Journal Logo

Original Article

Atorvastatin Inhibits ABCA1 Expression and Cholesterol Efflux in THP-1 Macrophages by an LXR-dependent Pathway

Qiu, Guosong PhD; Hill, John S PhD

Author Information
Journal of Cardiovascular Pharmacology: April 2008 - Volume 51 - Issue 4 - p 388-395
doi: 10.1097/FJC.0b013e318167141f
  • Free

Abstract

INTRODUCTION

Macrophages, in particular, foam cells, have been purported to have a role at early and late stages of the development of atherosclerotic lesions.1 Foam cell development is thought to arise from an imbalance between lipid uptake and efflux. Multiple mechanisms are involved in lipid efflux, including free diffusion, membrane adenosine triphosphate (ATP)-binding cassette (ABC) transporters, and lipoprotein receptors.2

The importance of the ABC transporter ABCA1 is underscored by the marked accumulation of lipid in peripheral tissues observed in Tangier disease, in which mutant ABCA1 is incapable of inducing cholesterol efflux.3-5 Furthermore, ABCA1 mutations in humans are also positively correlated with aortic intima thickness.6 ABCA1-deficient macrophages in particular have been shown to promote atherosclerosis, whereas human ABCA1 transgenic mice [BAC(+)] with apoE(−/−) genetic background developed dramatically smaller, less-complex lesions as compared with their apoE(−/−) counterparts.7 Thus, ABCA1 has become a promising therapeutic target for the treatment of atherosclerosis.

Statins, sharing an analogous structural moiety with 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA), are potent competitive inhibitors of HMG-CoA reductase so as to decrease endogenous cholesterol synthesis. The beneficial actions of statins are not restricted to their lipid-lowering effects; they also extend to a variety of so-called pleiotropic effects.8 Statin's nonlipid-lowering effects are largely mediated by their action on intracellular signaling molecules. Two intermediates for de novo cholesterol synthesis, oxysterols and isoprenoids [such as farnesylpyrophosphate (FPP) and geranylgeranylpyrophosphate (GGPP)], regulate the activities of nuclear receptor-like liver X receptor (LXR) and small G proteins (Rho, Rac, Rab, and Rap), respectively. The inhibition of oxysterols and isoprenoids by statins therefore affects signaling pathways and downstream gene expression.

Conflicting results have been reported for the effects of atorvastatin on ABCA1 expression in macrophages. Wong et al reported that statins decrease the expression of ABCA1 and cholesterol efflux via oxysterols-LXRα pathway.9 However, Argmann et al report an increase in macrophage ABCA1 messenger ribonucleic acid (mRNA) following atorvastatin treatment in acetylated LDL (acLDL)-loaded THP-1 macrophages through RhoA inhibition.10 Given the differences observed in the studies described thus far, we attempt to clarify the effect of atorvastatin on ABCA1 expression and cholesterol efflux and mechanisms with regard to both Rho and LXRα signaling pathways in macrophages.

METHODS

Reagents

Atorvastatin was provided by Pfizer; Mevalonate (MEV, M4667), farnesylpyrophosphate (FPP, F6892), and lysophosphatidic acid (LPA, L7260) were purchased from Sigma Aldrich, Clostridium botulinum exoenzyme C3 (Exo-C3, G-130, BioMol) was used to inhibit Rho proteins. T01901317 (575310, CalBiochem) and 22(R)-hydroxycholesterol (HO-Chol, 89355, Cayman Chemical) were used as synthetic and natural activators of LXRα, respectively.

Cell Culture and Treatment

THP-1 monocytes (American Type Culture Collection) were cultured in RPMI 1640 supplemented with 10% fetal bovine serum, 1% antibiotics-antimycotics (Invitrogen), 1 mM sodium pyruvate, and 1.5% sodium bicarbonate at 37°C, 95% air and 5% CO2, and used within 20 passages for experiments. Phorbol 12-myristate 13-acetate (PMA; P8139 from Sigma Aldrich) was added at a final concentration of 100 nM for 48 hours to differentiate THP-1 monocytes into macrophages. Human monocytes were purchased from the Lonza Corporation. Cells were cultured in RPMI 1640 medium with 10% FBS, 1% antibiotics-antimycotics, 1 mM sodium pyruvate, 1.5% sodium bicarbonate, and 10 ng/ml M-CSF (R & D Systems) for 8 days. Atorvastatin was then added to cultured macrophages at various concentrations. Macrophages were treated with Rho activators (200 μM of MEV, 20 μg/mL of FPP, 2.5 μM of LPA), Rho inhibitor (5 units/ml of Exo-C3), or LXRα activators (1 μM of T0901317 and 2.5 μM of HO-Chol) in the absence or presence of atorvastatin.

Cholesterol Efflux

THP-1 monocytes (5 × 105 cells) were seeded into 12-well plates and differentiated by PMA treatment for 48 hours. Macrophages were subsequently treated with atorvastatin and PMA for an additional 24 hours prior to cholesterol labeling. Thereafter, 3H-cholesterol (Amersham) in basal medium (RMPI 1640 with 100 nM PMA, 0.2% BSA and 1% antibiotics-antimycotics) was added into macrophage cell culture at a final concentration of 1 μCi/mL for 24 hours. Then, 3H-cholesterol-labeled macrophages were replenished with basal medium for 2-hour equilibration, and cholesterol efflux was thereafter induced in efflux medium (10 μg/mL of apoAI into basal medium) for 4 hours. Efflux medium (medium cholesterol; 200 μL) and isopropanol extract of air-dried cells (cellular cholesterol) were counted for radioactivity. Cholesterol efflux was represented as the percentage of medium cpm of total cpm (medium + cellular cpm). The apoAI-mediated cholesterol efflux is calculated as the total cholesterol efflux in the presence of apoAI minus basal cholesterol efflux in the absence of apoAI. Besides, apoAI-mediated cholesterol efflux was repeated in acLDL-transformed foam cells following the procedure of Argmann.10 Foam cell transformation was implemented by incubation with 100 μg/ml acetylated LDL and 3H-cholesterol (1 μCi/mL) in basal medium for 24 hours.

Rho Activation by Guanosine-5′-O-(γ-thio)-triphosphate

Monocytes (5 × 105 cells) were seeded in a 12-well plate and differentiated into macrophages by the stimulation with 100 nM PMA for 3 days. Thereafter, differentiated macrophages were washed twice with buffer A (100 mM KCl, 5.6 mM glucose, 1 mg/mL bovine serum albumin, 1.3 mM CaCl2, 1.3 mM CaCl2, 2 mM EGTA, 0.1 mM MgCl2, 1 mM ATP, 10 mM Hepes, pH 7.2). Permeabilization was then undertaken in the presence of 0.5 IU/mL streptolysin in buffer A for 10 minutes; guanosine-5′-O-(γ-thio)-triphosphate (GTPγS, 20-176, Upstate) was also added to a final concentration of 100 uM during permeabilization. After washing 2 times with buffer A, cells were recultured in RPMI 1640 medium for various times, and mRNA was extracted for quantitative reverse transcriptase-polymerase chain reaction (qRT-PCR).

RNA Extraction and Real-Time qRT-PCR

RNA was extracted using RNeasy plus mini kit (74134, Qiagen). Real-time qRT-PCR was performed using QuantiTectTM Probe RT-PCR kit (204443, Qiagen), and target primer set (18S RNA, Hs99999901_s1, and ABCA1, Hs00194045_mL, Applied Biosystem), and 1 μL of RNA sample on ABI Prism® 7900 platform with the following settings: 50°C for 30 min, 95°C for 15 min, and then 40 cycles of 95°C for 15 seconds and 60°C for 60 seconds. ABCA1 mRNA was normalized for 18S RNA.

Rho Pull-down Assay

A Rho activation assay kit (17-294, Upstate) was used to perform a Rho pull-down assay, which identifies RhoA, RhoB, and RhoC. Briefly, 2 × 107 macrophage cells under different treatment conditions were lysed in Mg2+ lysis/wash (MLB) buffer on ice for 15 minutes after 2 washes of ice-cold tris-buffered saline. Cell lysates were then collected after removing cell debris by centrifugation. A GTPase protein binding domain/agarose slurry (30 μL) was added to cell lysates to bind Rho protein and incubated for 45 minutes at 4°C with gentle agitation. Rho-bound agarose beads were pelleted by brief centrifugation (10 second, 14,000 × g, 4°C), washed with 1 mL of MLB 3 times, then resuspended in 40 μL of 2× Laemmli reducing sample buffer plus 2 μL of 1M dithiothreitol, and then boiled for 5 minutes prior to gel electrophoresis and Western blotting.

Western Blotting

Cell lysates (20 μg) and Rho/agarose mixture were loaded and electrophoresed on NuPAGE gels (3%-8% Tris-Acetate gel for ABCA1 blot, and 10% Bis-Tris gel for LXR and Rho blots) for 45 to 60 minutes at 200 volts. Gels were blotted onto polyvinylidene flouride (PVDF) membranes for 1.5 hours at 100 volts. After 1 hour blocking with Superblock buffer (Pierce), PVDF membranes were incubated with primary antibody (1:1000 dilution) of mouse anti-ABCA1 monoclonal antibody (ab18180, Abcam), rabbit LXR polyclonal antibody (NB 400-157, Novus Biologicals), or mouse Rho antibody (05-778, Upstate) overnight at 4°C, and then incubated with horseradish peroxidase (HRP)-conjugated mouse or rabbit secondary antibody (1:1000 dilution, Pierce) for 1 hour after 3 vigorous washes. Blots were developed by chemiluminescence using Supersignal west femto maximum sensitivity substrate (Pierce) and quantitated using ChemiGenius2 imaging system with normalization for internal control.

RESULTS

Atorvastatin Decreases ApoAI-mediated Cholesterol Efflux and ABCA1 Expression in THP-1 Macrophages

When atorvastatin was used at concentrations ranging from 2 μM to 40 μM, we observed a dose-dependent inhibitory effect on apoAI-mediated cholesterol efflux with decreases of 25% and 43% at 20 μM and 40 μM, respectively (Fig. 1A). Corresponding with decreases in cholesterol efflux, atorvastatin was also associated with a dose-dependent decrease in ABCA1 protein as observed on Western blotting (Fig. 1B), with ABCA1 protein reductions of >60% at 20-μM and >75% at 40-μM concentrations. RT-PCR analysis showed a decrease of 55% in ABCA1 mRNA level at 20 μM atorvastatin (Fig. 1C). The dose-dependent reduction of ABCA1 mRNA was also observed in human monocyte-derived macrophages, with a ~50% reduction beyond the concentration of 5 μM (Fig. 1D).

FIGURE 1
FIGURE 1:
Atorvastatin decreases apo AI-mediated cholesterol efflux and ABCA1 expression in human macrophages. Cholesterol efflux was conducted using 10 ug/mL of apoAI in3 H-cholesterol labeled macrophages for 4 hours. Cell lysates and mRNA were collected for Western blot and real-time qRT-PCR, respectively. Panel A, ApoAI-mediated cholesterol efflux in THP-1 macrophages after 24-hour atorvastatin treatment in a range from 2 μM to 40 μM (n = 6 from 2 independent experiments). Panel B, THP-1 macrophage ABCA1 protein level as shown on Western blot after the treatment of 0-40 μM atorvastatin (representative of 3 independent experiments). Panel C, Real-time qRT-PCR results of ABCA1 mRNA in THP-1 macrophages at 20 uM atorvastatin (n = 6 from 3 independent experiments, ***P < 0.001). Panel D, Real-time qRT-PCR results of ABCA1 mRNA in monocyte-derived macrophages at 0-40 uM atorvastatin (n = 4 from 2 independent experiments, *P < 0.05, **P < 0.01).

Atorvastatin Decreases LXRα and Rho Activity in Macrophages

LXR and the small G protein Rho are 2 targets that are influenced by cholesterol derivatives and intermediate metabolites.11,12 Thus, we examined the effect of atorvastatin on LXRα expression and Rho protein activities using immunoprecipitation and Rho-GTPase binding protein pull-down assay, respectively. An increasing concentration of atorvastatin was associated with decreases in LXRα protein, with a reduction of ~60% at 20 μM and ~70% at 40 μM (Fig. 2A). As expected, the level of activated Rho protein was also suppressed by atorvastatin in a dose-dependent manner, with a decrease of >80% at 20 μM atorvastatin (Fig. 2B).

FIGURE 2
FIGURE 2:
Atorvastatin decreases LXRα and Rho activity in THP-1 macrophages. Macrophages were differentiated by 100 nM PMA for 2 days prior to 24-hour atorvastatin treatment at various concentrations. Panel A, Western blot of LXR (representative of 3 independent experiments). Panel B, Rho pull-down assay: Activated Rho protein was suppressed dose-dependently by atorvastatin (representative of 2 independent experiments).

ABCA1 Expression and ApoAI-mediated Cholesterol Efflux After Rho Activation and Inhibition

Because Rho protein and ABCA1 were suppressed following atorvastatin treatment, the question was raised as to whether atorvastatin decreased ABCA1 expression through Rho inhibition. To test this hypothesis, Rho activators mevalonate (200 μM), FPP (20 μg/mL), and LPA (2.5 μM) were used in an attempt to salvage apoAI-mediated cholesterol efflux and ABCA1 expression after atorvastatin treatment.

In the presence of atorvastatin (20 μM), Rho activators were not able to mediate any improvements in apoAI-mediated cholesterol efflux, which was even further decreased by 18% (P < 0.01) with FPP treatment (Fig. 3A). Through the analysis of ABCA1 mRNA and protein, only mevalonate treatment partially restored ABCA1 mRNA (37%) and protein (40%) levels in the presence of atorvastatin (Fig. 3B and 3C). Rho activation by FPP did not exert appreciable effects on either ABCA1 mRNA or protein levels, whereas LPA treatment further decreased ABCA1 mRNA, although the protein level was not significantly changed (Fig. 3B, 3C).

FIGURE 3
FIGURE 3:
Cholesterol efflux and ABCA1 expression in macrophages after Rho activation in the presence of atorvastatin. PMA-differentiated macrophage cells were treated by either Rho activators (20 μM of MEV, 20 μg/mL of FPP, and 2.5 μM of LPA) for 24 hours in the presence (shaded bars) of 20 μM atorvastatin. Atorvastatin untreated with macrophages (clear bar) served for a control. ApoAI-mediated cholesterol efflux (Panel A, n ≥ 6), ABCA1 mRNA (Panel B, n ≥ 4), and ABCA1 Western blot (Panel C, representative of 3 individual experiments) were analyzed. (Comparison to the control: ***P < 0.001; Comparison to atorvastatin treatment: ΔP < 0.05, ΔΔP < 0.01).

To address inconsistencies raised earlier after Rho activation in the presence of atorvastatin, the experiment was repeated in the absence of atorvastatin. Consistently, Rho activation did not increase but decreased apoAI-mediated cholesterol efflux (reduction of 19% for mevalonate, 31% for FPP, and 30% for LPA, Fig. 4A). Corresponding to the effects of Rho activators on cholesterol efflux, ABCA1 mRNA levels were also reduced by 14%, 14%, and 10% for mevalonate, FPP, and LPA, respectively (Fig. 4B). In parallel, Western blot showed mevalonate and LPA decreased ABCA1 protein by ~20% and ~40% (Fig. 4C). In addition, a more specific Rho agonist, GTPγS, led to a moderate decrease in ABCA1 mRNA at 4 hours. As a result of its rapid degradation, the GTPγS effect disappeared after 8 hours (Fig. 4D). These results indicate that the activation of Rho protein was inclined to decrease ABCA1 expression.

FIGURE 4
FIGURE 4:
Cholesterol efflux and ABCA1 expression in THP-1 macrophages after Rho activation or inhibition in the absence of atorvastatin. PMA-differentiated THP-1 macrophage cells were treated by either Rho activators (20 μM of MEV, 20 μg/mL of FPP, and 2.5 μM of LPA) or inhibitor (5 units/mL of Exo-C3) for 24 hours in the absence of atorvastatin. Apo AI-mediated cholesterol efflux (Panel A, n ≥ 6), ABCA1 mRNA (Panel B, n ≥ 4), ABCA1 Western blot (Panel C, representative of 3 individual experiments), and time course of ABCA1 expression after Rho activation by GTPγS (Panel D, n = 4) were analyzed. (Comparison to the control: *P < 0.05, **P < 0.01, ***P < 0.001).

To further clarify the relationship between Rho protein and ABCA1 expression, we applied Clostridium botulinum C3 toxin (C3 exoenzyme, 5 units/mL), which interferes with the ribosylation of Rho protein, to inhibit Rho activity in the absence of atorvastatin (confirmed by Rho pull-down assay, data not shown). Rho inhibitor clostridium botulinum C3 toxin (C3 exoenzyme, 5 units/mL) was associated with a 48% (P < 0.01) increase in ABCA1 mRNA levels (Fig. 4B), and the protein level was also increased by ~25% (Fig. 4C). Combined with previously mentioned findings, an inverse correlation between Rho and ABCA1 expression can be established. In spite of an increase in ABCA1 expression, Rho inhibition was associated with a 17% (P < 0.01) decrease in apoAI-mediated cholesterol efflux (Fig. 4A), indicating Rho inhibition may influence apoAI-mediated cholesterol efflux in pathways independent of ABCA1.

LXRα Agonists Stimulate ApoAI-mediated Cholesterol Efflux, Increase ABCA1 mRNA and Protein, and Decrease Rho Expression

Because Rho inhibition did not appear to be the mechanism by which atorvastatin suppressed ABCA1 expression, we then examined the potential role of LXRα. In the presence of atorvastatin, LXRα agonists 22(R)-hydroxycholesterol (OH-Chol) and T0901317 rescued apoAI-mediated cholesterol efflux, and, in fact, efflux rates were even slightly higher than control (shaded bars in Fig. 5A); both activators also completely recovered the decreased ABCA1 mRNA level associated with atorvastatin treatment (shaded bars in Fig. 5B); and ABCA1 protein on Western blot increased by 4.5- and 3.5-fold accordingly (Fig. 5C).

FIGURE 5
FIGURE 5:
LXRα activation and cholesterol efflux in THP-1 macrophages in the absence or presence of atorvastatin. PMA-differentiated THP-1 macrophages were treated with either synthetic (1 μM of T01901317) or natural activator (2.5 μM of 22(R)-hydroxycholesterol, OH-Chol) of LXRα for 24 hours in the presence (shaded bars) or absence (clear bars) of atorvastatin. Panel A, ApoAI-mediated cholesterol efflux after LXRα activation in the absence (n = 6 from 2 independent experiments) or presence (n = 6 from 2 independent experiments) of atorvastatin treatment. Panel B, ABCA1 Western blot after LXRα activation in the presence of atorvastatin treatment (representative of 3 independent experiments). Panel C, ABCA1 mRNA after LXRα activation in the absence (n ≥ 4) or presence (n ≥ 4) of atorvastatin treatment. Panel D, ABCA1 Western blot after LXRα activation in the absence of atorvastatin treatment (representative of 3 independent experiments). (Comparison to the control: *P < 0.05, **P < 0.01, ***P < 0.001; Comparison to atorvastatin treatment: ΔΔΔP < 0.001)

Not surprisingly, in the absence of atorvastatin, T01901317 and OH-Chol increased apoAI-mediated cholesterol efflux by 52% and 17% and ABCA1 mRNA levels by 45% and 52%, respectively (the empty bars in Fig. 5A, B); ABCA1 protein levels were consistent with mRNA changes with 1.8- and 2.5-fold increase for T01901317 and 22(R)-hydroxycholesterol, respectively (Fig. 5D). These findings indicate that LXRα was the target signaling molecule involved in atorvastatin-induced ABCA1 down-regulation.

Interaction Between Rho and LXRα

Because LXRα and Rho protein exerted opposite effects on ABCA1 expression, we then investigated their relationship with one another in THP-1 macrophages. Rho activation by LPA consistently decreased LXRα levels as shown by Western blot, whereas Rho inhibition by C3 exoenzyme was associated with an elevated LXR level (Fig. 6A). By contrast, LXRα activation by 22(R)-hydroxycholesterol and T0901317 significantly decreased Rho level by ~50% (Fig. 6B).

FIGURE 6
FIGURE 6:
Reciprocal inhibition of Rho and LXRα in THP-1 macrophages. Panel A, Western blot of LXRα (representative of 2 individual blots). Panel B, Rho activity was evaluated by pull-down assay and Western blot, as shown in figure. Rho activity was significantly decreased after LXRα activation by OH-Chol and T01901317 (representative of 2 individual blots).

AcLDL Loading Attenuated Atorvastatin Effects on Cholesterol Efflux and ABCA1 Expression

Cholesterol loading can largely change the gene expression pattern in macrophages,13 and Argmann et al have reported an increase in ABCA1 expression.10 We therefore implemented experiments with acLDL-loaded macrophages. Compared with nonfoamy cells where atorvastatin treatment significantly reduced apoAI-mediated cholesterol efflux and ABCA1 mRNA (Fig. 7A, B), acLDL loading significantly increased apoAI-mediated cholesterol efflux (Fig. 7A). In these foam cells, the inhibitory effect of atorvastatin on apoAI-mediated cholesterol efflux was abolished, although atorvastatin seemingly ended to decrease ABCA1 expression (Fig. 7A, B).

FIGURE 7
FIGURE 7:
ApoAI-mediated cholesterol efflux in foam cells. In ac-LDL loaded THP-1 macrophages, cholesterol efflux was induced at 10 μg/mL apoAI for 4 hours in the presence of 20 μM atorvastatin. Macrophages incubated with 1 μCi/mL [3H]-cholesterol but no acLDL served for comparison. Panel A, ApoAI-mediated cholesterol efflux in macrophages loaded with or without acLDL (n = 6 from 2 independent experiments). Panel B, ABCA1 mRNA in macrophages loaded with or without acLDL (n = 6 from 2 independent experiments). (Comparison to the control: *P < 0.05, ***P < 0.001; comparison between acLDL-loaded and unloaded macrophages in the absence of atorvastatin: ΔΔΔP < 0.001).

DISCUSSION

In the present study, we report that atorvastatin decreased the ABCA1 expression in both THP-1 and human monocyte-derived macrophages. As tested in THP-1 macrophages, this suppressive effect of atorvastatin was mediated through LXRα but not Rho suppression. However, this observation is metabolic status-specific because acLDL loading attenuates atorvastatin's effect on ABCA1 expression. Parallel to ABCA1 reduction, atorvastatin treatment decreased apoAI-mediated cholesterol efflux in nonloaded THP-1 macrophages.

The present study displayed an inverse relationship between Rho activation and ABCA1 expression and apoAI-mediated cholesterol efflux. By contrast, Rho inhibition by C3 exoenzyme increased ABCA1 expression. Consistent with a role of Rho in the regulation of ABCA1, Utech et al observed that GTP-binding proteins of the Rho family (RhoA, RhoB, RhoG, Rac-1) were enriched in fibroblasts from patients with Tangier disease (ABCA1 deficiency).14 Furthermore, treatment with GGPP, a Rho activator, resulted in a dose- and time-dependent reduction of ABCA1 expression, and inhibitors of geranylgeranyl transferase and Rho proteins significantly increased the promoter activity and expression of ABCA1 in Caco-2 and THP-1 cells.15 Despite the contradictory results where atorvastatin treatment increased ABCA1 expression in acLDL-loaded macrophages, Argmann displayed that Rho inhibition by C3 exoenzyme could increase ABCA1 expression.10 Because of the reciprocal nature of the relationship of Rho with ABCA1, the reduction in both Rho and ABCA1 expression following atorvastatin treatment indicated that the reduction of ABCA1 was not Rho-dependent.

Despite the marked increases in ABCA1 expression, treatment with Rho inhibitor C3 exoenzyme did not lead to an increased apoAI-mediated cholesterol efflux, suggesting Rho protein can regulate cholesterol efflux in an ABCA1-independent pathway. One of Rho proteins, Cdc42, can facilitate the intracellular transport and ultimate efflux of lipids by regulating cytoskeletal reorganization.16,17 Therefore, the inhibition of Rho/Cdc42 may contribute to decreased cholesterol efflux after treatment with Rho inhibitor C3 exoenzyme.

It appears that the action of statins on ABCA1 is linked to LXRα, in line with Wong's finding.9 LXRα activation by the synthetic activator T0901317 and the natural activator 22(R) hydroxycholesterol not only rescued reductions in ABCA1-mRNA in the presence of atorvastatin, but also increased ABCA1 expression and apoAI-mediated cholesterol efflux in the absence of atorvastatin. This finding can be supported by similar findings in glomerular mesangeal cells,18 Caco-2 cells,19 fibroblasts, and macrophages.20 The presence of LXRα binding sites in the ABCA1 promoter was reported,15 and LXRα deficiency abolished pitavastatin's effect on ABCA1 expression in macrophages,21 suggesting ABCA1 is one of the target genes of LXR. Taken together, LXRα inactivation is the mechanism by which atorvastatin reduces ABCA1 and apoAI-mediated cholesterol efflux. Also, we demonstrated a reciprocal relationship between LXRα and Rho for the first time in macrophages, which may explain why activators of each have opposite effects on ABCA1 expression.

Mevalonate is the common intermediate for the production of both isoprenoids and oxysterols; therefore, complete recovery of ABCA1 would be expected after mevalonate addition in the presence of atorvastatin. It is intriguing that mevalonate only partially increased ABCA1 expression in the presence of atorvastatin but inhibited ABCA1 expression in the absence of atorvastatin, suggesting atorvastatin may have other effects independent of HMG-CoA reductase. Previous reports have indicated that statins can mediate effects that appear to be mevalonate insensitive.22,23 In particular, it has been reported that these reductase-independent effects may be related by the ability of statins to bind to a β2-integrin.23

Furthermore, the inhibitory effect of atorvastatin on ABCA1 expression was attenuated such that the decreased apoAI-mediated cholesterol efflux by atorvastatin was normalized when macrophages were transformed into foam cells by acLDL loading. The similar observation was also reported by Wong et al, where the addition of 24(s), 25-epoxycholesterol normalized expression of ABCA1 and cholesterol efflux.9 The blunted effect of atorvastatin on ABCA1 expression in cholesterol-loaded macrophages could be a result of the increased production of endogenous oxysterol and subsequent LXRα activation, which consequently offset the effect of atorvastatin.13,24,25 Furthermore, Argmann et al showed that ABCA1 expression was even increased in acLDL-loaded macrophages after atorvastatin treatment.10 In that study, THP-1 monocytes were stimulated with a different differentiating agent, phorbol 12,13-dibutyrate (300 nM), for a longer time (7 days) in contrast to the present study (2 days). Also, a longer incubation time (24 hours) with apoAI was applied for cholesterol efflux studies compared with 4 hours in the present study and Rho activators were withdrawn during 24-hour cholesterol efflux. These variables may account for, at least in part, the observed disparity in results.

In conclusion, we have demonstrated that atorvastatin decreases ABCA1 expression and apoAI-mediated cholesterol efflux in THP-1 macrophages and decreases LXRα and Rho activity. LXRα activation not only increased ABCA1 expression and cholesterol efflux in the absence of atorvastatin, but also restored these functions in the presence of atorvastatin. By contrast, Rho activation was inversely correlated with ABCA1 expression and cholesterol efflux and had an opposing role compared to LXRα activation. The cholesterol loading by acLDL largely abolished the inhibitory effect of atorvastatin on ABCA1 expression. Furthermore, the suppressive effect of atorvastatin on ABCA1 expression was also observed in human monocyte-derived macrophages. Taken together, this study resolves the inconsistency regarding atorvastatin effect on ABCA1 expression, specifically in THP-1 and human monocyte-derived macrophages. This study also provides evidence that atorvastatin's effect on ABCA1 expression and apoAI-mediated cholesterol efflux in noncholesterol-loaded macrophages is mediated through an LXRα but not Rho dependent pathway, an effect compromised by cholesterol loading.

REFERENCES

1. Ross R. Atherosclerosis-an inflammatory disease. N Engl J Med. 1999;340:115-126.
2. Cuchel M, Rader DJ. Macrophage reverse cholesterol transport: key to the regression of atherosclerosis? Circulation. 2006;113:2548-2555.
3. Oram JF, Heinecke JW. ATP-binding cassette transporter A1: a cell cholesterol exporter that protects against cardiovascular disease. Physiol Rev. 2005;85:1343-1372.
4. Bodzioch M, Orso E, Klucken J, et al. The gene encoding ATP-binding cassette transporter 1 is mutated in Tangier disease. Nat Genet. 1999;22:347-351.
5. Rust S, Rosier M, Funke H, et al. Tangier disease is caused by mutations in the gene encoding ATP-binding cassette transporter 1. Nat Genet. 1999;22:352-355.
6. van Dam MJ, de Groot E, Clee SM, et al. Association between increased arterial-wall thickness and impairment in ABCA1-driven cholesterol efflux: an observational study. Lancet. 2002;359:37-42.
7. Singaraja RR, Fievet C, Castro G, et al. Increased ABCA1 activity protects against atherosclerosis. J Clin Invest. 2002;110:35-42.
8. Yildirir A, Muderrisoglu H. Non-lipid effects of statins: emerging new indications. Curr Vasc Pharmacol. 2004;2:309-318.
9. Wong J, Quinn CM, Brown AJ. Statins inhibit synthesis of an oxysterol ligand for the liver x receptor in human macrophages with consequences for cholesterol flux. Arterioscler Thromb Vasc Biol. 2004;24:2365-2371.
10. Argmann CA, Edwards JY, Sawyez CG, et al. Regulation of macrophage cholesterol efflux through hydroxymethylglutaryl-CoA reductase inhibition: a role for RhoA in ABCA1-mediated cholesterol efflux. J Biol Chem. 2005;280:22212-22221.
11. Edwards PA, Ericsson J. Sterols and isoprenoids: signaling molecules derived from the cholesterol biosynthetic pathway. Annu Rev Biochem. 1999;68:157-185.
12. Liao JK. Isoprenoids as mediators of the biological effects of statins. J Clin Invest. 2002;110:285-288.
13. Shiffman D, Mikita T, Tai JT, et al. Large scale gene expression analysis of cholesterol-loaded macrophages. J Biol Chem. 2000;275:37324-37332.
14. Utech M, Hobbel G, Rust S, et al. Accumulation of RhoA, RhoB, RhoG, and Rac1 in fibroblasts from Tangier disease subjects suggests a regulatory role of Rho family proteins in cholesterol efflux. Biochem Biophys Res Commun. 2001;280:229-236.
15. Gan X, Kaplan R, Menke JG, et al. Dual mechanisms of ABCA1 regulation by geranylgeranyl pyrophosphate. J Biol Chem. 2001;276:48702-48708.
16. Hirano K, Matsuura F, Tsukamoto K, et al. Decreased expression of a member of the Rho GTPase family, Cdc42Hs, in cells from Tangier disease-the small G protein may play a role in cholesterol efflux. FEBS Lett. 2000;484:275-279.
17. Tsukamoto K, Hirano K, Yamashita S, et al. Retarded intracellular lipid transport associated with reduced expression of Cdc42, a member of Rho-GTPases, in human aged skin fibroblasts: a possible function of Cdc42 in mediating intracellular lipid transport. Arterioscler Thromb Vasc Biol. 2002;22:1899-1904.
18. Wu J, Zhang Y, Wang N, et al. Liver X receptor-alpha mediates cholesterol efflux in glomerular mesangial cells. Am J Physiol Renal Physiol. 2004;287:F886-895.
19. Murthy S, Born E, Mathur SN, et al. Liver-X-receptor-mediated increase in ATP-binding cassette transporter A1 expression is attenuated by fatty acids in CaCo-2 cells: effect on cholesterol efflux to high-density lipoprotein. Biochem J. 2004;377:545-552.
20. Arakawa R, Tamehiro N, Nishimaki-Mogami T, et al. Fenofibric acid, an active form of fenofibrate, increases apolipoprotein A-I-mediated high-density lipoprotein biogenesis by enhancing transcription of ATP-binding cassette transporter A1 gene in a liver X receptor-dependent manner. Arterioscler Thromb Vasc Biol. 2005;25:1193-1197.
21. Zanotti I, Poti F, Favari E, et al. Pitavastatin effect on ATP binding cassette A1-mediated lipid efflux from macrophages: evidence for liver X receptor (LXR)-dependent and LXR-independent mechanisms of activation by cAMP. J Pharmacol Exp Ther. 2006;317:395-401.
22. Wagner AH, Gebauer M, Guldenzoph B, et al. 3-hydroxy-3-methylglutaryl coenzyme A reductase-independent inhibition of CD40 expression by atorvastatin in human endothelial cells. Arterioscler Thromb Vasc Biol. 2002;22:1784-1789.
23. Weitz-Schmidt G, Welzenbach K, Brinkmann V, et al. Statins selectively inhibit leukocyte function antigen-1 by binding to a novel regulatory integrin site. Nat Med. 2001;7:687-692.
24. Fu X, Menke JG, Chen Y, et al. 27-hydroxycholesterol is an endogenous ligand for liver X receptor in cholesterol-loaded cells. J Biol Chem. 2001;276:38378-38387.
25. Langmann T, Klucken J, Reil M, et al. Molecular cloning of the human ATP-binding cassette transporter 1 (hABC1): evidence for sterol-dependent regulation in macrophages. Biochem Biophys Res Commun. 1999;257:29-33.
Keywords:

atorvastatin; ATP-binding cassette transporter A1; cholesterol efflux; Rho protein; liver X receptor; macrophage

© 2008 Lippincott Williams & Wilkins, Inc.