TLR2/6-dependent stimulation by MALP-2 induces GM-CSF release from HUVECs. (A) Pictures represent protein array membranes specific for angiogenesis-related factors. Membranes were incubated with supernatants from HUVECs not stimulated (control) or stimulated with MALP-2 (1 μg/mL) for 24 hours. Representative pictures of 2 independent experiments are shown. The position of GM-CSF is indicated by a white box. POS indicates positive control; NEG, negative control. (B) Quantification of the dot intensity relative to the positive control (n = 2). b.t. indicates below threshold. (C) GM-CSF levels in supernatants of HUVECs after MALP-2 (1 μg/mL) stimulation for 24 hours as assessed by ELISA. Unspecific IgG (50 μg/mL) was used as antibody control (n = 4). AB indicates antibody (25 μg/mL each). **P < .01 vs control. ##P < .01 vs MALP-2. (D) Knockdown of TRAF-6 protein levels after siRNA application (10 nM) in HUVECs was confirmed by Western blot. GAPDH expression is shown as loading control. Representative pictures of 3 independent experiments are shown. (E) GM-CSF levels in supernatants of HUVECs after siRNA application (10 nM) and MALP-2 (1 μg/mL) stimulation for 24 hours determined by ELISA (n = 4). **P < .01 vs control. ##P < .01 vs MALP-2 plus control siRNA. (F) Activation of the MAPK and the NF-κB pathway after MALP-2 (1 μg/mL) stimulation for the indicated time points was demonstrated by Western blot. Representative pictures of 3 independent experiments are shown. (G) GM-CSF levels in supernatants of HUVECs after MALP-2 (1 μg/mL) stimulation for 24 hours in the presence of pharmacologic inhibitors of the MAPK and the NF-κB pathway are assessed by ELISA (n = 5). **P < .01 vs control. ^#P < .01 vs MALP-2. (C,E,G) Error bars represent mean ± SEM.