(A) Whole-blood CXCL16 and tumor necrosis factor (TNF)-alpha messenger ribonucleic acid (mRNA) and (B) plasma protein levels were measured serially for 20 h before and 24 h after intravenous lipopolysaccharide (LPS) (3 ng/kg) in 6 healthy volunteers. Repeated-measures ANOVA revealed significant effects on CXCL16 mRNA (F = 6.89, p < 0.001) and sol-CXCL16 (F = 6.15, p < 0.001) levels.

Down-regulation of (A) CXCL16 mRNA (F = 19.4, p = 0.001) and (B) CXCL16 protein secretion (F = 71.9, p < 0.001) by aspirin (ASA) in human macrophages. Cells were pre-treated with aspirin for 2 h and then for 24 h with LPS (1 µg/ml). **p < 0.01, ***p < 0.001 by t test compared to LPS-induced CXCL16 levels. (C) Down-regulation of CXCL16 mRNA in human macrophages by the nuclear factor-kappa-B inhibitor SN50 (t test p = 0.002). Cells were pretreated with SN50 or control peptide at 100 µg/ml for 2 h, then for 24 h with LPS (1 µg/ml). (D) Induction of CXCL16 in human macrophages by adenovirus over-expression (24 h) of activated I-kappa-B kinase compared to control virus (*t test p = 0.03). Results of representative experiments with triplicate samples are expressed as mean ± (SEM). GFP = green fluorescent protein; other abbreviations as in Figure 1.

(A) Down-regulation of sol-CXCL16 protein by rosiglitazone (Rosi) in mouse macrophages (*F = 38.4, p = 0.005; p < 0.01 rosiglitazone/LPS versus LPS alone). Cells were pretreated with rosiglitazone (1 µM x 24 h) and then with LPS (1 µg/ml x 24 h). (B) LPS-dependent induction of plasma sol-CXCL16 is blocked by rosiglitazone in mice (F = 8.83, p = 0.01). LDLR–/– mice (8 per group) were treated for 7 days with rosiglitazone (30 mg/kg) or vehicle and injected intraperitoneally with LPS (200 ng/g). Abbreviations as in Figure 1.