Knockdown of CYLD expression impairs endothelial tube formation and sprouting. (A) Immunoblot analysis of CYLD and tubulin expression in HUVECs transfected with CYLD or control siRNAs for 72 hours. (B) HUVECs transfected with CYLD or control siRNAs were plated onto matrigel, and photographs were taken 4 and 16 hours later. Objective lens used was A-Plan 10×/0.25 NA dry (Carl Zeiss Inc). (C) Experiments were performed as in panel B, and the cumulative tube length was measured. Data are the mean and standard error from 3 experiments with each performed in triplicate (*P < .05 vs control). (D) Immunoblot analysis of CYLD and GFP expression in HUVECs transfected with pEGFPC1 and pSUPER or pSUPER-CYLD plasmids for 72 hours. (E) HUVECs transfected with pEGFPC1 and pSUPER or pSUPER-CYLD plasmids were plated onto matrigel, and photographs were taken 2 hours later. GFP was used to mark transfected cells. White arrowheads indicate cells that align end-to-end, and black arrowheads indicate cells without such a property. Objective lens used was A-Plan 20×/0.3 NA dry (Carl Zeiss Inc). (F) Capillary-like sprout formation from spheroids generated from HUVECs transfected with CYLD or control siRNAs. Objective lens used was A-Plan 10×/0.25 NA dry (Carl Zeiss Inc). (G) Experiments were performed as in panel F, and the cumulative sprout length was measured. Data are the mean and SE from 2 experiments with each performed in triplicate (*P < .05 vs control). All images in this figure and in the following figures (except Figure 3E and Figure 4A) were taken with the Axio Observer A1 fluorescence microscope (Carl Zeiss Inc), equipped with the AxioCam MRm Rev.3 CCD camera and the AxioVision Rel. 4.1 software. All images were processed with the Adobe Photoshop CS8.0 software (Adobe Systems).