CXCL16/CXCR6 is involved in LPS-induced acute lung injury via P38 signalling

Abstract

Although several chemokines play key roles in the pathogenesis of acute lung injury (ALI), the roles of chemokine (C-X-C motif) ligand 16 (CXCL16) and its receptor C-X-C chemokine receptor type 6 (CXCR6) in ALI pathogenesis remain to be elucidated. The mRNA and protein expression of CXCL16 and CXCR6 was detected after lipopolysaccharide (LPS) stimulation with or without treatment with the nuclear factor-κB (NF-κB) inhibitor pyrrolidine dithiocarbamate (PDTC). Lung injury induced by LPS was evaluated in CXCR6 knockout mice. CXCL16 level was elevated in the serum of ALI patients (n = 20) compared with healthy controls (n = 30). CXCL16 treatment (50, 100, and 200 ng/mL) in 16HBE cells significantly decreased the epithelial barrier integrity and E-cadherin expression, and increased CXCR6 expression, reactive oxygen species (ROS) production, and p38 phosphorylation. Knockdown of CXCR6 or treatment with the p38 inhibitor SB203580 abolished the effects of CXCL16. Moreover, treatment of 16HBE cells with LPS (5, 10, 20 and 50 μg/mL) significantly increased CXCL16 release as well as the mRNA and protein levels of CXCL16 and CXCR6. The effects of LPS treatment (20 μg/mL) were abolished by treatment with PDTC. The results of the luciferase assay further demonstrated that PDTC treatment markedly inhibited the activity of the CXCL16 promoter. In conclusion, CXCL16, whose transcription was enhanced by LPS, may be involved in ROS production, epithelial barrier dysfunction and E-cadherin down-regulation via p38 signalling, thus contributing to the pathogenesis of ALI. Importantly, CXCR6 knockout or inhibition of p38 signalling may protect mice from LPS-induced lung injury by decreasing E-cadherin expression.

Keywords: 16HBE; CXCL16; NF-κB; acute lung injury; p38 signal.

Figure 1

Serum levels of CXCL16. Serum levels of CXCL16 were elevated in ALI patients as demonstrated by ELISA (***P < 0.001 vs healthy controls)

Figure 2

Effects of CXCL16 on epithelial barrier integrity, ROS production and p38 phosphorylation. 16HBE cells were exposed to CXCL16 (50, 100 and 200 ng/mL), and control cells did not receive any treatment. A, Epithelial permeability was measured by using FITC‐conjugated dextran at 0 h before and 24 h after CXCL16 treatment. B, ROS production was determined by DCFH‐DA staining and flow cytometry analysis at 24 h after CXCL16 treatment. C, The levels of CXCR6, p‐p38, p38 and E‐cadherin were detected by western blotting at 24 h after CXCL16 treatment. Representative blots and quantification from three experiments are shown (^a P < 0.05, ^aa P < 0.01, ^aaa P < 0.001 vs control cells; ^bb P < 0.01, ^bbb P < 0.001 vs cells treated with 50 ng/mL CXCL16; ^c P < 0.05, ^cc P < 0.01, ^ccc P < 0.001 vs cells treated with 100 ng/mL CXCL16)

Figure 3

CXCR6 and p38 signalling mediate the effects of CXCL16 on epithelial barrier dysfunction. A, 16HBE cells were transfected with CXCR6 siRNA (siCXCR6‐1, 2, or 3) or control siRNA (siNC) and CXCR6 expression was determined after 48 h. Representative blots and quantification from three experiments are shown (***P < 0.001 vs siNC cells). B and C, 16HBE cells were divided into four groups: Group 1, siNC; Group 2, siNC + CXCL16; Group 3, siCXCR6 + CXCL16; Group 4, siNC + SB+CXCL16. The cells in Group 3 were transfected with CXCR6 siRNA (siCXCR6), whereas cells in the other groups were transfected with control siRNA (siNC). At 24 h post‐transfection, cells in Group 2 and 3 were treated with 100 ng/mL CXCL16, whereas cells in Group 4 were treated with 1 μmol/L SB203580 and 100 ng/mL CXCL16. Epithelial permeability was measured at 0 h before and 24 h after CXCL16 treatment (B). The levels of CXCR6, p‐p38, p38, and E‐cadherin were detected at 24 h after CXCL16 treatment (C). Representative blots and quantification from three experiments are shown (^aaa P < 0.001 vs siNC; ^bbb P < 0.001 vs CXCL16 + siNC)

Figure 4

LPS promotes the secretion of CXCL16 and the intracellular expression of CXCL16/CXCR6. 16HBE cells were exposed to LPS (5, 10, 20 and 50 μg/mL) for 24 h, and control cells did not receive any treatment. CXCL16 concentration in the supernatant (A) and the protein (B) and mRNA (C) (^a P < 0.05, ^aa P < 0.01, ^aaa P < 0.001 vs control cells; ^b P < 0.05, ^bb P < 0.01, ^bbb P < 0.001 vs cells treated with 5 μg/mL LPS; ^c P < 0.05, ^cc P < 0.01, ^ccc P < 0.001 vs cells treated with 10 μg/mL LPS)

Figure 5

LPS‐mediated CXCL16/CXCR6 pathway was dependent on NF‐κB. A‐C, 16HBE cells were exposed to LPS (20 μg/mL) and NF‐κB inhibitor PDTC (10 μmol/L)/vehicle (DMSO) for 24 h. Cytosol and nuclear NF‐κB p65 (A, β‐actin and H3 as loading controls, respectively) and the protein (B) and mRNA (C) levels of CXCL16/CXCR6 were detected. D, Luciferase assays were performed in 16HBE cells treated with PDTC or DMSO (***P < 0.001 vs DMSO; ^## P < 0.05, ^### P < 0.001 vs LPS + DMSO)

Figure 6

CXCR6 knockout or SB203580 treatment inhibits LPS‐induced ALI in vivo. LPS‐induced lung injury model was established in knockout (n = 5, KO/LPS group) and WT mice (n = 5, WT/LPS group). Mice in the WT/control group (n = 5) received 0.9% NaCl. Mice in the WT/LPS + SB group (n = 5) were pretreated with SB203580 and then with LPS. HE staining (A), E‐cadherin immunohistochemistry staining (B), western blotting results of E‐cadherin (C), and CXCL16 concentration in the serum (D) are shown (***P < 0.001 vs WT; ^## P < 0.001 vs WT/LPS). E, The Kaplan‐Meier plot of survival duration are shown (n = 12)