GTC
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Genes to Cells (2008) 13, 889-904. doi:10.1111/j.1365-2443.2008.01213.x
© 2008 Blackwell Publishing or its licensors
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Tanaka, A. R.
Right arrow Articles by Murata, M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Tanaka, A. R.
Right arrow Articles by Murata, M.

Formation of cholesterol-enriched structures by aberrant intracellular accumulation of ATP-binding cassette transporter A1

 Arowu R. Tanaka1, Fumi Kano1, Akitsugu Yamamoto2, Kazumitsu Ueda3 and Masayuki Murata1,*

1 Department of Life Sciences, Graduate School of Arts and Sciences, The University of Tokyo, Tokyo 153-8902, Japan
2 Department of Bio-Science, Faculty of Bio-Science, Nagahama Institute of Bio-Science and Technology, Shiga 526-0829, Japan
3 Laboratory of Cellular Biochemistry, Division of Applied Life Sciences, Graduate School of Agriculture, Kyoto University, Kyoto 606-8502, Japan


   Abstract
Top
 Abstract
Introduction
 Results
 Discussion
 Experimental procedures
 References
 
ATP-binding cassette transporter A1 (ABCA1) is a key transporter associated with excess cellular lipid efflux. Here, we report that in HEK293 cells ABCA1 functions in intracellular compartments along the endocytic pathway. Inhibition of ABCA1-GFP degradation with proteasome inhibitors induced the internalization of ABCA1 and the formation of intracellular round-shaped structures, designated "A1 bodies". Importantly, cholesterol was selectively accumulated in A1 bodies, and this depended on the cholesterol efflux activity of ABCA1. Treatment with either lactacystin or acetylated LDL, which reduces proteasome activity, resulted in internalization of ABCA1 in mouse peritoneal macrophages. By performing array analysis on macrophages treated with these reagents, we identified Rab4 as a key protein involved in the internalization and aberrant accumulation of ABCA1 in HEK cells. Treatment of the cells with proteasome inhibitors inhibited the degradation of Rab4, and Rab4 over-expression induced the formation of small A1 bodies. Furthermore, A1 bodies formation was substantially inhibited by silencing of the endogenous Rab4 gene. Taken together, our findings suggest that the endocytic ABCA1 possesses cholesterol efflux activity, and thus the cellular control of post-endocytic sorting, retention or recycling of functional ABCA1 in the endocytic vesicles, which is in part regulated by Rab4, is important for cholesterol metabolism in living cells.


   Introduction
Top
 Abstract
 Introduction
Results
 Discussion
 Experimental procedures
 References
 
Cellular cholesterol efflux is indispensable for maintaining cholesterol homeostasis in peripheral cells. ATP-binding cassette transporter A1 (ABCA1) is a key membrane protein with respect to the efflux of cellular cholesterol and phospholipids, and known to be involved in their transfer to extracellular lipid-poor apolipoproteins such as apoA-I, initiating the formation of high density lipoprotein (HDL; Oram & Yokoyama 1996; Mendez 1997; Takahashi et al. 2005). Thus, Dysfunction of ABCA1, involving the loss of its ability to bind HDL or to bind or hydrolyse ATP, results in Tangier disease, a genetic disorder typified by reduced levels of circulating HDL (Bodzioch et al. 1999; Brooks-Wilson et al. 1999; Rust et al. 1999). Studies of patients with Tangier disease and mouse models led to the hypothesis that over-expression of ABCA1 could have a protective role against atherosclerosis (McNeish et al. 2000; Orso et al. 2000).

Previously, we have reported that the defect in HDL assembly in two point mutants of ABCA1 (R587W, Q597R) responsible for familial HDL deficiency is due to the impaired localization to the plasma membrane (Tanaka et al. 2003). Many of the human ABC superfamily genes have been linked to disorders displaying Mendelian inheritance (Dean 2005), and in some of these diseases, mislocalization caused by mutations plays a critical role in the pathogenesis as for ABCA1. For example, the CFTR/ABCC7 {Delta}F508 mutation, found in approximately 70% of cystic fibrosis patients, results in defective chloride channel activity because of retention of the protein in the ER by the ER quality control system (Kerem et al. 1989; Amaral 2004). SUR1/ABCC8 mutants identified in hyperinsulinemic hypoglycemia are also trapped in the ER and missing its potassium channel regulatory activity (Yan et al. 2004). Thus, mislocalization or problems with the trafficking pathway are important common causes of pathogenicity with regard to ABC proteins. In particular, this issue might be more critical for ABCA1 because its shuttling between the plasma membrane and endocytic vesicles is assumed to be associated with its cholesterol efflux function.

Time-lapse fluorescence microscopic observations using green fluorescence protein (GFP)-fused ABCA1 revealed that ABCA1 is mainly localized to the plasma membrane but some ABCA1 shuttle between the plasma membrane and intracellular compartments (early- or late-endosomes) (Neufeld et al. 2001; Tanaka et al. 2003). These data raise the possibility that endocytosed apoA-1, which undergoes partial cellular uptake, present in ABCA1-positive endosomes plays a role in the first step of HDL formation, and the resulting HDL is subsequently recycled to the plasma membrane for release to the surrounding medium (Takahashi & Smith 1999; Smith et al. 2002; Neufeld et al. 2004). In this case, the cholesterol in nascent HDL might be derived in part from the endosomal pool. However, the source of cholesterol remains unclear and the mechanism of shuttling of ABCA1 and apoA-I in HDL assembly is still controversial.

Here, we describe that proteasome inhibitors cause the aberrant accumulation of ABCA1 in endocytic vesicles and result in the abnormal accumulation of cholesterol in these vesicles in HEK293 cells. The accumulation of cholesterol depends on the cholesterol efflux activity of ABCA1. By performing array analysis on macrophages treated with acetylated LDL and lactacystin, we have identified a small GTPase, Rab4, as a key protein that regulates the localization of ABCA1 in endocytic vesicles via the ubiquitin-proteasome system. In addition, silencing of the Rab4 gene by siRNA substantially inhibits the ABCA1 accumulation induced by proteasome inhibitors. These results indicated that the accurate cellular localization of ABCA1 is regulated in part by Rab4 and the mislocalization might induce the disturbance of cholesterol metabolism in peripheral cells.


   Results
Top
 Abstract
 Introduction
 Results
Discussion
 Experimental procedures
 References
 
Proteasome inhibitors perturb the localization of ABCA1 and result in the aberrant accumulation of ABCA1 in the endocytotic pathway

In the previous study, we established a cell line HEK-A1WT from HEK293 cells, which stably expressed ABCA1 fused with GFP (ABCA1-GFP), to monitor dynamic intracellular behavior and localization of the protein by time-lapse confocal microscopy (Tanaka et al. 2003). That study and others demonstrated that ABCA1 shuttled between the plasma membrane and intracellular compartments (Neufeld et al. 2001). In the present study, we found that some proteasome inhibitors caused perturbation of the intracellular localization of ABCA1-GFP. We treated HEK-A1WT cells with lactacystin, a specific proteasome inhibitor, and followed ABCA1-GFP localization by confocal microscopy. After an 8-h incubation, the amount of ABCA1-GFP on the cell surface decreased and extensive intracellular accumulation was observed in large round-shaped membranous structures (Fig. 1A). After 12 h, these structures grew larger with diameters of 2–5 µm (Fig. 1A). We designated these structures ‘A1 bodies’. They were also formed with the proteasome inhibitor MG132 or ALLN, a thiol protease and proteasome inhibitor known to prevent ABCA1 degradation (Fig. 1B) (Arakawa & Yokoyama 2002). In contrast, treatment with NH4Cl, a lysosomal enzyme inhibitor, induced ABCA1 internalization, but did not induce the formation of these structures (Fig. 1B, right). In addition, we treated with HEK-A1WT with leupeptin, a lysosomal enzymes inhibitor. As shown in Fig. S3A Supplementary Material, A1 body was not generated as well as NH4Cl treatment. All of the inhibitors described above both caused internalization of ABCA1-GFP and inhibited its degradation (Fig. 1C), but only the proteasome inhibitors induced the A1 body formation.


Figure 1
View larger version (42K):
[in this window]
[in a new window]

  		Figure 1  A1-body formation in HEK-A1WT cells. (A) HEK-A1WT cells were treated with 10 µM lactacystin for the indicated times. (B) HEK-A1WT cells were treated with 5 µM MG132, 10 µM ALLN or 5 mM NH4Cl. (C) Extracts from HEK-A1WT cells treated with the indicated reagents were analyzed by immunoblotting with anti-ABCA1 antibodies and anti-{alpha}-tubulin antibodies. Bar = 10 µm.

 
To characterize the origin of A1 bodies, immunofluorescence analysis was performed using specific markers of organelles in the endocytotic pathways. The small ABCA1-positive compartments colocalized with EEA1, a marker as specific for a set of early endosomes, but A1 bodies themselves did not (Fig. 2A). LAMP2, a marker of late endosomes/lysosomes, did localize to A1 bodies (Fig. 2C). Next, we incubated HEK-A1WT cells with ALLN for 12 h to generate A1 bodies, and then further incubated the cells with tetramethyl-rhodamine-dextran (TRD:10 000 MW), a endocytosis tracer, for 5 h followed by a 1 h chase without TRD in the presence of ALLN. Almost all of the TRD did not colocalize with A1 bodies, but a portion of the TRD accumulated not only in small ABCA1-positive compartments but also in A1 bodies (Fig. 2B). The subpopulation of TRD accumulated in ABCA1-positive vesicles appeared to deviate from the pathway to lysosomes. In addition, we treated HEK-A1WT with MG132 in the presence of cycloheximide (CHX). As shown in Supplementary Fig. S3B, A1 body was generated even in the CHX-treated cells, suggesting ABCA1 in A1 body was derived from the plasma membrane but not from other sources, such as the newly synthesized pool and a population that had somehow escaped lysosomal degradation. Taken together, these results suggested that A1 bodies correspond to late endosomes/lysosomes and the formation of A1 bodies is attributable to the impairment of endocytic trafficking of ABCA1.


Figure 2
View larger version (40K):
[in this window]
[in a new window]

  		Figure 2  A1-body characterization. (A) HEK-A1WT cells were treated with 10 µM ALLN and stained with anti-EEA1 antibodies. ABCA1-GFP (green) and EEA1 (red) are overlaid. (B) HEK-A1WT cells treated with 10 µM ALLN were incubated with tetramethyl-rhodamine-dextran (TRD), washed and incubated with normal medium for 1 h. ABCA1-GFP (green) and TRD (red) are overlaid. (C) HEK-A1WT cells were treated with 10 µM ALLN and stained with anti-LAMP2 antibodies. ABCA1-GFP (green) and LAMP2 (red) are overlaid. Bar = 10 µm.

 
An abnormal amount of free cholesterol accumulates in A1 bodies

If functional ABCA1 accumulates in A1 bodies, free cholesterol, which is a major substrate of ABCA1, should be transported actively into them. Therefore we examined whether cholesterol accumulated in A1 bodies by staining HEK-A1WT cells with filipin that detect free cholesterol. A1 bodies in HEK-A1WT cells treated with ALLN were extensively stained by filipin but no extensively stained organelles were detected in wild-type HEK293 cells (Fig. 3, control). These results indicated that an abnormal amount of free cholesterol accumulated in A1 bodies.


Figure 3
View larger version (23K):
[in this window]
[in a new window]

  		Figure 3  Cholesterol is accumulated in A1 bodies. HEK-A1WT cells or HEK293 were treated with 10 µM ALLN for 12 h, fixed and stained with filipin. ABCA1-GFP (green) and filipin (red) are overlaid. Bar = 10 µm.

 
The abnormal accumulation of cholesterol in A1 bodies was also confirmed by a biochemical assay. We isolated A1 bodies using anti-GFP antibody-conjugated magnetic beads to analyze their lipid components by thin layer chromatography (TLC). These isolated A1 bodies from ALLN-treated HEK-A1WT cells maintained their round-shaped structure, as shown by confocal microscopy (Fig. 4A). Consistent with the results of filipin staining of A1 bodies in living cells, TLC analysis showed that cholesterol was predominantly concentrated in the immuno-isolated A1 bodies (Fig. 4B, right), although there was no difference in the proportion of total cell lipids between ALLN-treated or untreated HEK-A1WT cells (Fig. 4B, left).


Figure 4
View larger version (34K):
[in this window]
[in a new window]

  		Figure 4  Purified A1 bodies selectively contain cholesterol. (A) Purification of A1 bodies. Purified A1 bodies and control microsomes contained ABCA1-GFP as shown by confocal microscopy. Bar = 1 µm. (B) Total lipids from HEK-A1WT cells treated or untreated with ALLN were quantified by thin-layer chromatography (left). HEK-A1WT cells treated or untreated with ALLN were physically homogenized. Microsomes were fractionated by centrifugation and purified by anti-GFP antibody conjugated beads. Lipids in purified fractions were quantified by thin-layer chromatography (right). FC = free cholesterol, PE = phosphatidylethanolamine, PC = phosphatidylcholine.

 
Next, we investigated the origin of the cholesterol accumulated in A1 bodies. When HEK-A1WT cells were incubated with lipoprotein-deficient serum (LPDS) for the depletion of LDL-derived cholesterol or with mevastatin, a HMG-CoA reductase inhibitor, for the depletion of de novo synthesized cholesterol, the formation of A1 bodies was reduced to < 89% or 79% of that observed in control cells, respectively (Fig. 5B). In contrast, when the mevastatin-treated cells were incubated with LPDS the formation of A1 bodies was reduced drastically to less than 31% of that seen in control cells. These results suggested that the inhibitory effects of mevastatin treatment and LPDS on the formation of A1 bodies were synergistic, and that free cholesterol in A1 bodies is derived from both intracellular biosynthetic pathways and serum in extracellular medium.


Figure 5
View larger version (26K):
[in this window]
[in a new window]

  		Figure 5  The source of cholesterol in A1 bodies. (A) HEK-A1WT cells were treated with 10 µM ALLN for 24 h. The cells were cultured in DMEM/10% FCS, DMEM/5% LPDS or DMEM/5% LPDS + 40 µM mevastatin. (B) Effects of LPDS and/or mevastatin on A1 body formation were quantified. Cells having at least one A1 body > 1 µm in diameter were counted. The formation rate of A1 bodies in culture using FCS was defined as 100%. More than 150 cells were examined, and experiments were performed three times. Data represent the mean ± SD of at least three independent experiments. Bar = 10 µm.

 
Functional ABCA1 is essential for abnormal cholesterol accumulation in A1 bodies

To determine whether ABCA1 function is required for the formation of A1 bodies, we studied the effect of the K939M ABCA1-GFP mutant on A1-body formation. ABCA1 has ATP binding/hydrolysis activity (Tanaka et al. 2001; Takahashi et al. 2006) and the mutant protein, in which lysine 939 in the Walker A motif of NBF1 is changed to methionine, is known to have lost ABCA1 function (Hamon et al. 2000), that is, the mutant does not have the ability to mediate apoA-I induced cholesterol efflux (Fig. 6A, WT+ and KM+), although both K939M ABCA1-GFP and the wild-type protein localized mainly to the plasma membrane (Fig. 6B, 0 h). Upon ALLN treatment, K939M ABCA1-GFP was translocated from the plasma membrane to intracellular compartments. However, in contrast to wild-type ABCA1-GFP, the mutant protein did not cause A1-body formation. The intracellular vesicles containing the internalized mutant protein remained small and did not grow into A1 bodies (Fig. 6B, 24 h). The expression level of KM protein was lower than that of WT protein (Fig. 6A). To confirm the effect of the quantity of ABCA1 protein on the formation of A1 bodies, we used RNA interference (RNAi). As shown in Supplementary Fig. S5A, the gene silencing reduced the protein level of wild-type ABCA1-GFP in HEK-A1WT to lower than 25% of control HEK-A1WT. In the ABCA1 knockdown cells, the A1 body formation by proteasome inhibitor treatment was observed (Supplementary Fig. S5B, arrows). Next, we have still observed A1 body formation in the CHX-treated HEK-A1 cells, in which expression level of ABCA1-GFP was reduced, in the presence of MG132 (Supplementary Fig. S3B). Taken together, we conclude that KM is not capable of forming A1 body in the endosomes not because of the low expression level of KM protein but of the defect in ApoA-I induced cholesterol efflux activity. These results suggest that the cholesterol efflux activity of ABCA1 in the plasma membrane could also be exerted within the endocytic compartments, and that functional ABCA1 is necessary for the accumulation of free cholesterol in A1 bodies.


Figure 6
View larger version (24K):
[in this window]
[in a new window]

  		Figure 6  Functional ABCA1 is necessary for A1-body formation. (A) ApoA-I-induced cellular cholesterol release activity was measured. Data represent the mean ± SD of three independent experiments. Extracts from HEK-A1WT cells or HEK-K939M cells were analyzed by immunoblotting with anti-ABCA1 antibodies and anti-{alpha}-tubulin antibodies. (B) HEK-A1WT cells and HEK-K939M cells were treated with 10 µM ALLN for the indicated times. Bar = 10 µm.

 
To define the fine structural features of A1 bodies, electron microscopy analysis was performed. As shown in Fig. 7, we found many unique vesicles of 0.2–3 µm diameter, consisting of lipid bilayers, in ALLN-treated but not untreated HEK-A1WT cells (Fig. 7A,B,C). There were two populations of such vesicles, as shown in Fig. 7B (the larger ones, arrow) and Fig. 7C (the smaller ones: arrow head). The smaller vesicles often aggregated and appeared to fuse to form larger ones, > 1 µm in diameter (Fig. 7D), implying that the larger type of vesicles, corresponding to A1 bodies, were derived from the smaller ones, probably corresponding to both ABCA1- and EEA1-positive small compartments (Fig. 2A). Interestingly, the smaller vesicles often contained dozens of granules with higher electron densities and constant diameters (Fig. 7C). We have also observed by light microscopy that A1 bodies included fluorescence-conjugated apoA-I (data not shown). Taken together, we speculate that maturating A1 bodies could include cholesterol-containing particles, for example, HDL, which are generated by functional ABCA1 (see Discussion).


Figure 7
View larger version (86K):
[in this window]
[in a new window]

  		Figure 7  Structural features of A1 bodies. HEK-A1WT cells treated with DMSO (A) or ALLN (B–D) for 12 h were fixed with glutaraldehyde and studied by electron microscopy. Unique vesicles of diameter 0.2–3 µm were observed (B, arrow). The smaller vesicles contained dozens of granules (C, arrowhead). The smaller vesicles aggregated and appeared to fuse to form larger ones (D, arrow). Bar = 1 µm.

 
Aberrant internalization of ABCA1 in proteasome inhibitor-treated macrophages and in acetylated LDL-induced foam macrophages

Our findings described above indicated that aberrant accumulation of functional ABCA1 within the endocytic pathway caused A1-body formation, and that the disturbance of ABCA1 localization might disrupt cholesterol or lipid metabolism. To investigate this further, we focused on mouse peritoneal foam macrophages, which are induced by acetylated LDL (AcLDL)- or oxidized LDL (OxLDL)-treatment, for the following reasons. First, abnormal accumulation of cholesterol or lipids is frequently observed in foam macrophages although the accumulated structures are not necessarily correspond to A1 bodies. Second, Vieira et al. have reported that OxLDL reduces proteasome activity and induces the intracellular accumulation of ubiquitinated proteins (Vieira et al. 2000). This report prompted us to examine the relationship between proteasome activity and the localization of ABCA1 in foam macrophages. Thirdly, we could investigate the effect of proteasome inhibitors on the changes in localization of endogenous ABCA1 by using macrophages.

To test whether proteasome inhibitors induce the internalization and aberrant accumulation of endogenous ABCA1 or not, we used the cell surface biotinylation method to estimate the extent of plasma membrane-localized ABCA1 in control or foam macrophages. Macrophage surface proteins were labeled with biotin by sulfo-NHS-biotin and then ABCA1 was immunoprecipitated with anti-ABCA1 antibodies. ABCA1 protein was separated by SDS-PAGE and blotted with anti-ABCA1 antibodies for total applied ABCA1 or avidin-conjugated horseradish peroxidase for applied surface ABCA1. Endogenous ABCA1 in the plasma membrane of lactacystin-treated macrophages decreased to 58.9 ± 19.6% of the control value (Fig. 8A, right). This indicated that the proteasome inhibitor caused internalization of endogenous ABCA1 in macrophages as has been observed with ABCA1-GFP in HEK-A1WT cells. Interestingly, in the macrophages treated with lactacystin for 3 days we observed many cytoplasmic droplets stained with Sudan III, a lipid droplets staining reagent (Supplementary Fig. S2). We have tried to confirm the intracellular accumulation of endogenous ABCA1 in macrophages using fluorescence microscopy, but have been unable to do so owing to a lack of anti-ABCA1 antibodies effective at detecting the endogenous protein.


Figure 8
View larger version (13K):
[in this window]
[in a new window]

  		Figure 8  Effects of the proteasome on ABCA1 internalization. (A) Mouse peritoneal macrophages were incubated with lactacystin or AcLDL for 3 days. Macrophage surface proteins were labeled with biotin and then ABCA1 was immunoprecipitated with anti-ABCA1 antibodies. Applied surface ABCA1 was analyzed by avidin-conjugated horseradish peroxidase and total applied ABCA1 was immuno-blotted with anti-ABCA1 antibodies. Data represent the mean ± SD of three independent experiments. (B) Macrophages treated with AcLDL for 3 days were lysed and incubated with Suc-LLVY-AMC, in the presence or absence of clasto-lactacystin-β-lactone (clbl). The fluorescence of the liberated AMC was measured with a spectrofluorometer. The proteasome activity of normal macrophages was defined as 100%. Data represent the mean ± SD of three independent experiments.

 
Next, we measured the proteasome activity of AcLDL-treated foam macrophages using sLLVY-MCA, a fluorogenic synthetic peptide substrate for 20S proteasomes. We found that 20S proteasome activity decreased to 85% of that in normal macrophages (Fig. 8B). In addition, the ratio of ABCA1 in the plasma membrane to total ABCA1 in AcLDL-treated cells decreased to 80.8 ± 3.3% of the control value, although the total amounts of ABCA1 increased as compared with untreated macrophages (Fig. 8A, lower). We also found that OxLDL had the similar effects on the proteasome activity and internalization of ABCA1 (data not shown). Taken together, these results suggested that a common intracellular signaling pathway might be involved at some level in the aberrant internalization and accumulation of ABCA1 in lactacystin-treated macrophages or in AcLDL-treated foam macrophages.

Micro-array analysis to identify possible candidate proteins encoded by lactacystin- or AcLDL-inducible genes in mouse peritoneal macrophages

As observed in both proteasome inhibitor-treated HEK-A1WT cells and AcLDL-induced foam macrophages, accurate localization of ABCA1 in these cells is essential for its correct metabolism of cholesterol and lipids. Therefore, to search for the protein(s) that cause the internalization of ABCA1 in proteasome inhibitor-treated HEK-A1WT cells, we narrowed down the candidate proteins encoded by lactacystin- or AcLDL-inducible genes in mouse peritoneal macrophages using micro-array analyses. Remarkably, genes induced by lactacystin correlated highly with those induced in AcLDL-treated cells (correlation coefficient = 0.7719) (Supplementary Fig. S2). By using the k-means algorithm (MacQueen 1967), we categorized our micro-array data into 50 clusters and selected 139 potentially interesting genes from 11 clusters. Of these, 77 were prominent up-regulated (5 clusters) and 62 were prominent down-regulated (6 clusters) by both lactacystin and AcLDL treatment (Supplementary Table S1). Next, we examined whether these genes cause ABCA1 translocation from the plasma membrane to intracellular compartments. For the up-regulated group, HEK293 cells were co-transfected with both the ABCA1-GFP vector and an expression vector containing the gene of interest. Reciprocally, for the down-regulated group, RNA interference-mediated knockdown of the genes of interest in HEK-A1WT cells was performed.

As a result, we found that Rab4B, the mRNA for which was up-regulated 1.7-fold in AcLDL-treated and 2.4-fold in lactacystin-treated macrophages as compared to BSA-treated controls, caused the aberrant accumulation of ABCA1-GFP in HEK293 cells when over-expressed. Rab4B is a Ras-related GTP-binding protein that has 87% high-grade homology with Rab4A, and is reported to be involved in the regulation of endosomal sorting (Jones et al. 2006). As have been expected, over-expressed Rab4A also caused the accumulation of ABCA1-GFP in HEK293 cells. The effects of Rab4A on the localization of ABCA1-GFP in the cells were indistinguishable from those of Rab4B (Fig. 9C). Therefore, we focused on both Rab4A and Rab4B as potential candidates for further analyses.


Figure 9
View larger version (25K):
[in this window]
[in a new window]

  		Figure 9  Effects of Rab4 on ABCA1 localization. (A) Extracts from HEK-A1WT cells treated with DMSO or 5 µM MG132 for 12 h were analyzed by immunoblotting with anti-ABCA1 antibodies and anti-{alpha}-tubulin antibodies. (B) HEK293 cells were co-transfected with both ABCA1-GFP and HA-Rab4A (1 : 1). ABCA1-GFP (green) and HA-Rab4A (red) are overlaid. Bar = 10 µm. (C) Among ABCA1-GFP expressing cells, the percentages of cells with three types of ABCA1-GFP localization, large intracellular vesicles, the ER or the PM, were assessed. In the graph, the percentage for mock vector-transfected cells (white) or HA-Rab4A/B vector-transfected cells (black) is indicated. More than 450 cells were examined, and data represent the mean ± SD of three independent experiments. (D) Filipin staining for cholesterol accumulation in Rab4-transfected cells. HEK293 cells were co-transfected with both ABCA1-GFP and HA-Rab4A (1 : 1) or only HA-Rab4A, fixed and stained with filipin. Bar = 10 µm.

 
Rab4 is one of the regulators responsible for the aberrant accumulation of ABCA1 in proteasome inhibitor-treated HEK-A1WTls

First, we estimated the total amount of endogenous Rab4A/Rab4B protein in HEK-A1WT by Western blotting using a rabbit anti-Rab4A polyclonal antibody because we did not have an adequate anti-Rab4B antibody. The amount of Rab4A was increased by MG132 treatment (Fig. 9A). Next, we co-transfected both ABCA1-GFP and HA-tagged Rab4A into HEK293 cells and found that ABCA1-GFP present in cells also containing HA-Rab4A was translocated, from the plasma membrane, into large intracellular vesicles of 1–2 µm diameter, that is, smaller than A1 bodies (Fig. 9B). Rab4B had the same effect on ABCA1 localization (data not shown). Interestingly, we have observed three types of localization of ABCA1-GFP in HEK293 cells, in which Rab4A or Rab4B was over-expressed: normal localization to the plasma membrane, localization to large intracellular vesicles and localization to the ER network. We performed morphometric analysis to quantify the effect of over-expressed Rab4A/Rab4B on the ABCA1 localization in cells (Fig. 9C). Over-expressed Rab4A or Rab4B caused the translocation of ABCA1 from the plasma membrane to large intracellular vesicles; with Rab4A, there was an increase from 14% to 39% in the amount of cells where ABCA1-GFP was found in the large vesicles; for Rab4B the corresponding change was 8%–32%. The large intracellular vesicles, which were induced by the over-expression of Rab4A or Rab4B, stained with filipin, indicating that free cholesterol was substantially accumulated in them. However, the degree of accumulation was not as great in the vesicles as in A1 bodies (Fig. 9D). Interestingly, aberrant accumulation of ABCA1-GFP in the ER (18% due to Rab4A or 22% due to Rab4B) was also observed. This raises the possibility that Rab4A or Rab4B is involved in the exit of proteins from the ER network. These results suggest that an increase in Rab4A or Rab4B protein, for example, under condition of lower proteasome activity, could disturb the localization of ABCA1-GFP in the cells and result in the abnormal accumulation of cholesterol in endocytic compartments. It should be noted that the vesicles induced by Rab4 over-expression was not completely identical to A1 bodies with regard to the average size and the cholesterol content.

Finally, we used RNAi to confirm whether Rab4A directly affects the formation of A1 bodies. Rab4A protein in HEK-A1WT cells were knocked down by RNAi (Fig. 10A). In Rab4A-kockdown cells, the formation of A1 bodies was substantially inhibited upon treatment with the proteasome inhibitor, MG132 (Fig. 10B). The percentage of cells, in which A1 bodies were induced by MG132, decreased from 69% for the control to 40% for Rab4A-knockdown cells (Fig. 10C). In addition, aggresomes were markedly observed in Rab4A-knockdown cells (Fig. 10B, arrow heads and Supplementary Fig. S7) (Johnston et al. 1998).


Figure 10
View larger version (13K):
[in this window]
[in a new window]

  		Figure 10  Knockdown of Rab4A by RNAi reduced the formation of A1 bodies. (A) Extracts from HEK-A1WT cells transfected with control siRNA (100 nM: lane 1) or siRNA specific for hRab4A (100 nM: lane 2) were analyzed by immunoblotting with anti-Rab4A antibodies and anti-{alpha}-tubulin antibodies. (B) HEK-A1WT cells transfected with control or hRab4A siRNA were treated with 5 µM MG132. Aggresomes were markedly observed (arrow heads). Bar = 10 µm. (C) Cells having at least one A1 body > 1 µm in diameter were counted. More than 100 cells were examined, and experiments were performed three times. Data represent the mean ± SD of at least three independent experiments.

 
Taken together, these results suggest that Rab4A, and probably Rab4B, is one of the key proteins that regulate the accurate localization of ABCA1-GFP in HEK293 cells, and its dysfunction or altered amount might cause the aberrant accumulation of ABCA1 in endocytic vesicles, resulting in the formation of A1 bodies.


   Discussion
Top
 Abstract
 Introduction
 Results
 Discussion
Experimental procedures
 References
 
In general, precise localization in the cell is crucially important for all proteins to fulfill their functions optimally. The disturbance of localization can cause dysfunction of the protein, resulting in a huge perturbation of normally homeostatic cellular activities: metabolism, the cell cycle or signal transduction. On the other hand, a mislocalized protein could be fully active but its abnormal location might result in the worsening of cellular conditions.

In this paper, we focused on the intracellular localization of ABCA1, which is a key membrane protein for cellular cholesterol release, and found that an aberrant accumulation of functional ABCA1 in endocytic vesicles (referred to as A1 bodies), which were induced by proteasome inhibitors, caused an abnormal accumulation of cholesterol in these A1 bodies. This resulted in the perturbation of cholesterol or lipid homeostasis as observed in foam macrophages. In addition, we found that Rab4, which is known to be regulator of endocytic pathways, is involved in the aberrant accumulation of ABCA1 in the endocytic vesicles.

First, we found that ABCA1-GFP was internalized and accumulated in the round-shaped membrane structures, A1 bodies, in the HEK-A1WT cells, when we treated these cells with proteasome inhibitors such as lactacystin or ALLN, but not lysosomal protease inhibitors NH4Cl or leupeptin. (Fig. 1 and Supplementary Fig. S3). These results suggest that Ubiquitin–Proteasome system is involved in the A1 body formation. The stable expression level of ABCA1-GFP in HEK-A1WT is about less than twofold of foam mouse peritoneal macrophages (Supplementary Fig. S1), indicating that the over-expression of ABCA1 has limited effects on HEK cells. The morphological and biochemical analyses shown in Figs 2–4 revealed that A1 bodies contained abnormal amount of free cholesterol originating from both extracellular serum and intracellular pools of cholesterol (Fig. 5).

We hypothesized that the ABCA1-GFP that accumulated in A1 bodies is capable of transporting cholesterol and induces atypical cholesterol influx into A1 bodies. By using the K939M ABCA1-GFP mutant protein, which does not have the ability to mediate apoA-I induced cholesterol release, we confirmed that functional ABCA1-GFP plays a crucial role in the formation of A1 bodies, indicating that ABCA1-GFP has the ability to transport cellular cholesterol into A1 bodies or to keep LDL cholesterol into A1 bodies. Morphological analysis using light microscopy and EM revealed that developing A1 bodies in HEK-A1WT cells included apoA-I molecules and particles resembling precursor HDL. These findings support the idea that ABCA1, which shuttles between the plasma membrane and early/late endosomes, is involved in cholesterol efflux with the aid of apoA-I (Takahashi & Smith 1999; Smith et al. 2002; Neufeld et al. 2004). In fact, it has been reported that other ABCA subfamily proteins except for ABCA7 are localized to and function in intracellular compartments (Weng et al. 1999; Nagata et al. 2004; Abe-Dohmae et al. 2006).

The relationship between A1 bodies and HDL formation remains unclear. However, we assume that particles generated by functional ABCA1 in A1 bodies are partially analogous to precursor HDL. There were several reports arguing about the model that ABCA1-containing endosomes might be involved in HDL formation (Takahashi & Smith 1999; Smith et al. 2002; Neufeld et al. 2004). In addition, Chen et al. also reported that cholesterol deposited in late endosomes/lysosomes preferentially acts as a source of cholesterol for ABCA1-mediated cholesterol efflux (Chen et al. 2001). Other is that the relationship between ABCA1 and NPC1, which facilitates the egress of cholesterol from late endosomes/lysosomes to other cellular compartments (Boadu & Francis 2006). ABCA1 may be normally active inside the cells, especially in endosomes. Although further biochemical and morphological studies are required to elucidate the precise role of apoA-I in ABCA1 positive endosomes, our findings support the hypothesis that ABCA1 could pump the cholesterol into the endosomes, probably with the aid of apoA-1, resulting in HDL formation.

Interestingly, a work reported by Vieira et al. indeed allows us to focus our attention on proteasome activity in macrophage foam cells, in which the abnormal accumulation of cholesterol and lipids is frequently observed (Vieira et al. 2000). They have reported that OxLDL, an inducer of macrophage foam cells, reduces proteasome activity. Therefore, we examined the effect of lactacystin and AcLDL, another inducer of macrophage foam cells, on the localization of ABCA1, and found that both induced the internalization of endogenous ABCA1 in macrophages (Fig. 8A). Notably, proteasome activity was decreased substantially in AcLDL-treated macrophages (Fig. 8B).

To determine whether the decrease of proteasome activity by AcLDL affects ABCA1 degradation or sorting directly, the ubiquitination of ABCA1 was measured. We could not detect ubiquitinated ABCA1 (Supplementary Fig. S4). Next, we tried to examine indirect roles of the ubiquitin–proteasome system in ABCA1 localization by using micro-array analysis. The genes induced by lactacystin treatment correlated highly with those induced by AcLDL treatment. This suggested that the effect of AcLDL-treatment on proteasome activity could markedly influence the gene-expression profile of macrophages. We selected 139 genes from the most up- and down-regulated groups (Supplementary Table S1). Over-expression by vector transfection or knockdown by RNA interference of a gene of interest in each of the respective groups was performed and five candidates were selected using the internalization assay of ABCA1-GFP.

After performing extensive internalization assays, we finally identified Rab4B and its homologue Rab4A as promising candidates for further analyses. The over-expressed Rab4B, as well as its homologue Rab4A, actually induced the internalization and accumulation of ABCA1-GFP in HEK293 cells without a change in proteasome activity (Fig. 9B) and these Rab4 proteins are expected to regulate the endocytic pathway.

A group of Rab proteins, Rab5, Rab11 and Rab4, are known to regulate some steps of both the endocytic and exocytic pathways (Sonnichsen et al. 2000). Rab4 is localized to early endosomes and involved in the recycling of cargo such as transferrin from early endosomes back to the plasma membrane (van der Sluijs et al. 1992). For some internalized membrane proteins such as the β-adrenergic receptor or glucose transporter 4 (Glut4), their translocation to the plasma membrane are known to be facilitated by the over-expression or activation of Rab 4 (Cormont & Le Marchand-Brustel 2001; Imamura et al. 2003). In addition, using in vitro reconstitution assays, it has been demonstrated that Rab4 can regulate the formation of recycling vesicles from endosomes (Pagano et al. 2004).

In contrast, it has been reported that Rab4 over-expression increased the intracellular localization of Glut4 in adipocytes not stimulated with insulin. Cormont et al. speculated that Rab4 has a specialized function that accelerates the sorting of Glut4 in its storage compartment (Cormont et al. 1996). In addition, cell surface expression of CFTR/ABCC7, a member of the ABC transporter family, was decreased by Rab4 over-expression (Saxena et al. 2006). Thus, opposing views about the effect of Rab4 on the protein recycling have been demonstrated. Our results in this paper support the latter idea: namely, that Rab4 participates in the intracellular retention of ABCA1, but does not facilitate its recycling.

We found that ABCA1-GFP was endocytosed and accumulated into large intracellular vesicles of 1–2 µm diameter in HEK293 cells, in which HA-Rab4A or B was over-expressed. These vesicles, which contained HA-Rab4A or B, were smaller than the mature A1 bodies induced by proteasome inhibitors. Substantial accumulation of free cholesterol was detected but to a lesser degree than observed in mature A1 bodies. These results suggest that Rab4A or Rab4B accelerates the endocytosis of ABCA1 from the plasma membrane to endocytic compartments or inhibits the recycling step from endosomes to plasma membranes. However, neither Rab4A nor Rab4B activates the progression of ABCA1-accumulated endocytic vesicles to A1 bodies. As shown in Fig. 1, in the earlier phase of A1 body formation a large number of small ABCA1-positive vesicles were observed in the proteasome-treated HEK-A1WT cells. In the later phase of A1 body formation, the decrease in the numbers of ABCA1-positive vesicles was observed as larger A1 bodies were formed. In addition, as shown in Fig. 7, electron microscopic observation revealed that the smaller vesicles often aggregated and appeared to fuse to form larger A1 bodies. From these morphological data, we concluded that A1 body was derived from the smaller A1 bodies, probably corresponding to both ABCA1- and EEA1-positive small vesicles. However, KM mutant-accumulated endosomes did not grow larger even in the presence of proteasome inhibitors (Fig. 6B), suggesting that the swelling process by the aberrant accumulation of cholesterol in endosomes is necessary for the complete A1 body formation. So, we assume that both fusion process of vesicles and concurrent swelling process by cholesterol are involved in A1 body formation. Vesicle swelling or fusion might be induced by the uptake of cholesterol or lipids into the A1 bodies via ABCA1. Other factors apart from Rab4A or B might be required for the swelling or fusion step. Perhaps the genes involved in the swelling or fusion step may be included in the lactacystin- or AcLDL-inducible genes of interest described above. To test this, multi-transfection or multi-RNAi experiments are now being performed in our laboratory.

The essential role of Rab4A and B in A1-body formation, acting at the accumulation step of ABCA1 in endocytic vesicles, was confirmed by a gene silencing experiment using siRNA (Fig. 10). Knockdown of Rab4A by siRNA reduced the formation of A1 bodies. This result supports the idea that Rab4A either is essential for trapping ABCA1 in endocytic vesicles or inhibits its recycling to the plasma membrane. Under normal conditions, this Rab4A-dependent trapping may be advantageous to the transfer of cholesterol between ABCA1 and apoA-I for HDL assembly in endocytic vesicles. In contrast, under abnormal conditions, in which excess ABCA1 accumulates in A1 bodies, the aberrant trapping causes the abnormal accumulation of cholesterol or lipids in endocytic vesicles that eventually form A1 bodies. On the other hand, we detected that Rab4 over-expression perturbed the recycling of transferrin (Supplementary Fig. S6). Therefore, Rab4 (or proteasome inhibitors) may regulate recycling in nonspecific manner. It raises the possibility that the exocytosis of (pre)HDL assembled in endosomes by ABCA1 might be the cause for the cholesterol accumulation in the A1 body by Rab4 (or proteasome inhibitors). In other words, we cannot rule out the possibility that the arrest of the exocytic clearance of endo/lysosomal cholesterol plays a role in A1-body formation. For efficient cholesterol transfer to apoA-I or exocytosis of (pre)HDL, ABCA1 may need to reside in the endocytic vesicles for a particular time period, with Rab4 regulating the length of this period.

Interestingly, besides the accumulation of ABCA1-GFP in endosomal vesicles in Rab4-over-expressing cells (Fig. 9B), accumulation in the ER was observed concomitantly. Recently, Audhya et al. showed that depletion of Rab5 results in an ER tubular morphology defect in Caenorhabditis elegans (Audhya et al. 2007). The role of Rab5 in ER structure is independent of its requirement during endocytosis but dependent on its GTPase activity. Based on the report that the ER contacts several organelles such as the Golgi, the plasma membrane, lysosomes and late endosomes, they propose the hypothesis that active Rab5 on endosomes transiently interacts with effectors on ER membranes to promote their homotypic fusion (Voeltz et al. 2002). As both Rab4 and Rab5 are reported to colocalize in early endosomes, Rab4A may play a crucial role in the ER exit of ABCA1-GFP. Taken together, the effect of Rab4 on the ER retention of ABCA1 can be explained by the same ‘trans-action’ mechanism as that advocated by Audhya et al. The possibility that Rab4 is involved in ER functions such as folding or the exit of proteins needs to be tested in the future.

At this time, it is difficult to conclude that A1 body found in HEK-A1WT cells and A1-like body in macrophages stained with Sudan III are identical functionally and biochemically. However, we would like to note the similarities of A1 body in two cells. Cholesterol was accumulated in A1 (or A1-like) body, and A1 (or A1-like) body was induced by treatment of proteasome inhibitor in both HEK-A1WT cells and macrophages. Therefore, the same regulations and mechanisms may work in A1 body formation in two cells. Furthermore, DNA chip analysis shows that the up-regulated and down-regulated gene sets are almost similar in macrophages treated with lactacystin or AcLDL (Supplementary Fig. S2). AcLDL perturbs the lipid homeostasis in the cells and induces the formation of foam macrophages, which is the important initial process of atherosclerosis. This data indicates that the same signals and regulators may be involved in the formation of A1 body by lactacystin and foam cells by AcLDL. So we assume that studying the mechanisms of A1 body formation would allow us to understand the pathogenic cause of foam cell formation and regulation of lipid homeostasis in macrophage, and that HEK-A1WT cells are useful cell system to elucidate the mechanisms.

More analyses based on the micro-array data will allow us to find the common protein networks involved in the intracellular accumulation of ABCA1, and elucidate the relationship between the ubiquitin–proteasome system in normal and foam macrophages. In this paper, we focus on the correlation between the defect in ABCA1 trafficking and cholesterol metabolism in pathogenic conditions. Under the pathogenic conditions such as foam macrophage, active ABCA1 becomes rather harmful to the cells because the recycling of ABCA1 from endosomes to the plasma membrane is perturbed and it results in the aberrant accumulation of cholesterol in the cells (in endosomes). At this time, we cannot conclude whether the A1 body formation in proteasome inhibitor-treated HEK-A1WT cells are pathogenic or not, we assume that the aberrant accumulation of cholesterol (A1 body) due to ABCA1 being active in endosomes might be harmful to living cells, because the cholesterol accumulation in macrophages plays a critical role in causing atherosclerosis. Our results concerning A1 bodies suggest that the appropriate localization of ABCA1 is one of the most important factors for the correct metabolism of cholesterol and lipids.

In addition, these results lead to interesting implications about the effectiveness of preventive methods using ABCA1 as an anti-atherosclerosis factor. ABCA1 appears to be a double-edged sword. The cholesterol efflux activity of ABCA1 could be altered to a cholesterol ‘influx’ activity when proteasome activity is lower than normal. Therefore, we should not underestimate the possibility that the ABCA1 over-expression strategy may backfire in anti-atherosclerosis therapy.


   Experimental procedures
Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
References
 
Materials and antibodies

ApoA-I was purchased from Chemicon and acetylated low density lipoproteins (AcLDL) from Biomedical Technologies. EZ-Link sulfo-NHS-biotin reagents and VECTASTAIN Elite ABC kits were purchased from Pierce and Vector Laboratories, respectively. The other reagents were purchased from Sigma. Anti-ABCA1 C-terminus antibody was generated as described previously (Tanaka et al. 2003). Anti-{alpha}-tubulin antibody was purchased from Sigma and anti-EEA1 antibody from Transduction Laboratories. Anti-HA antibody was purchased from Roche. Anti-LAMP2 antibody was purchased from Iowa University. Mice were purchased from SLC, Japan.

K939M mutant ABCA1-GFP and HA-Rab4 construction

Wild-type ABCA1-GFP was constructed as described previously (Tanaka et al. 2003). The DNA fragment (FbaI-AatII) containing the K939M missense mutation was generated using the polymerase chain reaction method with the following primer pairs: (5'-CTCAGTGGCTGTGATCATCAAGGGCATCG and 5'-GTGGTCGTCaTCCCCGCTCCATTGTGGCCC):(5’-GGGC CACAATGGAGCGGGGAtGACGACCAC and 5'-CTGTCCC CCAGGACGTCCGCTTCATCCATG), where the mutated nucleotide is shown in lower case. The DNA fragment was used to replace the FbaI-AatII fragment of human ABCA1. Wild-type human Rab4A and Rab4B were each cloned from liver cDNA into the pGEM-T Easy vector using the following primer pairs: 5'-GAATTCGGATGTCGCAGACGGCCATGTCCG and 5'-CTCGAGCTAACAACCACACTCCTGAGCGTT (Rab4A) 5'-GAATTCGGATGGCTGAGACCTACGACTTC CTCTTC and 5'-CTCGAGTCAGCAGCCACACGGCTGAG GGGCCAC (Rab4B). The EcoRI-XhoI fragment was sub-cloned into the pCMV-HA vector.

Cell culture, transfection, establishment of stable transfectants and cellular cholesterol release assay

HEK293 cells stably expressing wild-type or K939M mutant ABCA1-GFP were selected and maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal calf serum (FCS), 100 U/mL penicillin G, 100 µg/mL streptomycin, 0.25 µg/mL fungizone and 300 µg/mL geneticin at 37 °C in a 5% CO2 incubator. Transfection of plasmids or siRNA was performed using Lipofectamine or Lipofectamine 2000 (Invitrogen), respectively. For cellular cholesterol release assays, samples were processed according to a previous protocol (Tanaka et al. 2003). In brief, cells cultured in 6-well plates were washed and incubated in DMEM supplemented with 0.1% bovine serum albumin and 15 µg/mL apoA-I. The lipid content in the medium was determined after a 24-h incubation as described previously (Tanaka et al. 2003). Total protein content was measured by BCA Protein Assay Kit (Pierce) and the lipid content in the medium was corrected.

Fluorescence and electron microscopy

Cells were grown on a 35-mm glass-base dish (Iwaki). For immunofluorescence, cells were fixed with 3% paraformaldehyde, permeabilized with 0.1% Triton X-100 and pre-incubated with phosphate-buffered saline (PBS) containing 5% skim milk. They were incubated with primary antibodies in PBS/5% skim milk. After being washed, they were incubated with the fluorescent-labeled secondary antibodies. The cells were directly viewed with x63 Plan-NEOFLUAR oil immersion objective or x40 C-Apochromat water immersion objective on a Zeiss confocal microscope LSM510. For filipin staining, cells were fixed in 3% paraformaldehyde, washed in PBS, quenched with 50 mM NH4Cl and incubated in PBS containing fresh filipin (0.05 mg/mL) and saponin (0.05%). After being washed, cells were visualized for filipin fluorescence using a 405 nm blue diode laser. For transferrin cycling experiments, cells incubated with 150 µg/mL transferrin Alexa Fluor 488 conjugate in DMEM/10% FCS for 30 min were fixed in 3% paraformaldehyde, washed in PBS and quenched with 50 mM NH4Cl. For Sudan III staining, mouse peritoneal macrophages were fixed with 3% paraformaldehyde. The cells were washed with PBS, followed by 50% ethanol, and then washed with 70% ethanol. They were incubated for 20 min with 0.2% Sudan III (Wako) in 70% ethanol. After being washed with 70% ethanol and PBS, cells were visualized for Sudan III fluorescence using a He–Ne 543 nm laser. For electron microscopy, samples were fixed with 2.5% glutaraldehyde at room temperature for 2 h, and processed according to a previous protocol (Hatsuzawa et al. 2000).

Purification of A1 bodies and TLC

HEK-A1WT cells were treated with 10 µM ALLN or DMSO for 24 h. Cells were washed in PBS and resuspended in purification buffer (10 mM Hepes–KOH (pH 7.4), 1.5 mM MgCl2, 10 mM KCl, 1 mM DTT, 1 mM EGTA and protease inhibitors). Cells were passed through a 22-gauge needle x30 and then centrifuged at 1000 g. The supernatant was centrifuged at 20 000 g. The pellet was resuspended in 400 µL purification buffer and incubated for 4 h with anti-GFP antibody-conjugated beads (Santa Cruz). The precipitates were washed five times with purification buffer and then dissolved in chloroform : methanol (1 : 2). The eluates were filtered and dried. Lipids were separated by TLC with a developer (chloroform : methanol : H2O : 50% aqueous ammonia = 120 : 75 : 4 : 2).

Biotinylation and immunoprecipitation

Mouse peritoneal macrophages (3 x 106 cells/well) harvested from ICR mice (retired breeders) were incubated with lactacystin or AcLDL for 3 days. The Cells were washed three times with ice-cold PBS (pH 8.0). The cells were then gently agitated at 4 °C for 30 min in PBS containing 1 mg/mL sulfo-NHS-biotin and then washed three times in PBS with 100 mM glycine. The cells were lysed on ice in lysis buffer (50 mM Tris–HCl (pH 7.4), 150 mM NaCl and 1% Triton X-100, protease inhibitors). Cell lysates were incubated for 4 h at 4 °C with anti-ABCA1 antibody and protein G-Sepharose beads (Pharmacia). The precipitates were then washed three times with lysis buffer. The proteins were separated by SDS-PAGE and analyzed by immunoblotting and VECTASTAIN Elite ABC kits.

In vitro measurement of proteasome activity

Mouse peritoneal macrophages harvested from female ICR mice (retired breeders) were lysed in lysis buffer (25 mM Tris–HCl (pH 7.8), 0.5% Triton X-100, protease inhibitors). The soluble fraction was used immediately for determining proteasome activity. The assay mixture contained 50 µL of buffer (25 mM Tris–HCl (pH 7.8), 20 mM KCl, 5 mM MgCl2, and 0.1 mM DTT), 250 µM of sLLVY-MCA (Biomol), a proteasome substrate, and 50 µL of cell lysate (containing 15 µg of protein) in the presence or absence of 10 µM clasto-lactacystin-β-lactone (clbl), a direct proteasome inhibitor. After 30 min at 37 °C, the reaction was stopped by adding 0.9 mL of 0.2 M glycine buffer (pH 10) and the fluorescence of the liberated 7-amino-4-methylcoumarin was measured (Hitachi spectrofluorometer, excitation 365 nm, emission 466 nm).

Micro-array analysis

Total RNA was prepared from mouse peritoneal macrophages treated with DMSO (control), 10 µM lactacystin or 20 µg/mL AcLDL for 3 days. GeneTrack Mouse 10 K Oligo Micro-arry (GenomicTree) targeting about 10 000 genes was used and data on 3875 genes were obtained. The expression ratio of log (AcLDL/control) to log (lactacystin/control) for each gene was plotted. This correlation coefficient was calculated as 0.7719.

Rab4 morphometric assay

HEK293 cells were co-transfected with ABCA1-GFP expression vector and Rab4A/B expression vector (1 : 3). After a 24-h incubation, cells expressing ABCA1-GFP were classified by means of the localization of ABCA1-GFP such as the ER, punctate pattern or the plasma membrane. The intermediate type cells were prioritized as the ER  > punctate pattern  > the plasma membrane.

RNA interference assay

HEK-A1WT cells were transfected with the following siRNAs (100 nM) using Lipofectamine 2000. Silencer Negative control #1 (Ambion) and the following human Rab4A specific siRNA was used: 5'-CGAUUCAGGUCCGUGACGATT (sense) and 5'-UCGUCACGGACCUGAAUCGTT (anti-sense). After a 48-h incubation, cells were treated with 5 µM MG132 for 12 h. Cells having at least one A1 body > 1 µm in diameter were counted.


   Acknowledgements
 
Authors thank Dr Hisanori Horiuchi at Kyoto University for providing Rab4 antibodies and useful discussion. A.R. Tanaka is supported by a Research Fellowship of the Japan Society for the Promotion of Science for Young Scientists. This work was supported, in part, by grants from the Ministry of Education, Culture, Sports, Science and Technology of Japan (15GS0310 to M.M. and 18770167 to F.K.).


   Footnotes
 
Communicated by: Akihiko Nakano

Correspondence: mmurata{at}bio.c.u-tokyo.ac.jp


   References
Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
 References
 
Abe-Dohmae, S., Ueda, K. & Yokoyama, S. (2006) ABCA7, a molecule with unknown function. FEBS Lett. 580, 1178–1182.[CrossRef][Medline]

Amaral, M.D. (2004) CFTR and chaperones: processing and degradation. J. Mol. Neurosci. 23, 41–48.[CrossRef][Medline]

Arakawa, R. & Yokoyama, S. (2002) Helical apolipoproteins stabilize ATP-binding cassette transporter A1 by protecting it from thiol protease-mediated degradation. J. Biol. Chem. 277, 22426–22429.[Abstract/Free Full Text]

Audhya, A., Desai, A. & Oegema, K. (2007) A role for Rab5 in structuring the endoplasmic reticulum. J. Cell Biol. 178, 43–56.[Abstract/Free Full Text]

Boadu, E. & Francis, G.A. (2006) The role of vesicular transport in ABCA1-dependent lipid efflux and its connection with NPC pathways. J. Mol. Med. 84, 266–275.[CrossRef][Medline]

Bodzioch, M., Orso, E., Klucken, J., et al. (1999) The gene encoding ATP-binding cassette transporter 1 is mutated in Tangier disease. Nat. Genet. 22, 347–351.[CrossRef][Medline]

Brooks-Wilson, A., Marcil, M., Clee, S.M., et al. (1999) Mutations in ABC1 in Tangier disease and familial high-density lipoprotein deficiency. Nat. Genet. 22, 336–345.[CrossRef][Medline]

Chen, W., Sun, Y., Welch, C., Gorelik, A., Leventhal, A.R., Tabas, I. & Tall, A.R. (2001) Preferential ATP-binding cassette transporter A1-mediated cholesterol efflux from late endosomes/lysosomes. J. Biol. Chem. 276, 43564–43569.[Abstract/Free Full Text]

Cormont, M. & Le Marchand-Brustel, Y. (2001) The role of small G-proteins in the regulation of glucose transport. Mol. Membr. Biol. 18, 213–220.[CrossRef][Medline]

Cormont, M., Bortoluzzi, M.N., Gautier, N., Mari, M., van Obberghen, E. & Le Marchand-Brustel, Y. (1996) Potential role of Rab4 in the regulation of subcellular localization of Glut4 in adipocytes. Mol. Cell. Biol. 16, 6879–6886.[Abstract]

Dean, M. (2005) The genetics of ATP-binding cassette transporters. Methods Enzymol. 400, 409–429.[Medline]

Hamon, Y., Broccardo, C., Chambenoit, O., Luciani, M.F., Toti, F., Chaslin, S., Freyssinet, J.M., Devaux, P.F., McNeish, J., Marguet, D. & Chimini, G. (2000) ABC1 promotes engulfment of apoptotic cells and transbilayer redistribution of phosphatidylserine. Nat. Cell Biol. 2, 399–406.[CrossRef][Medline]

Hatsuzawa, K., Hirose, H., Tani, K., Yamamoto, A., Scheller, R.H. & Tagaya, M. (2000) Syntaxin 18, a SNAP receptor that functions in the endoplasmic reticulum, intermediate compartment, and cis-Golgi vesicle trafficking. J. Biol. Chem. 275, 13713–13720.[Abstract/Free Full Text]

Imamura, T., Huang, J., Usui, I., Satoh, H., Bever, J. & Olefsky, J.M. (2003) Insulin-induced GLUT4 translocation involves protein kinase C-{lambda}-mediated functional coupling between Rab4 and the motor protein kinesin. Mol. Cell. Biol. 23, 4892–4900.[Abstract/Free Full Text]

Johnston, J.A., Ward, C.L. & Kopito, R.R. (1998) Aggresomes: a cellular response to misfolded proteins. J. Cell Biol. 143, 1883–1898.[Abstract/Free Full Text]

Jones, M.C., Caswell, P.T. & Norman, J.C. (2006) Endocytic recycling pathways: emerging regulators of cell migration. Curr. Opin. Cell Biol. 18, 549–557.[CrossRef][Medline]

Kerem, B.S., Buchanan, J.A., Durie, P., Corey, M.L., Levison, H., Rommens, J.M., Buchwald, M. & Tsui, L.C. (1989) DNA marker haplotype association with pancreatic sufficiency in cystic fibrosis. Am. J. Hum. Genet. 44, 827–834.[Medline]

MacQueen, J.B. (1967) Some methods for classification and analysis of multivariate observations. Proceedings of 5th Berkeley Symposium on Mathematical Statistics and Probability, University of California Press, Berkely, CA, pp. 281–297.

McNeish, J., Aiello, R.J., Guyot, D., Turi, T., Gabel, C., Aldinger, C., Hoppe, K.L., Roach, M.L., Royer, L.J., de Wet, J., Broccardo, C., Chimini, G. & Francone, O.L. (2000) High density lipoprotein deficiency and foam cell accumulation in mice with targeted disruption of ATP-binding cassette transporter-1. Proc. Natl. Acad. Sci. USA 97, 4245–4250.[Abstract/Free Full Text]

Mendez, A.J. (1997) Cholesterol efflux mediated by apolipoproteins is an active cellular process distinct from efflux mediated by passive diffusion. J. Lipid Res. 38, 1807–1821.[Abstract]

Nagata, K., Yamamoto, A., Ban, N., Tanaka, A.R., Matsuo, M., Kioka, N., Inagaki, N. & Ueda, K. (2004) Human ABCA3, a product of a responsible gene for abca3 for fatal surfactant deficiency in newborns, exhibits unique ATP hydrolysis activity and generates intracellular multilamellar vesicles. Biochem. Biophys. Res. Commun. 324, 262–268.[CrossRef][Medline]

Neufeld, E.B., Remaley, A.T., Demosky, S.J., Stonik, J.A., Cooney, A.M., Comly, M., Dwyer, N.K., Zhang, M., Blanchette-Mackie, J., Santamarina-Fojo, S. & Brewer, H.B. Jr. (2001) Cellular localization and trafficking of the human ABCA1 transporter. J. Biol. Chem. 276, 27584–27590.[Abstract/Free Full Text]

Neufeld, E.B., Stonik, J.A., Demosky, S.J. Jr., Knapper, C.L., Combs, C.A., Cooney, A., Comly, M., Dwyer, N., Blanchette-Mackie, J., Remaley, A.T., Santamarina-Fojo, S. & Brewer, H.B. Jr (2004) The ABCA1 transporter modulates late endocytic trafficking: insights from the correction of the genetic defect in Tangier disease. J. Biol. Chem. 279, 15571–15578.[Abstract/Free Full Text]

Oram, J.F. & Yokoyama, S. (1996) Apolipoprotein-mediated removal of cellular cholesterol and phospholipids. J. Lipid Res. 37, 2473–2491.[Abstract]

Orso, E., Broccardo, C., Kaminski, W.E., et al. (2000) Transport of lipids from golgi to plasma membrane is defective in tangier disease patients and ABC1-deficient mice. Nat. Genet. 24, 192–196.[CrossRef][Medline]

Pagano, A., Crottet, P., Prescianotto-Baschong, C. & Spiess, M. (2004) In vitro formation of recycling vesicles from endosomes requires adaptor protein-1/clathrin and is regulated by rab4 and the connector rabaptin-5. Mol. Biol. Cell 15, 4990–5000.[Abstract/Free Full Text]

Rust, S., Rosier, M., Funke, H., Real, J., Amoura, Z., Piette, J.C., Deleuze, J.F., Brewer, H.B., Duverger, N., Denèfle, P. & Assmann, G. (1999) Tangier disease is caused by mutations in the gene encoding ATP-binding cassette transporter 1. Nat. Genet. 22, 352–355.[CrossRef][Medline]

Saxena, S.K., Kaur, S. & George, C. (2006) Rab4GTPase modulates CFTR function by impairing channel expression at plasma membrane. Biochem. Biophys. Res. Commun. 341, 184–191.[CrossRef][Medline]

Smith, J.D., Waelde, C., Horwitz, A. & Zheng, P. (2002) Evaluation of the role of phosphatidylserine translocase activity in ABCA1-mediated lipid efflux. J. Biol. Chem. 277, 17797–17803.[Abstract/Free Full Text]

Sonnichsen, B., De Renzis, S., Nielsen, E., Rietdorf, J. & Zerial, M. (2000) Distinct membrane domains on endosomes in the recycling pathway visualized by multicolor imaging of Rab4, Rab5, and Rab11. J. Cell Biol. 149, 901–914.[Abstract/Free Full Text]

Takahashi, K., Kimura, Y., Kioka, N., Matsuo, M. & Ueda, K. (2006) Purification and ATPase activity of human ABCA1. J. Biol. Chem. 281, 10760–10768.[Abstract/Free Full Text]

Takahashi, K., Kimura, Y., Nagata, K., Yamamoto, A., Matsuo, M. & Ueda, K. (2005) ABC proteins: key molecules for lipid homeostasis. Med. Mol. Morphol. 38, 2–12.[CrossRef][Medline]

Takahashi, Y. & Smith, J.D. (1999) Cholesterol efflux to apolipoprotein AI involves endocytosis and resecretion in a calcium-dependent pathway. Proc. Natl. Acad. Sci. USA 96, 11358–11363.[Abstract/Free Full Text]

Tanaka, A.R., Abe-Dohmae, S., Ohnishi, T., Aoki, R., Morinaga, G., Okuhira, K., Ikeda, Y., Kano, F., Matsuo, M., Kioka, N., Amachi, T., Murata, M., Yokoyama, S. & Ueda, K. (2003) Effects of mutations of ABCA1 in the first extracellular domain on subcellular trafficking and ATP binding/hydrolysis. J. Biol. Chem. 278, 8815–8819.[Abstract/Free Full Text]

Tanaka, A.R., Ikeda, Y., Abe-Dohmae, S., Arakawa, R., Sadanami, K., Kidera, A., Nakagawa, S., Nagase, T., Aoki, R., Kioka, N., Amachi, T., Yokoyama, S. & Ueda, K. (2001) Human ABCA1 contains a large amino-terminal extracellular domain homologous to an epitope of Sjogren's Syndrome. Biochem. Biophys. Res. Commun. 283, 1019–1025.[CrossRef][Medline]

van der Sluijs, P., Hull, M., Webster, P., Male, P., Goud, B. & Mellman, I. (1992) The small GTP-binding protein rab4 controls an early sorting event on the endocytic pathway. Cell 70, 729–740.[CrossRef][Medline]

Vieira, O., Escargueil-Blanc, I., Jurgens, G., Borner, C., Almeida, L., Salvayre, R. & Nègre-Salvayre, A. (2000) Oxidized LDLs alter the activity of the ubiquitin-proteasome pathway: potential role in oxidized LDL-induced apoptosis. FASEB J. 14, 532–542.[Abstract/Free Full Text]

Voeltz, G.K., Rolls, M.M. & Rapoport, T.A. (2002) Structural organization of the endoplasmic reticulum. EMBO Rep. 3, 944–950.[CrossRef][Medline]

Weng, J., Mata, N.L., Azarian, S.M., Tzekov, R.T., Birch, D.G. & Travis, G.H. (1999) Insights into the function of Rim protein in photoreceptors and etiology of Stargardt's disease from the phenotype in ABCR knockout mice. Cell 98, 13–23.[CrossRef][Medline]

Yan, F., Lin, C.W., Weisiger, E., Cartier, E.A., Taschenberger, G. & Shyng, S.L. (2004) Sulfonylureas correct trafficking defects of ATP-sensitive potassium channels caused by mutations in the sulfonylurea receptor. J. Biol. Chem. 279, 11096–11105.[Abstract/Free Full Text]

Received: 4 December 2007
Accepted: 14 May 2008





This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Tanaka, A. R.
Right arrow Articles by Murata, M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Tanaka, A. R.
Right arrow Articles by Murata, M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS