Activation of liver X receptor attenuates lysophosphatidylcholine-induced IL-8 expression in endothelial cells via the NF-κB pathway and SUMOylation

Abstract

The liver X receptor (LXR) is a cholesterol-sensing nuclear receptor that has an established function in lipid metabolism; however, its role in inflammation is elusive. In this study, we showed that the LXR agonist GW3965 exhibited potent anti-inflammatory activity by suppressing the firm adhesion of monocytes to endothelial cells. To further address the mechanisms underlying the inhibition of inflammatory cell infiltration, we evaluated the effects of LXR agonist on interleukin-8 (IL-8) secretion and nuclear factor-kappa B (NF-κB) activation in human umbilical vein endothelial cells (HUVECs). The LXR agonist significantly inhibited lysophosphatidylcholine (LPC)-induced IL-8 production in a dose-dependent manner without appreciable cytotoxicity. Western blotting and the NF-κB transcription activity assay showed that the LXR agonist inhibited p65 binding to the IL-8 promoter in LPC-stimulated HUVECs. Interestingly, knockdown of the indispensable small ubiquitin-like modifier (SUMO) ligases Ubc9 and Histone deacetylase 4 (HDAC4) reversed the increase in IL-8 induced by LPC. Furthermore, the LPC-induced degradation of inhibitory κBα was delayed under the conditions of deficient SUMOylation or the treatment of LXR agonist. After enhancing SUMOylation by knockdown SUMO-specific protease Sentrin-specific protease 1 (SENP1), the inhibition of GW3965 was rescued on LPC-mediated IL-8 expression. These findings indicate that LXR-mediated inflammatory gene repression correlates to the suppression of NF-κB pathway and SUMOylation. Our results suggest that LXR agonist exerts the anti-atherosclerotic role by attenuation of the NF-κB pathway in endothelial cells.

Keywords: NF-κB; SUMOylation; liver X receptor; lysophosphatidylcholine.

Figure 1

LPC‐induced IL‐8 expression. (A) HUVECs were treated with LPC (0, 40, 60 or 80 μM) for 3 hrs. IL‐8 was analysed by ELISA. (B) HUVECs were treated with the indicated concentrations of LPC for 3 hrs, and cell viability was determined. Cytotoxic assessment of HUVECs was measured by the CCK‐8 assay. (C) HUVECs were treated with 40 μM LPC for 0.5, 1 and 3 hrs. IL‐8 mRNA was analysed by real‐time PCR. *P < 0.05; **P < 0.01; ***P < 0.001. One‐way anova was used to compare the differences in A and B. Two‐way anova was used to compare the differences in C.

Figure 2

LXR agonist inhibited LPC‐induced IL‐8 production in HUVECs. (A and B) HUVECs were pretreated with 2.5 μM of GW3965 for 8, 12 or 24 hrs and then treated with 40 μM of LPC. IL‐8 protein and mRNA levels were analysed. (C and D) HUVECs were pretreated with GW3965 (0.05 or 2.5 μM) for 24 hrs and then treated with 40 μM of LPC. The level of the IL‐8 protein in the supernatant was measured by ELISA, and IL‐8 mRNA was analysed by real‐time PCR. (E) HUVECs were pretreated with GW3965 (0.05, 0.5, 2.5, 3, 4 or 5 μM) for 24 hrs and then treated with 40 μM of LPC for 3 hrs and cell viability was determined. (F) LXRαβ knockdown cells were under the stimulation of LPC or vehicle for 1 hr after incubation with or without GW3965, and IL‐8 gene expression was determined by real‐time PCR. *P < 0.05; **P < 0.01; ***P < 0.001, compared with the control. ^# P < 0.05; ^## P < 0.01; ^### P < 0.001, compared with the LPC‐treated group. One‐way anova was used to compare the differences.

Figure 3

LXR agonist reversed LPC‐induced NF‐κB activation and inhibited IL‐8 translation. (A) HUVECs were treated with vehicle or 2.5 μM of GW3965 for 24 hrs and then stimulated with PBS or LPC (40 μM) for 5 min. Relative expression of p‐IKKα was normalized to IKKα, p‐IKKβ was normalized to IKKβ and the expression of p‐IκBα was normalized to IκBα in HUVECs. (B) Cells were processed for Immunofluorescence staining to assess the translocation of NF‐κB p65 into the nucleus. (C) Nuclear proteins were extracted for the NF‐κB transcription factor activity assay. (D and E) HUVECs were cotransfected with NF‐κB luciferase reporter plasmid pNF‐κB‐TA‐luc or ‐162 IL‐8 reporter construct, together with the Renilla luciferase reporter as an internal control. Transfected cells were treated with vehicle or 2.5 μM of GW3965 for 24 hrs and then stimulated with PBS or 40 μM of LPC for 3 hrs. The Firefly luciferase activities were analysed by the dual‐luciferase assay and normalized to Renilla luciferase expression. (F) HUVECs were transfected with either the empty pGL3‐Basic plasmid, the ‐162 IL‐8 reporter construct or IL‐8 promoter construct containing mutated AP‐1 or NF‐κB sites (ΔAP‐1 or ΔNF‐κB). The cells were treated with or without 40 μM of LPC for 3 hrs. **P < 0.01; ***P < 0.001, compared with the control. ^## P < 0.01; ^### P < 0.001, compared with the LPC‐treated group. ^† P < 0.05 compared with the control. One‐way anova was used to compare the differences.

Figure 4

LXR agonist antagonized LPC‐induced IL‐8 expression by inhibiting SUMOylation. (A) HUVECs were transfected with siUbc9, siHDAC4 or siControl under the stimulation of LPC (40 μM) for 1 hr. IL‐8 expressed was assessed by real‐time PCR. (B) Cells were stimulated by LPC for 0, 15, 30 and 60 hrs after transfecting with siUbc9 or siControl, and IκB levels were detected by immunoblotting. (C) HUVECs were stimulated by LPC for 0, 15, 30 and 60 hrs after treating with Vehicle or LXR agonist for 24 hrs and IκB levels were detected by Western Blotting. (D) SENP1 knockdown cells were under the stimulation of LPC or vehicle for 1 hr after incubation with or without GW3965. **P < 0.01; ***P < 0.001, compared with the control. ^# P < 0.01; ^## P < 0.01 compared with the LPC‐treated group. One‐way anova was used to compare the differences.

Figure 5

LXR agonist reversed LPC‐induced monocyte adhesion via an IL‐8 dependent mechanism in HUVECs. (A) An in vitro adhesion assay was performed to measure THP‐1 adhesion on vehicle‐stimulated or LPC‐stimulated HUVECs with or without pretreatment with GW3965 for 24 hrs. An IL‐8 neutralizing antibody was used to confirm the role of IL‐8 in THP‐1 adhesion induced by LPC. (B) Data shown were pooled from five independent experiments. ***P < 0.001, compared with the control. ^## P < 0.01, compared with the LPC‐treated group. One‐way anova was used to compare the differences.