Cholesterol efflux potential and antiinflammatory properties of high-density lipoprotein after treatment with niacin or anacetrapib

Abstract

Objective: To examine the effects of treatments with niacin or anacetrapib (an inhibitor of cholesteryl ester transfer protein) on the ability of high-density lipoprotein (HDL) to promote net cholesterol efflux and reduce toll-like receptor-mediated inflammation in macrophages.

Methods and results: A total of 18 patients received niacin, 2 g/d, for 4 weeks; 20 patients received anacetrapib, 300 mg/d, for 8 weeks; and 2 groups (n=4 and n=5 patients) received placebo. HDL samples were isolated by polyethylene glycol precipitation or ultracentrifugation, tested for the ability to promote cholesterol efflux in cholesterol-loaded THP-I or mouse peritoneal macrophages, or used to pretreat macrophages, followed by lipopolysaccharide exposure. HDL cholesterol levels were increased by 30% in response to niacin and by approximately 100% in response to anacetrapib. Niacin treatment increased HDL-mediated net cholesterol efflux from foam cells, primarily by increasing HDL concentration, whereas anacetrapib treatment increased cholesterol efflux by both increasing HDL concentration and causing increased efflux at matched HDL concentrations. The increased efflux potential of anacetrapib-HDL was more prominent at higher HDL cholesterol concentrations (>12 microg/mL), which was associated with an increased content of lecithin-cholesterol acyltransferase (LCAT) and apolipoprotein E and completely dependent on the expression of ATP binding cassette transporters (ABCA1 and ABCG1). Potent antiinflammatory effects of HDL were observed at low HDL concentrations (3 to 20 microg/mL) and were partly dependent on the expression of ABCA1 and ABCG1. All HDL preparations showed similar antiinflammatory effects, proportionate to the HDL cholesterol concentration.

Conclusions: Niacin treatment caused a moderate increase in the ability of HDL to promote net cholesterol efflux, whereas inhibition of cholesteryl ester transfer protein via anacetrapib led to a more dramatic increase in association with enhanced particle functionality at higher HDL concentrations. All HDLs exhibited potent ability to suppress macrophage toll-like receptor 4-mediated inflammatory responses, in a process partly dependent on cholesterol efflux via ABCA1 and ABCG1.

Figure 1

Dose-response curve for cholesterol efflux induced by similar concentration or volume of control and PEG-HDL post post-niacin and –anacetrapib treatment. THP-1 macrophage were treated with 50μg/mL acLDL and 3μmol/L TO901317 for 24h. Then, increased concentrations of pooled PEG-HDL (12, 36, 72μg/mL cholesterol) from niacin treatment (A) and anacetrapib treatment (B) or increased volumes of pooled control PEG-HDL (34±4mg/dL cholesterol) and niacin-PEG-HDL (45±4mg/dL cholesterol) (C) were added to media for 6h before TC, FC and CE mass analysis. Values are means±SEM of an experiment performed in triplicate. *P<0,05, significant difference vs control PEG-HDL.

Figure 2

Effect of niacin-PEG-HDL on cholesterol efflux from human THP-1 macrophage foam cells treated with 50μg/mL acLDL and 3μmol/L TO901317 for 24h. The TC, FC, and CE mass in media were determined 6h after incubation of 25μl of placebo PEG-HDL (35±4mg/dL and 35±2mg/dL cholesterol before and after treatment, respectively) (A) or niacin-PEG-HDL (34±4mg/dL and 45±4mg/dL cholesterol before and after treatment, respectively) (B) in 500μl of media. Values are means±SEM of 5 individuals determinations for niacin placebo groups and 18 individuals determinations for niacin group. *P<0,05, significant difference vs control PEG-HDL.

Figure 3

Effect of anacetrapib-PEG-HDL on cholesterol efflux from human THP-1 macrophage foam cells treated with 50μg/mL acLDL and 3μmol/L TO901317 for 24h. The TC, FC, and CE mass in media were determined 6h after incubation of similar concentration of PEG-HDL (50μg/mL cholesterol) from placebo (C) or anacetrapib group (D) before and after 2 week treatment. Values are means±SEM of 5 individuals determinations for anacetrapib placebo groups and 20 individuals determinations for anacetrapib group. **P<0.001, significant difference vs control PEG-HDL.

Figure 4

Dose response curve for cholesterol efflux induced by similar concentration or volume of ultracentrifugated control and HDL-2 post-niacin and -anacetrapib treatment in WT and Abca1^−/−Abcg1^−/− macrophages. Bone-marrow-derived macrophages from WT and Abca1^−/−Abcg1^−/− mice were treated for 16h with 50μg/mL acLDL and 2μCi/mL of [3H]-cholesterol. Then, increased concentrations of pooled HDL-2 (3, 8, 20μg/mL cholesterol) from niacin treatment (A and B) and anacetrapib treatment (E and F) or increased volumes of pooled control HDL-2 (34±4mg/dL and 38±2mg/dL cholesterol for control niacin and control anacetrapib, respectively), niacin-HDL-2 (45±4mg/dL cholesterol, C and D) and anacetrapib-HDL-2 (81±4mg/dL cholesterol, G and H) were added to 500μl media for 3h before isotopic cholesterol efflux. Control, niacin and anacetrapib-HDL significantly increased cholesterol efflux along with enhanced concentration or volume of HDL. Values are means±SEM of an experiment performed in triplicate. *P<0.05, significant difference vs control HDL-2.

Figure 5

Dose response curve for suppression of toll-like receptor 4-mediated inflammation induced by similar concentration or volume of ultracentrifugated control and niacin-HDL in WT and Abca1^−/−Abcg1^−/− macrophages. Bone-marrow-derived macrophages from WT and Abca1^−/−Abcg1^−/− mice were treated for 16h with 50μg/mL acLDL. Then, increased concentrations of pooled HDL-2 (3, 8, 20μg/mL cholesterol) from niacin treatment (A) or increased volumes of pooled control HDL-2 (34±4mg/dL cholesterol) and niaicn-HDL-2 (45±4mg/dL cholesterol) (B) were added to 500μl of 0.2%BSA media for 3h before LipidA treatment for 3h more hours as described in the Methods. At the end of the incubation, inflammatory transcript levels (TNFα, IL-6, MIP1α and MIP2) were quantified and normalized to m36B4. Results are means±SEM and expressed as arbitrary units (a.u) from an experiment performed in triplicate. *P<0,05, significant difference vs control HDL-2.

Figure 6

Dose response curve for suppression of toll-like receptor 4-mediated inflammation induced by similar concentration or volume of ultracentrifugated control and anacetrapib-HDL in WT and Abca1^−/−Abcg1^−/− macrophages. Bone-marrow-derived macrophages from WT and Abca1^−/−Abcg1^−/− mice were treated for 16h with 50μg/mL acLDL. Then, increased concentrations of pooled HDL-2 (3, 8, 20μg/mL cholesterol) from 300mg anacetrapib treatment (A) or increased volumes of pooled control HDL-2 (38±2mg/dL cholesterol) and anacetrapib-HDL-2 (81±4mg/dL cholesterol) (B) were added to 500μl of 0.2%BSA media for 3h before LipidA treatment for 3h more hours as described in the Methods. At the end of the incubation, inflammatory transcript levels (TNFα, IL-6, MIP1α and MIP2) were quantified and normalized to m36B4. Results are means±SEM and expressed as arbitrary units (a.u) from an experiment performed in triplicate. *P<0,05, significant difference vs control HDL-2.