Inhibition of Tumor Necrosis Factor α–Stimulated Monocyte Adhesion to Human Aortic Endothelial Cells by AMP-Activated Protein Kinase
Objective— Proatherosclerotic adhesion of leukocytes to the endothelium is attenuated by NO. As AMP-activated protein kinase (AMPK) regulates endothelial NO synthesis, we investigated the modulation of adhesion to cultured human aortic endothelial cells (HAECs) by AMPK.
Methods and Results— HAECs incubated with the AMPK activator, AICAR, or expressing constitutively active AMPK demonstrated reduced TNFα-stimulated adhesion of promonocytic U-937 cells. Rapid inhibition of TNFα-stimulated U-937 cell adhesion by AICAR was NO-dependent, associated with unaltered cell surface adhesion molecule expression, and reduced MCP-1 secretion by HAECs. In contrast, inhibition of TNFα-stimulated U-937 cell adhesion by prolonged AMPK activation was NO-independent and associated with reduced cell surface adhesion molecule expression.
Conclusions— AMPK activation in HAECs inhibits TNFα-stimulated leukocyte adhesion by a rapid NO-dependent mechanism associated with reduced MCP-1 secretion and a late NO-independent mechanism whereby adhesion molecule expression, in particular E-selectin, is suppressed.
Recent studies have demonstrated AMPK-activated protein kinase (AMPK) to be a key regulator of NO synthesis in cultured endothelial cells in response to stimuli including AICAR.1,2 The functional effects of AMPK-mediated NO production in the endothelium remain poorly characterized, however. Endothelial NO inhibits leukocyte adhesion and the expression of intercellular adhesion molecule-1 (ICAM-1), vascular cell adhesion molecule (VCAM)-1, and E-selectin in response to proinflammatory stimuli such as TNFα.3 AICAR has recently been reported to reduce TNFα-stimulated monocyte adhesion and mRNA expression of ICAM-1, VCAM-1, and E-selectin in cultured endothelial cells and postischemic leukocyte rolling and adhesion in mice.4,5 However, these studies did not address whether the effects of AICAR are mediated by NO or AMPK, as AICAR has been demonstrated to have AMPK-independent effects.6 We have, therefore, investigated the molecular mechanisms that underlie reduced monocyte adhesion in response to AICAR.
Cell Systems and Infection
HAECs, cultured as described previously,2 were infected with recombinant adenoviruses expressing constitutively active (Ad.AMPK-CA) mutant AMPK, dominant negative (Ad.AMPK-DN) mutant AMPK, or control adenoviruses (Ad.control) as described previously.2
Monocyte Adhesion and Chemokine Secretion Assay
To ensure that observed effects of AMPK activation were specific to HAECs, HAEC monolayers were washed thoroughly with serum-free medium before overlay with U-937 cells or chemokine assay. Adhesion was assessed microscopically and chemokine secretion by immunoassay.
Cell surface E-Selectin, ICAM-1, and VCAM-1 expression were quantified using anti–VCAM-1, anti–ICAM-1, or anti–E-Selectin antibodies and fluorescein isothiocyanate (FITC)- or phycoerythrin-labeled secondary antibodies.
For details and complete figure legends, see the supplemental materials (available online at http://atvb.ahajournals.org).
Incubation of HAECs with AICAR reduced TNFα-stimulated U-937 cell adhesion in a time- and dose-dependent manner without affecting basal U-937 cell adhesion (Figure 1A and 1C; supplemental Figure I) or HAEC viability as assessed by Annexin V staining (data not shown). Infection with Ad.AMPK-CA completely abrogated TNFα-stimulated U-937 cell adhesion, without altering basal adhesion (Figure 1B). Preincubation of HAECs with the eNOS inhibitor, L-NAME, abrogated the rapid (45 to 90 minutes) inhibition of TNFα-stimulated U-937 cell adhesion by AICAR, but had no effect on inhibition of adhesion in response to long term (120 to 360 minutes) AICAR treatment or Ad.AMPK-CA (Figure 1A and 1B). Similar effects were observed after siRNA-mediated downregulation of eNOS (supplemental Figure II). Infection with Ad.AMPK-DN attenuated AMPK activity (supplemental Figure III) and the inhibition of TNFα-stimulated U-937 cell adhesion by AICAR (Figure 1C).
Incubation of HAECs with AICAR (240 minutes) or infection with Ad.AMPK-CA inhibited TNFα-stimulated cell surface expression of ICAM-1, VCAM-1, and E-selectin, yet incubation with AICAR for 45 minutes did not affect adhesion molecule expression (Figure 2A). Infection with Ad.AMPK-DN reversed AICAR-mediated inhibition of E-selectin expression (Figure 2B) and partially reversed inhibition of ICAM-1 and VCAM-1 expression (supplemental Figure IV). L-NAME did not affect expression of adhesion molecules (Figure 2B and supplemental Figure IV).
As acute incubation with AICAR reduced U-937 cell adhesion independent of HAEC adhesion molecule expression, we reasoned AICAR may inhibit chemokine secretion by HAECs. Incubation of HAECs with AICAR for 45 minutes significantly reduced TNFα-stimulated MCP-1 secretion, an effect partially abrogated by coincubation of HAECs with L-NAME (Figure 2C) or infection with Ad.AMPK-DN (Figure 2D).
The central finding of this study is that incubation of HAECs with AICAR markedly reduces TNFα-stimulated monocyte adhesion in a biphasic and AMPK-dependent manner. These data suggest that the rapid adhesion molecule–independent effects of AICAR are mediated by AMPK-stimulated NO synthesis reducing secretion of MCP-1 and potentially secretion of other chemokines by HAECs in response to TNFα. Prolonged incubation with AICAR reduces TNFα-stimulated E-selectin expression in an AMPK-dependent NO-independent manner, whereas AICAR reduces TNFα-stimulated ICAM-1 and VCAM-1 expression by a mechanism only partially dependent on AMPK. A recent study reported that AICAR reduced postischemic leukocyte adhesion in an eNOS-independent manner, yet leukocyte rolling in response to prolonged AICAR treatment was eNOS-dependent.5 The differences between the current study and these data may reflect the different species studied and lack of assessment of the AMPK-dependence and site of action of AICAR in vivo.
The evidence presented in this study clearly implicates AMPK activation as a potential antiatherogenic mechanism. Atherosclerosis is commonly associated with type 2 diabetes and insulin resistance, and we propose that AMPK, in addition to its potential to correct the metabolic defects present in type 2 diabetes and insulin resistance,7 is an attractive candidate therapeutic target for reducing associated atherogenesis.
Sources of Funding
This work was supported by the British Heart Foundation (04/051/17026), a Wellcome Trust PhD Studentship (C.F.K), and Diabetes UK R.D. Lawrence Fellowship (to I.P.S).
Original received June 20, 2008; final version accepted September 3, 2008.
Morrow VA, Foufelle F, Connell JMC, Petrie JR, Gould GW, Salt IP. Direct activation of AMP-activated protein kinase stimulates nitric oxide synthesis in human aortic endothelial cells. J Biol Chem. 2003; 278: 31629–31639.
De Caterina R, Libby P, Peng HB, Thannickal VJ, Rajavashisth TB, Gimbrone MA Jr, Shin WS, Liao JK. Nitric oxide decreases cytokine-induced endothelial activation. Nitric oxide selectively reduces endothelial expression of adhesion molecules and proinflammatory cytokines. J Clin Invest. 1995; 96: 60–68.
Gaskin FS, Kamada K, Yusof M, Korthuis RJ. 5′-AMP-activated protein kinase activation prevents postischemic leukocyte-endothelial cell adhesive interactions. Am J Physiol Heart Circ Physiol. 2007; 292: H326–H332.
Towler MC, Hardie DG. AMP-activated protein kinase in metabolic control and insulin signaling. Circ Res. 2007; 100: 328–341.