CREB is required for maintaining basal endothelial barrier function and for preventing persistent increases in endothelial permeability postthrombin challenge. (A-C) Depletion of CREB impairs endothelial barrier function. (A) HPAE monolayers were transfected with 2.4 μg of either scrambled (SiSc) or CREB siRNA (SiCREB) for 48 hours. Cell lysates were immunoblotted 48 hours after transfection using anti-CREB antibody to analyze CREB expression. Immunoblot with anti-actin antibody was used as a loading control. (B) Cells plated on gold electrodes were transfected with either SiSc or SiCREB and after 24 hours after transfection, TEER was determined for indicated times. Data represent mean ± SD from 3 experiments performed in duplicates. Asterisk (*) indicates values different from SiSc-transfected endothelial monolayers (P < .05). (C) HPAE monolayers were transfected with SiSc or SiCREB for 48 hours after which cells were simulated with 50nM thrombin, and changes in TEER were recorded overtime. The effects of CREB inhibition on endothelial barrier recovery postthrombin challenge were examined after normalization of TEER values after 48 hours of transfection to 100%. Data represent mean ± SD from 3 experiments performed in duplicates. Asterisk (*) indicates values different from SiSc-transfected endothelial monolayers (P < .05). (D) HPAE cells seeded on Transwell plates were transfected with SiSc or SiCREB and 48 hours after transfection, EBA clearance was determined after without or with thrombin challenge. Data represents the mean ± SD from 3 experiments. Asterisk (*) indicates difference from unstimulated SiSc monolayer, and double asterisk (**) indicates difference from thrombin-stimulated SiSc- or control SiCREB-expressing cells (P < .05). (E-G) Expression of dn-CREB mutant impairs endothelial barrier function. HPAE cells seeded on 6-well plates or gold-plated electrodes were transfected with control cDNA or dn-CREB mutant. After 24 hours after transfection, we determined changes in CREB phosphorylation (E) and TEER in naive monolayer (F) or after stimulation with 50nM thrombin (G) at indicated times. For determining CREB phosphorylation and expression of CREB mutant, cell lysates were immunoblotted with anti–Ser133-CREB antibody anti-CREB antibodies. Immunoblot with anti-actin antibody was performed for protein loading control. The effects of CREB inhibition on endothelial barrier recovery postthrombin challenge were examined after normalization of TEER values after 24 hours after transfection to 100%. Data represent mean ± SD from 3 experiments performed in duplicates. Asterisk (*) indicates values different from SiSc-transfected endothelial monolayers (P < .05). (H-I) CREB knockdown disrupts adherens junctions and increases actin stress fiber formation. HPAE cells expressing SiSc or SiCREB were left unstimulated or stimulated with 50nM thrombin for indicated times, fixed, and immunostained with anti–VE-cadherin antibody and rhodamine phalloidin as described in “Immunofluoresence.” Cells were visualized using an LSM confocal microscope. The images shown are representative of 3 independent experiments. (I) Plot shows time course of fold-change in intercellular gap areas in siSc- or SiCREB-expressing cells after thrombin stimulation. Gap area was quantified using National Institutes of Health ImageJ Version 1.44 software. Asterisk (*) indicates values different from values at time 0 in SiSc-transfected monolayers (P < .05), and double asterisk (**) indicates values different from corresponding thrombin-stimulated SiSc-transfected endothelial monolayers (P < .05). (J) Effect of CREB depletion on actin polymerization. SiSc or SiCREB cells were lysed and centrifuged to separate G- and F-actin. Lysates were then immunoblotted with anti-actin antibody to determine actin polymerization. Numbers indicate densitometric values from 3 individual experiments.