Dynamic Modulation of Interendothelial Gap Junctional Communication by 11,12-Epoxyeicosatrienoic Acid by Rüdiger Popp, Ralf P. Brandes, Gregor Ott, Rudi Busse, and Ingrid Fleming Circulation Research Volume 90(7):800-806 April 19, 2002 Copyright © American Heart Association, Inc. All rights reserved.

Figure 1. Bradykinin-induced changes in gap junctional communication. Figure 1. Bradykinin-induced changes in gap junctional communication. A and B, Time course of the bradykinin (10 nmol/L)–induced changes in electrical coupling (A) and the transfer of Lucifer yellow (B) between porcine coronary artery endothelial cells. Experiments were performed in the absence and presence of L-NA (300 μmol/L) and diclofenac (diclo, 10 μmol/L). C, Pharmacological characterization of the communication-enhancing factor in porcine coronary endothelial cells. Dye (Lucifer yellow) transfer between endothelial cells was assessed in cells pretreated with solvent, sulfaphenazole (Sulfa, 10 μmol/L), or the combination of charybdotoxin/apamin (CA, both 100 nmol/L) under basal conditions and 60 seconds after the application of bradykinin (100 nmol/L). Experiments were performed in the continuous presence of L-NA (300 μmol/L) and diclo (10 μmol/L), and the results represent the mean±SEM of data obtained in 8 separate experiments. *P<0.05 and **P<0.01 vs control. Rüdiger Popp et al. Circ Res. 2002;90:800-806 Copyright © American Heart Association, Inc. All rights reserved.

Figure 2. Time course of the 11,12-EET–induced changes in gap junctional communication between human umbilical vein endothelial cells. Figure 2. Time course of the 11,12-EET–induced changes in gap junctional communication between human umbilical vein endothelial cells. A and B, Endothelial cells were stimulated with 11,12-EET (3 μmol/L) for the time shown, and the transfer of Lucifer yellow (A) and the electrical coupling (B) between endothelial cells were assessed. C, Dye transfer between human endothelial cells was assessed in cells pretreated with solvent, Sulfa (10 μmol/L), or CA (both 100 nmol/L) under basal conditions and 60 seconds after the application of 11,12-EET (3 μmol/L). Experiments were performed in the continuous presence of L-NA (300 μmol/L) and diclofenac (10 μmol/L), and the results represent the mean±SEM of data obtained in 8 separate experiments. *P<0.05 and **P<0.01 vs control (CTL). Rüdiger Popp et al. Circ Res. 2002;90:800-806 Copyright © American Heart Association, Inc. All rights reserved.

Figure 3. Effect of enhancing CYP 2C expression on gap junctional communication in endothelial cells. Figure 3. Effect of enhancing CYP 2C expression on gap junctional communication in endothelial cells. Porcine coronary artery endothelial cells (A) and human umbilical vein endothelial cells (B) were incubated with either solvent (CTL) or nifedipine (Nif, 0.1 μmol/L; 18 hours), and dye (Lucifer yellow) coupling was determined in the absence and presence of Sulfa (10 μmol/L). Experiments were performed in the continuous presence of L-NA (300 μmol/L) and diclofenac (10 μmol/L), and the results represent the mean±SEM of data obtained in 8 to 12 separate experiments. **P<0.01. Rüdiger Popp et al. Circ Res. 2002;90:800-806 Copyright © American Heart Association, Inc. All rights reserved.

Figure 4. Effect of cAMP elevation on gap junctional communication and the Triton X-100 solubility of Cx43 in human umbilical vein endothelial cells. Figure 4. Effect of cAMP elevation on gap junctional communication and the Triton X-100 solubility of Cx43 in human umbilical vein endothelial cells. A, Dye (Lucifer yellow) coupling in cultured human endothelial cells after the application of forskolin (For, 10 μmol/L; 20 minutes) or a caged cAMP (cAMP, 50 μmol/L; flash-activated at 360 nm for 1 minute). Experiments were performed in the continuous presence of L-NA (300 μmol/L) and diclofenac (10 μmol/L), and the results represent the mean±SEM of data obtained in 6 to 10 separate experiments. **P<0.01. B, Western blot with an antibody recognizing the Cx43 showing the time course of the changes in the recovery of Cx43 in the Triton X-100–insoluble cell fraction in cells incubated with a caged cAMP (cAMP, 50 μmol/L; flash-activated at λ 360 nm). Experiments were performed in the absence and presence of the PKA inhibitor Rp-cAMPS (10 μmol/L). Identical results were obtained in 2 additional experiments. Rüdiger Popp et al. Circ Res. 2002;90:800-806 Copyright © American Heart Association, Inc. All rights reserved.

Figure 5. Role for bradykinin-induced and 11,12-EET–induced changes in cAMP levels in the regulation of gap junctional communication. Figure 5. Role for bradykinin-induced and 11,12-EET–induced changes in cAMP levels in the regulation of gap junctional communication. A, Effect of solvent (open bars), bradykinin (10 nmol/L, 60 seconds; shaded bars), and 11,12-EET (1 μmol/L, 60 seconds; solid bars) on the accumulation of cAMP in confluent cultures of porcine coronary endothelial cells. Experiments were performed in the continuous presence of L-NA (300 μmol/L) and diclofenac (10 μmol/L) and in the absence and presence of Sulfa (10 μmol/L). The results represent the mean±SEM of data obtained in 4 separate experiments. B and C, Effect of inhibiting adenylyl cyclase and PKA on the bradykinin-induced increase in gap junctional communication. Coronary artery endothelial cells were stimulated with either solvent or bradykinin (10 nmol/L, 60 seconds), and gap junctional communication was assessed by the intercellular transfer of Lucifer yellow. Experiments were performed in the absence and presence of 2′,5′-DDA (30 nmol/L) (B) and Rp-cAMPS (Rp, 10 μmol/L) (C). D, Effect of inhibiting PKA on the cAMP-induced and 11,12-EET–induced increase in gap junctional communication. Dye coupling in cultured human endothelial cells after the release of a caged cAMP (cAMP, 50 μmol/L; flash-activated at λ 360 nm, for 1 minute; shaded bars) or after the application of 11,12-EET (1 μmol/L, 60 seconds; solid bars) is shown. Experiments were performed in the continuous presence of L-NA (300 μmol/L) and diclofenac (10 μmol/L) and in the absence (solvent) and presence of KT5720 (KT, 1 μmol/L). The results represent the mean±SEM of data obtained in 10 separate experiments. *P<0.05 and **P<0.01 vs control. Rüdiger Popp et al. Circ Res. 2002;90:800-806 Copyright © American Heart Association, Inc. All rights reserved.

Figure 6. Role of ERK1/2 in mediating the delayed bradykinin-induced and 11,12-EET–induced uncoupling of endothelial cells. Figure 6. Role of ERK1/2 in mediating the delayed bradykinin-induced and 11,12-EET–induced uncoupling of endothelial cells. A, Capacitance measurements showing the effect of bradykinin (10 nmol/L, 10 minutes) on the coupling of porcine coronary endothelial cells in the absence and presence of solvent, Sulfa (10 μmol/L), CA (both 100 nmol/L), or PD 98059 (50 μmol/L). B, Representative experiment showing the time course of bradykinin (100 nmol/L)–induced changes in electrical coupling between coronary artery endothelial cells in the absence and presence of PD 98059. C, Concentration-dependent inhibition of endothelial cell dye (Lucifer yellow) coupling by 11,12-EET (1 to 10 μmol/L, 10 minutes). Experiments were performed in the absence and presence (striped bar) of PD 98059. The results represent the mean±SEM of data obtained in 10 separate experiments. *P<0.05, **P<0.01, and ***P<0.001 vs control. Rüdiger Popp et al. Circ Res. 2002;90:800-806 Copyright © American Heart Association, Inc. All rights reserved.

Figure 7. Western blots showing the effect of the MEK inhibitor U0126 on the bradykinin-induced and 11,12-EET–induced phosphorylation of Cx43 in human umbilical vein endothelial cells. Figure 7. Western blots showing the effect of the MEK inhibitor U0126 on the bradykinin-induced and 11,12-EET–induced phosphorylation of Cx43 in human umbilical vein endothelial cells. Confluent cultures of endothelial cells were stimulated with bradykinin (100 nmol/L) or 11,12-EET (1 μmol/L) in the absence and presence of U0126 (1 μmol/L) for the times indicated. Triton X-100–soluble cell fractions were subjected to SDS-PAGE, and Cx43 was identified by using a specific antibody. The Cx43 band can be separated into the nonphosphorylated protein (NP) and 2 phosphorylated forms (P1 and P2). To compare the time course of the Cx43 mobility shift with the activation of ERK1/2, each blot was reprobed with a specific antibody recognizing the phosphorylated form of ERK1/2 (pERK1/2) as well as total ERK1/2 protein. Identical results were obtained in 2 additional experiments. Rüdiger Popp et al. Circ Res. 2002;90:800-806 Copyright © American Heart Association, Inc. All rights reserved.