Hepatic Overexpression of Endothelial Lipase Lowers High-Density Lipoprotein but Maintains Reverse Cholesterol Transport in Mice: Role of Scavenger Receptor Class B Type I/ATP-Binding Cassette Transporter A1-Dependent Pathways.

OBJECTIVE: Reverse cholesterol transport (RCT) is a major mechanism by which HDL (high-density lipoprotein) protects against atherosclerosis. Endothelial lipase (EL) reportedly reduces HDL levels, which, in theory, would increase atherosclerosis. However, it remains unclear whether EL affects RCT in vivo. APPROACH AND RESULTS: Adenoviral vectors expressing EL or luciferase were intravenously injected into mice, and a macrophage RCT assay was performed. As expected, hepatic EL overexpression markedly reduced HDL levels. In parallel, plasma 3H-cholesterol counts from the EL-expressing mice decreased by 85% compared with control. Surprisingly, there was no difference in fecal 3H-cholesterol excretion between the groups. Kinetic studies revealed increased catabolism/hepatic uptake of 3HDL-cholesteryl ether, resulting in no change in fecal HDL-cholesteryl ester excretion in the mice. To explore underlying mechanisms for the preservation of RCT despite low HDL levels in the EL-expressing mice, we investigated the effects of hepatic SR-BI (scavenger receptor class B type I) knockdown. RCT assay revealed that knockdown of SR-BI alone reduced fecal excretion of macrophage-derived 3H-cholesterol. Interestingly, hepatic EL overexpression under SR-BI inhibition further attenuated fecal tracer counts as compared with control. Finally, we observed that EL overexpression enhanced in vivo RCT under pharmacological inhibition of hepatic ABCA1 (ATP-binding cassette transporter A1) by probucol. CONCLUSIONS: Hepatic EL expression compensates for reduced macrophage-derived cholesterol efflux to plasma because of low HDL levels by promoting cholesterol excretion to bile/feces via an SR-BI pathway, maintaining overall RCT in vivo. In contrast, EL-modified HDL might negatively regulate RCT via hepatic ABCA1. Despite extreme hypoalphalipoproteinemia, RCT is maintained in EL-expressing mice via SR-BI/ABCA1-dependent pathways.

Hepatic overexpression of idol increases circulating protein convertase subtilisin/kexin type 9 in mice and hamsters via dual mechanisms: sterol regulatory element-binding protein 2 and low-density lipoprotein receptor-dependent pathways.

OBJECTIVE: Low-density lipoprotein receptor (LDLR) is degraded by inducible degrader of LDLR (Idol) and protein convertase subtilisin/kexin type 9 (PCSK9), thereby regulating circulating LDL levels. However, it remains unclear whether, and if so how, these LDLR degraders affect each other. We therefore investigated effects of liver-specific expression of Idol on LDL/PCSK9 metabolism in mice and hamsters. APPROACH AND RESULTS: Injection of adenoviral vector expressing Idol (Ad-Idol) induced a liver-specific reduction in LDLR expression which, in turn, increased very-low-density lipoprotein/LDL cholesterol levels in wild-type mice because of delayed LDL catabolism. Interestingly, hepatic Idol overexpression markedly increased plasma PCSK9 levels. In LDLR-deficient mice, plasma PCSK9 levels were already elevated at baseline and unchanged by Idol overexpression, which was comparable with the observation for Ad-Idol-injected wild-type mice, indicating that Idol-induced PCSK9 elevation depended on LDLR. In wild-type mice, but not in LDLR-deficient mice, Ad-Idol enhanced hepatic PCSK9 expression, with activation of sterol regulatory element-binding protein 2 and subsequently increased expression of its target genes. Supporting in vivo findings, Idol transactivated PCSK9/LDLR in sterol regulatory element-binding protein 2/LDLR-dependent manners in vitro. Furthermore, an in vivo kinetic study using (125)I-labeled PCSK9 revealed delayed clearance of circulating PCSK9, which could be another mechanism. Finally, to extend these findings into cholesteryl ester transfer protein-expressing animals, we repeated the above in vivo experiments in hamsters and obtained similar results. CONCLUSIONS: A vicious cycle in LDLR degradation might be generated by PCSK9 induced by hepatic Idol overexpression via dual mechanisms: sterol regulatory element-binding protein 2/LDLR. Furthermore, these effects would be independent of cholesteryl ester transfer protein expression.

Retinoic acid receptor agonists regulate expression of ATP-binding cassette transporter G1 in macrophages.

ABC transporter G1 (ABCG1) plays a pivotal role in HDL-mediated cholesterol efflux and atherogenesis. We investigated whether, and how, retinoic acid receptors (RARs) regulate ABCG1 expression in macrophages. All-trans retinoic acid (ATRA), an RAR ligand, increased ABCG1 protein levels and apoA-I/HDL-mediated cholesterol efflux from the macrophages. Both ATRA and other RAR agonists, TTNPB and Am580, increased major transcripts driven by promoter B upstream of exon 5, though minor transcripts driven by promoter A upstream of exon 1 were only increased by ATRA. The stimulatory effects of ATRA on ABCG1 expression were completely abolished in the presence of RAR/RXR antagonists but were only partially canceled in the presence of an LXR antagonist. Adenovirus with overexpressed oxysterol sulfotransferase abolished the LXR pathway, as previously reported, and ATRA-responsiveness in ABCA1/ABCG1 expressions were respectively attenuated by 38 and 22% compared to the control virus. Promoter assays revealed that ABCG1 levels were regulated more by promoter B than promoter A, and ATRA activated promoter B in a liver X receptor-responsive element (LXRE)-dependent manner. Further, LXRE-B in intron 7, but not LXRE-A in intron 5, enhanced ATRA responsiveness under overexpression of all RAR isoforms-RARα/ß/γ. In contrast, the activation of promoter B by TTNPB depended on LXRE-B and RARα, but not on RARß/γ. Finally, chromatin immunoprecipitation and gel-shift assays revealed a specific and direct repeat 4-dependent binding of RARα to LXRE-B. In conclusion, RAR ligands increase ABCA1/G1 expression and apoA-I/HDL-mediated cholesterol efflux from macrophages, and modulate ABCG1 promoter activity via LXRE-dependent mechanisms.

Pioglitazone enhances cholesterol efflux from macrophages by increasing ABCA1/ABCG1 expressions via PPARγ/LXRα pathway: findings from in vitro and ex vivo studies.

OBJECTIVE: Pioglitazone, a peroxisome proliferator-activated receptor γ (PPARγ) agonist, reportedly reduces cardiovascular events in diabetic patients. ATP cassette binding transporters (ABC) A1 and G1 are pivotal molecules for cholesterol efflux (ChE) from macrophages and high density-lipoprotein biogenesis, and the A1 transporter is regulated by a PPARγ-liver receptor X (LXR) pathway. Also, pioglitazone induces ABCG1 expression, though the exact mechanism remains unclear. We therefore investigated the effects of pioglitazone on ABCA1/G1 expression in vitro and ex vivo. METHODS: The effects of pioglitazone on ChE and ABCA1/G1 expressions in macrophages were assessed. Then, mRNA was quantified in macrophages when PPARγ/LXR inhibition by siRNA or overexpression of oxysterol sulfotransferase was performed. ABCA1/G1 promoter activity with mutated LXR-responsive elements was also measured. As an ex vivo study, 15 type 2 diabetic patients were administered pioglitazone or placebo, and ChE assays and protein expressions were determined using macrophages cultured with the corresponding sera. RESULTS: Pioglitazone increased LXRα/ABCA1/G1 expressions, which enhanced ChE from macrophages. Inhibition of PPARγ/LXR pathways revealed that LXR was primarily involved in pioglitazone's transactivation of ABCA1 but only partially involved for ABCG1. Promoter assays showed that ABCG1 was regulated more by the promoter in intron 4 than that upstream of exon 1 but both promoters were responsive to LXR activation. Sera obtained after pioglitazone treatment promoted ChE and ABCA1/G1 expressions in macrophages. CONCLUSION: Pioglitazone enhanced ChE from macrophages by increasing ABCA1/G1 in LXR-dependent and -independent manners. Our comparable in vitro and ex vivo results shed new light on pioglitazone's novel anti-atherogenic property.

Proteasomal inhibition promotes ATP-binding cassette transporter A1 (ABCA1) and ABCG1 expression and cholesterol efflux from macrophages in vitro and in vivo.

OBJECTIVE: ATP-binding cassette transporter A1 (ABCA1) and ABCG1 are key molecules in an initial step of reverse cholesterol transport (RCT), a major antiatherogenic property of high-density lipoprotein (HDL). The ubiquitin-proteasome system (UPS) mediates nonlysosomal pathways for protein degradation and is known to be involved in atherosclerosis. However, little is known about the effects of the UPS on these molecules and overall RCT. We therefore investigated whether UPS inhibition affects ABCA1/G1 expression in macrophages and RCT in vitro and in vivo. METHODS AND RESULTS: Various proteasome inhibitors increased ABCA1/G1 expression in macrophages, translating into enhanced apolipoprotein A-I- and HDL-mediated cholesterol efflux from macrophages. ABCA1 and ABCG1 were found to undergo polyubiquitination in the macrophages and HEK293 cells overexpressing these proteins, and pulse-chase analysis revealed that proteasome inhibitors inhibited ABCA1/G1 protein degradation. In in vivo experiments, the proteasome inhibitor bortezomib increased ABCA1/G1 protein levels in mouse peritoneal macrophages, and RCT assays showed that it significantly increased the fecal (54% increase compared with saline) and plasma (23%) appearances of the tracer derived from intraperitoneally injected (3)H-cholesterol-labeled macrophages. CONCLUSIONS: The present study provided evidence that the UPS is involved in ABCA1/G1 degradation, thereby affecting RCT in vivo. Therefore, specific inhibition of the UPS pathway might lead to a novel HDL therapy that enhances RCT.

Effect of sulfonylurea agents on reverse cholesterol transport in vitro and vivo.

AIM: Reverse cholesterol transport (RCT) is a critical mechanism for the anti-atherogenic property of HDL. The inhibitory effect of the sulfonylurea agent (SUA) glibenclamide on ATP binding-cassette transporter (ABC) A1 may decrease HDL function but it remains unclear whether it attenuates RCT in vivo. We therefore investigated how the SUAs glibenclamide and glimepiride affected the functionality of ABCA1/ABCG1 and scavenger receptor class B type I (SR-BI) expression in macrophages in vitro and overall RCT in vivo. METHODS: RAW264.7, HEK293 and BHK-21 cells were used for in vitro studies. To investigate RCT in vivo, 3H-cholesterol-labeled and acetyl LDL-loaded RAW264.7 cells were injected into mice. RESULTS: High dose (500µM) of glibenclamide inhibited ABCA1 function and apolipoprotein A-I (apoA-I)-mediated cholesterol efflux, and attenuated ABCA1 expression. Although glimepiride maintained apoA-I-mediated cholesterol efflux from RAW264.7 cells, like glibenclamide, it inhibited ABCA1-mediated cholesterol efflux from transfected HEK293 cells. Similarly, the SUAs inhibited SR-BI-mediated cholesterol efflux from transfected BHK-21 cells. High doses of SUAs increased ABCG1 expression in RAW264.7 cells, promoting HDL-mediated cholesterol efflux in an ABCG1-independent manner. Low doses (0.1-100 µM) of SUAs did not affect cholesterol efflux from macrophages despite dose-dependent increases in ABCA1/G1 expression. Furthermore, they did not change RCT or plasma lipid levels in mice. CONCLUSION: High doses of SUAs inhibited the functionality of ABCA1/SR-BI, but not ABCG1. At lower doses, they had no unfavorable effects on cholesterol efflux or overall RCT in vivo. These results indicate that SUAs do not have adverse effects on atherosclerosis contrary to previous findings for glibenclamide.

Cilostazol enhances macrophage reverse cholesterol transport in vitro and in vivo.

OBJECTIVE: Recent failure of an HDL-cholesterol raising strategy using a cholesteryl ester transfer protein inhibitor highlights the importance of the anti-atherogenic function rather than plasma concentration of HDL. Cilostazol, a selective inhibitor of phosphodiesterase 3, has been widely used in patients with atherosclerotic diseases and is known to increase HDL-cholesterol. However, it remains unclear whether cilostazol enhances anti-atherogenic properties by promoting reverse cholesterol transport (RCT), a major anti-atherogenic function of HDL. METHODS AND RESULTS: We observed that treatment of THP-1 macrophages, human monocyte-derived macrophages, and RAW264.7 cells with cilostazol increased ABCA1 and ABCG1 expression in a concentration-dependent manner, translating into enhanced apoA-I- and HDL-mediated cholesterol efflux from the macrophages. However, other cyclic AMP (cAMP)-elevating agents did not increase ABCA1 gene expression in THP-1 macrophages. Cilostazol did not change intracellular cAMP levels in THP-1 macrophages and RAW264.7 cells, and a protein kinase A (PKA) inhibitor did not affect cilostazol-induced ABCA1 and ABCG1 expression. To further investigate RCT in vivo, (3)H-cholesterol-labeled and acetyl LDL-loaded RAW264.7 cells were intraperitoneally injected into mice and the appearance of the (3)H-tracer was monitored in plasma, liver, and feces. Supporting the in vitro data, cilostazol was found to significantly increase (3)H-tracer levels in both plasma and feces. CONCLUSIONS: These findings indicate that cilostazol might provide anti-atherosclerotic effects by promoting RCT through increased ABCA1/G1 expression in macrophages.

Coffee consumption enhances high-density lipoprotein-mediated cholesterol efflux in macrophages.

RATIONALE: Association of habitual coffee consumption with coronary heart disease morbidity and mortality has not been established. We hypothesized that coffee may enhance reverse cholesterol transport (RCT) as the antiatherogenic properties of high-density lipoprotein (HDL). OBJECTIVE: This study was to investigate whether the phenolic acids of coffee and coffee regulates RCT from macrophages in vitro, ex vivo and in vivo. METHODS AND RESULTS: Caffeic acid and ferulic acid, the major phenolic acids of coffee, enhanced cholesterol efflux from THP-1 macrophages mediated by HDL, but not apoA-I. Furthermore, these phenolic acids increased both the mRNA and protein levels of ATP-binding cassette transporter (ABC)G1 and scavenger receptor class B type I (SR-BI), but not ABCA1. Eight healthy volunteers were recruited for the ex vivo study, and blood samples were taken before and 30 minutes after consumption of coffee or water in a crossover study. The mRNA as well as protein levels of ABCG1, SR-BI, and cholesterol efflux by HDL were increased in the macrophages differentiated under autologous sera obtained after coffee consumption compared to baseline sera. Finally, effects of coffee and phenolic acid on in vivo RCT were assessed by intraperitoneally injecting [(3)H]cholesterol-labeled acetyl low-density lipoprotein-loaded RAW264.7 cells into mice, then monitoring appearance of (3)H tracer in plasma, liver, and feces. Supporting in vitro and ex vivo data, ferulic acid was found to significantly increase the levels of (3)H tracer in feces. CONCLUSIONS: Coffee intake might have an antiatherogenic property by increasing ABCG1 and SR-BI expression and enhancing HDL-mediated cholesterol efflux from the macrophages via its plasma phenolic acids.

Effects of atorvastatin on nuclear magnetic resonance-defined lipoprotein subclasses and inflammatory markers in patients with hypercholesterolemia.

AIM: Information about the effects of HMG-CoA reductase inhibitor (statin) treatment on lipoprotein subclasses has been severely limited. Nuclear magnetic resonance (NMR) spectrometry has emerged as a new methodology to quantify lipoprotein subclass concentrations. In the present study, we attempted to evaluate the hypolipidemic effects of atorvastatin utilizing this method. METHODS: Twenty-six patients were administered with atorvastain 10 mg daily for 4 weeks. Lipoprotein subclasses were measured by nuclear magnetic resonance (NMR) spectroscopy. Inflammation markers, including C-reactive protein (CRP), interleukin-6 (IL-6), and monocyte chemotactic protein-1 (MCP-1), were also determined. RESULTS: Additional to a marked reduction of LDL-C (-43%), atorvastatin treatment significantly decreased TG, RLP-C, apoC-II, apoC-III, and apoE by 27%, 49%, 25%, 15%, and 28%, respectively. NMR analysis revealed marked reductions of all LDL subclasses, resulting in a significant reduction of LDL particle number as well as an increase in LDL particle size. Further, some VLDL were decreased and HDL particle size was increased by atorvastatin. Among inflammation markers, MDA-LDL and IL-6 were marginally to significantly decreased. CONCLUSION: In addition to a strong LDL-C lowering function, atorvastatin exerts beneficial effects on TG-rich lipoproteins and inflammation in hypercholesterolemic patients.