Reconstituted High-Density Lipoprotein Reduces LPS-Stimulated TNFα

Abstract

Reconstituted high-density lipoprotein (rHDL), an artificial lipoprotein consisting of apolipoprotein A-I and phosphatidylcholine (1:150, molar ratios) dose-dependently reduces lipopolysaccharide (LPS)-induced tumor necrosis factor-α (TNF) production in in vitro, ex-vivo, and in-vivo model systems. In an in-vitro whole blood assay, rHDL (1 mg/ml) added concomitantly with LPS increased cellular resistence to LPS stimulation ∼1000-fold. Even with extremely high levels of LPS (10 μg/ml), rHDL, ⩾0.5 mg/ml caused >50% decrease in TNF production. Preincubation of rHDL with LPS was not required for activity. rHDL (⩾1 mg/ml) reduced TNF production by 50% even when added to cultures 2 hr after their stimulation with LPS (10 μg/ml). In an ex-vivo study, rabbits were infused with rHDL at doses of 25, 50, and 75 mg/kg. Blood was drawn and stimulated with LPS ex vivo and bioactive TNF was assessed using the L929 cytotoxicity assay. Fifteen minutes after rHDL infusion, there was a significant difference in ex-vivo-induced TNF activity between groups (750 ± 160, 170 ± 40, 80 ± 30, 60 ± 30 pg TNF/ml, for the control, 25, 50, and 75 mg/kg rHDL dose groups, respectively; P < 0.0001). The duration of ex-vivo TNF inhibition was dependent on the dose of rHDL. Even at 2 hr, rHDL showed a pronounced TNF inhibition (control: 950 ± 120 pg TNF/ml; 75 mg/kg: 140 ± 60 pg TNF/ml). Further studies showed that a prophylactic infusion of rHDL diminished LPS-induced TNF production in a rabbit endotoxemia model. Anesthetized rabbits were infused with saline or rHDL (25 or 50 mg/kg) 15 min prior to the infusion of LPS (5 μg/kg/hr for 2 hr). One hour after the beginning of LPS infusion, TNF was significantly decreased in the treatment groups (7.54 ± 1.89, 2.22 ± 1.17, 2.23 ± 0.72 ng TNF/ml for control, 25, and 50 mg/kg, respectively; P = 0.02). rHDL not only delayed TNF production, but caused a reduction in peak TNF levels (20.4 ng versus 6.7 and 4.4 ng TNF/ml in the saline recipients or 25 and 50 mg/kg groups, respectively, P = 0.058). Hypotension was also reduced. In aggregate, these data confirm rHDL ability to block LPS, modulating TNF release and the physiological sequelae associated with LPS shock. If the physiological benefit of rHDL is confirmed in clinical trials, rDHL may represent a new approach to the prohylaxis and treatment of gram-negative septic shock.