Ethyl pyruvate modulates acute inflammatory reactions in human endothelial cells in relation to the NF-kappaB pathway

Abstract

Background and purpose: Endothelial cell activation plays a critical role in regulating leukocyte recruitment during inflammation and infection. Ethanol (EtOH) reduces host defence systems, including cell adhesion. However, well-known side effects of EtOH limit its clinical use as an anti-inflammatory drug. Instead, ethyl pyruvate (EtP) may represent a better alternative. Here, we compared effects of EtP and EtOH on neutrophil recruitment and activation of human umbilical vein endothelial cells (HUVECs).

Experimental approach: Adhesion of neutrophils to HUVEC monolayers, surface expression of intercellular cell adhesion molecule, E-selectin, vascular cell adhesion molecule, release of interleukin (IL)-8 and granulocyte colony-stimulating factor (G-CSF) from HUVECs were assessed as well as translocation of interleukin-1 receptor-associated kinase (IRAK-1), the nuclear factor-kappa B (NF-kappaB) subunits p50, p65 and IkappaB-alpha. NF-kappaB activation was analysed with a luciferase reporter plasmid. Cells were stimulated with IL-1beta, lipopolysaccharide (LPS) or tumour necrosis factor-alpha.

Key results: EtP was several-fold more potent than EtOH in reducing adhesion of neutrophils to activated HUVECs, generation of IL-8 or G-CSF and surface expression of the adhesion molecules. This last reaction was decreased by EtP even when added after cytokines or LPS. Translocation of IRAK-1, IkappaBalpha and the NF-kappaB p65 subunit to the HUVEC nucleus was inhibited by EtP for all stimuli, whereas the diminished p50 translocation was stimulus specific. When p65 was constitutively expressed in Cos7 cells, stimulation of an NF-kappaB-dependent reporter gene was not affected by EtP, suggesting that EtP acted upstream of gene activation.

Conclusions and implications: EtP impedes adhesive, secretory and signalling events typical of the early inflammatory response in endothelial cells, suggesting EtP as a possible treatment for acute inflammatory conditions.

Figure 1

Effect of ethyl pyruvate (EtP), ethanol (EtOH) or sodium pyruvate (NaP) on adhesion of neutrophils to plastic (a) or human umbilical vein endothelial cells (HUVECs) (b and c). (Panel a) Polymorphonuclear neutrophil granulocyte (PMN), added to an albumin-coated plastic surface were treated with the drugs (as indicated) and then stimulated with 0.1 μM of fMet-Leu-Phe (in continued absence or presence of the EtP, EtOH or NaP). After 15 min, plates were analysed for adherent PMN. (Panels b and c) HUVECs were treated with EtP, EtOH or NaP (in concentrations indicated) and then stimulated for 4 h with 100 ng of LPS per mL (b) or 5 U of IL-1β per mL (c) (in continued absence or presence of EtP, EtOH or NaP). PMN was added and adherence analysed after 15 min. The data are means±s.e.mean (n=6–36). Asterisks above columns indicate a P-value of <0.05 for comparisons with cells stimulated with agonists only.

Figure 2

Effect of ethyl pyruvate (EtP), ethanol (EtOH) or sodium pyruvate (NaP) on surface expression of intercellular cell adhesion molecule (ICAM)-1 (panel a), E-selectin (panel b) or vascular cell adhesion molecule (VCAM)-1 (panels c and d) in human umbilical vein endothelial cells (HUVECs), induced by interleukin (IL)-1β. (Panels a–c) show the results when cells were treated with EtP, EtOH or NaP (in concentrations indicated) and then stimulated with 5 U of IL-1β per mL (in continued absence or presence of EtP, EtOH or NaP) for 4 h. (Panel d) shows results when HUVECs were stimulated for 2 h with IL-1β, then EtP, EtOH or NaP was added, remaining for the next 2 h. After a total of 4 h of incubation, antibodies were added and epitope abundance assayed. The data are means±s.e.mean (n=14–15). Asterisks above columns indicate a P-value of <0.05 for comparisons with cells stimulated with agonists only.

Figure 3

Effect of ethyl pyruvate (EtP), ethanol (EtOH) or sodium pyruvate (NaP) on secretion of interleukin (IL)-8 (panel a) or granulocyte colony-stimulating factor (G-CSF) (panel b) by human umbilical vein endothelial cells (HUVECs), induced by IL-1β. Cells were treated with EtP, EtOH or NaP (in concentrations indicated) and then stimulated with 5 U of IL-1β per mL (in continued absence or presence of EtP, EtOH or NaP). After 4 h of subsequent incubation, supernatants were analysed for the cytokines. The data are means±s.e.mean (n=12–40). Asterisks above columns indicate a P-value of <0.05 for comparisons with cells stimulated with agonists only.

Figure 4

Effect of ethyl pyruvate (EtP; 5 mM), ethanol (EtOH; 85 mM) on translocations of interleukin-1 receptor-associated kinase (IRAK)-1 (a) and of nuclear factor-kappa B (NF-κB) subcomponents p65 and p50 (b and c) and of inhibitor kappa B (IκB)α (d) from the cytoplasm to the nucleus in human umbilical vein endothelial cells (HUVECs). (Panel a) shows effects on IRAK-1 after stimulation with interleukin (IL)-1β for 15 min. The micrographs insert show results of lipopolysaccharide (LPS) stimulation. (Panel b) shows effects on p65 after stimulation with LPS for 60 min. (Panel c) shows effects on p50 after stimulation with IL-1β for 60 min or with LPS for 60 min. (Panel d) shows the level of IκBα in nucleus after stimulation with EtOH or EtP and LPS for 45 min. The data are means±s.e.mean (n=5–10). Asterisks above columns indicate a P-value of <0.05 for comparisons with cells stimulated with agonists only.