Introduction: At the early stage of atherosclerosis, endothelial cells (ECs) are activated by inflammatory mediators mainly at sites under turbulent flow and lowered shear stress, leading to enhancement of leukocyte adhesion to ECs. Interactions between leukocytes and activated ECs plays a crucial role in atherogenesis. Gap junction molecules including connexins (Cxs) enable a direct cell-cell interaction by allowing ions and small signaling molecules to move between adjacent cells. Among Cx subtypes, Cx37 and 40 were recently suggested to contribute to atheroprotection by inhibiting lymphocyte adhesion and platelet aggregation. In contrast, the role of Cx43 in atherogenesis has not been fully understood. We sought to address whether Cx 40 and 43 are involved in atherosclerosis development.

Methods: First, immune staining was performed using antibodies against Cx40, Cx43, and CD68 in cross-sections of human carotid plaques and their localization was investigated. Then, we assessed whether the expression of Cxs in freshly isolated human umbilical vein ECs (HUVECs) was regulated by inflammatory stimuli: (HUVECs) were seeded in laminar flow-through slides and exposed to laminar shear stress (~10 dyne) or lipopolysaccharides (LPS; 1-100mg/ml) under static conditions for 24 hours. Protein levels were determined fluorometrically using immunofluorescent staining while mRNA levels were determined analyzed using quantitative reverse-transcription PCR. Furthermore, the Cx expression in THP-1 monocytic cells was assessed after LPS stimulation or PMA (phorbol 12-myristate 13-acetate)-induced differentiation using flow cytometry. Finally, Cx43 in THP-1 cells were inhibited by GAP19 incubation and dynamic adhesion assays were performed to evaluate the effect of Cx43 in THP-1 cells on their adhesion to ECs.

Results: In human carotid plaques, Cx43 and Cx40 were detected mainly in and near ECs and Cx43 was colocalized with CD68-positive macrophages. In HUVECs, expressions of Cx43 was significantly higher after 24 hours of laminar shear stress than those after static conditions, both at protein levels (P<0.01, n=9) and at mRNA levels (P<0.001, n=6). In contrast, Cx40 protein expression was not significantly different between flow and static conditions (p=0.496, n=9), although Cx40 mRNA levels were higher after 24 hours of laminar shear stress compared with static conditions (P<0.05, n=6). LPS stimulation of HUVECs for 24 hours under static conditions also increased Cx43 mRNA levels (P<0.05, n=4), while Cx40 mRNA levels were decreased (P<0.05, n=4). In THP-1 cells, LPS-treatment increased the expression of Cx40 and Cx43 (P<0.05, n=4). After inducing THP-1 cell differentiation by PMA, expression of Cx43, but not Cx40, was increased (Cx43; p=0.029, n=4). Finally, dynamic adhesion assays showed that adherence of THP-1 cells to HUVECs at turbulent shear stress sites was slightly increased when THP-1 cells were treated with a Cx43 inhibitor (p=0.095, n=5).

Conclusion: Endothelial activation by inflammatory and mechanical stress upregulated Cx43, but not Cx40 expression in HUVECs. In THP-1 cells, inflammatory stimuli and PMA induced-differentiation upregulated Cx43 expression. These results suggest that Cx43 may be involved in atherogenesis and warrant further investigation.