 antibody (bottom panel). Adex indicates adenovirus. To show that EC markers are expressed in CPAEs, RT-PCR was carried out (supplemental Figure 6A). (E) E2-2 inhibits VEGF-induced Erk phosphorylation. Phospho-Erk1/2 (top panel), Erk2 (second panel), Myc-E2-2 (third panel), and β-actin (bottom panel) levels were analyzed by Western blots of total cell lysates. Adenoviruses expressing GFP were used as a negative control. The expression for phosphorylation of Erk1/2 was normalized using the intensity of the band corresponding to Erk2. Each relative intensity was calculated by the comparison of the value for cells infected with GFP-expressing adenoviruses in the absence of VEGF. (F) Graphical presentation for relative intensity of phospho-Erk1/2 levels in panel E: ● represents GFP-expressing adenoviruses; ○, Myc-E2-2–expressing adenovirusues. (G) E2-2 perturbs EC proliferation. Each experiment was performed in triplicate and repeated a few times. Values represent the mean plus or minus SD (n = 3).