ABCA1 contributes to macrophage deposition of extracellular cholesterol.

We previously reported that cholesterol-enriched macrophages excrete cholesterol into the extracellular matrix. A monoclonal antibody that detects cholesterol microdomains labels the deposited extracellular particles. Macrophage deposition of extracellular cholesterol depends, in part, on ABCG1, and this cholesterol can be mobilized by HDL components of the reverse cholesterol transport process. The objective of the current study was to determine whether ABCA1 also contributes to macrophage deposition of extracellular cholesterol. ABCA1 functioned in extracellular cholesterol deposition. The liver X receptor agonist, TO901317 (TO9), an ABCA1-inducing factor, restored cholesterol deposition that was absent in cholesterol-enriched ABCG1−/− mouse macrophages. In addition, the ABCA1 inhibitor, probucol, blocked the increment in cholesterol deposited by TO9-treated wild-type macrophages, and completely inhibited deposition from TO9-treated ABCG1−/− macrophages. Lastly, ABCA1−/− macrophages deposited much less extracellular cholesterol than wild-type macrophages. These findings demonstrate a novel function of ABCA1 in contributing to macrophage export of cholesterol into the extracellular matrix.


Male ABCG1
Ϫ / Ϫ mice on a C57BL/6J background were generated as described previously ( 25 ). Female ABCA1 Ϫ / Ϫ mice were generated from DBA/1-Abca1 tm1Jdm /J mice (#003897) obtained from Jackson Laboratory (Bar Harbor, ME). These mice were of mixed genetic background. The ABCA1 mutation was transferred to a C57BL/6N background by 10 consecutive crossings with C57BL/6N. Wild-type C57BL/6 control mice were substrain-, sex-and age-matched to ABCG1 Ϫ / Ϫ and ABCA1 Ϫ / Ϫ mice. Animal studies were conducted in conformity with the Public Health Service Policy on Humane Care and Use of Laboratory Animals, and were approved by the National Heart, Lung, and Blood Institute Institutional Animal Care and Use Committee . For culture of bone marrow-derived macrophages, femurs and tibias were isolated from mice and muscle was removed. Both ends of the bones were cut with scissors and then fl ushed with 5 ml of RMPI-1640 with a 25 gauge needle. Bone marrow cells were centrifuged and resuspended at a concentration of 4-6 × 10 6 cells/ml in 1 ml of freezing medium containing 90% FBS and 10% DMSO ( 26 ). Cells were stored in liquid nitrogen until use.
On the day of use, cells were thawed and suspended in 30 ml RPMI-1640 medium containing 100 U/ml penicillin, 0.1 mg/ml streptomycin, and 2 mM L-glutamine before centrifugation to remove DMSO. Then, cells were resuspended at a concentration of 1 × 10 5 cells/ml RPMI-1640 medium containing 100 U/ml penicillin, 0.1 mg/ml streptomycin, 2 mM L-glutamine, 10% FBS, and 50 ng/ml macrophage colony-stimulating factor (complete medium). Cells were seeded in a 75 cm 2 culture fl ask and incubated in a 37°C cell culture incubator with 5% CO 2 /95% air. On day 3, cultures were rinsed three times with RPMI-1640 medium containing 100 U/ml penicillin, 0.1 mg/ml streptomycin, and 2 mM L-glutamine, and then cultured in fresh complete medium. Medium was changed every 2 days until suffi cient macrophages had grown in the fl ask, which usually occurred by the seventh day. Next, experiments were initiated by harvesting macrophages at room temperature with 10 ml 0.25% trypsin-EDTA solution. After about 20-30 min, macrophages rounded, but mostly remained attached. Trypsinization was stopped by addition of 10 ml RMPI-1640 containing 100 U/ml penicillin, 0.1 mg/ml streptomycin, 2 mM L-glutamine, and 10% FBS. A cell lifter was used to retrieve macrophages from the culture surface. The cell suspension was centrifuged at 300 g for 5 min and the resulting cell pellet was resuspended in 1 ml complete medium. Macrophages were counted with a hemocytometer. 0.6 × 10 5 macrophages per milliliter were cultured in 12-well CellBIND culture plates containing 1.5 ml of complete medium per well. Macrophages were incubated overnight before experiments were initiated with complete medium and the indicated additions, but without FBS. Experimental incubations were carried out for 4 days with the medium and additions refreshed after 2 days.

Immunostaining of macrophages
Fixation, immunostaining, and microscopy were all performed with macrophages in their original CellBIND culture plates, and all steps were carried out at room temperature. Macrophage cultures were rinsed three times (5 min each rinse this and all subsequent times) in DPBS, fi xed for 10 min with 4% paraformaldehyde in DPBS, and then rinsed an additional three times in DPBS. Macrophages were then incubated 60 min with 5 g/ml purifi ed mouse anti-cholesterol microdomain MAb 58B1 IgM diluted in DPBS containing 0.1% BSA. Control staining was performed with 5 g/ml of an irrelevant purifi ed mouse anti-Clavibacter michiganense MAb (clone 9A1) IgM diluted in DPBS containing 0.1% BSA. MAb IgM fractions were purifi ed as previously described ( 16 ). Cultures were then rinsed three times in DPBS, followed by to HDL for potential reverse cholesterol transport (15)(16)(17). In those studies, we employed a unique monoclonal antibody (MAb 58B1) that labels cholesterol microdomains formed when cholesterol reaches high concentrations within membranes. While the MAb labels cholesterol crystals and cholesterol monolayers, it does not label individual cholesterol molecules (18)(19)(20)(21)(22). Thus, the antibody recognizes a structural motif presented by an ordered array of cholesterol molecules. Such ordered arrays of cholesterol, in the form of cholesterol crystalline microdomains, have been demonstrated with small-angle X-ray diffraction in biological membranes, including the membranes of cells isolated from atherosclerotic plaques ( 23,24 ).
With this anti-cholesterol microdomain antibody, we have shown that cultured cholesterol-enriched macrophages excrete spherical particles containing cholesterol microdomains into the extracellular matrix ( 17 ). Similar extracellular spherical particles labeled by the anti-cholesterol microdomain antibody are found in human atherosclerotic lesions ( 17 ). These extracellular particles showing cholesterol microdomains may function as an extracellular storage form of cholesterol. Macrophage deposition of cholesterol into the extracellular matrix could help maintain cholesterol homeostasis when macrophages accumulate excess cholesterol beyond that which can be stored in intracellular lipid droplets and cell membranes.
We previously reported that ABCG1 contributes to macrophage generation of these extracellular particles showing cholesterol microdomains ( 15 ). In the current work, we show that ABCA1 also contributes to generation of these extracellular deposited cholesterol microdomains. The generation of cholesterol microdomains may facilitate ABCA1-and ABCG1-mediated transport of cholesterol to HDL.
Cholesterol-enriched ABCG1 Ϫ / Ϫ mouse macrophages (AcLDL treatment in Fig. 2 ) showed very little extracellular cholesterol deposition, as we reported previously ( 15 ). However, when stimulated with TO9 (AcLDL + TO9 treatment in Fig. 2 ), these macrophages then deposited cholesterol into the extracellular matrix (7.6-fold more MAb 58B1 immunofl uorescence compared with macrophages incubated with AcLDL without TO9). Given that TO9 is known to upregulate ABCA1 expression ( 29 ), that probucol inhibits human macrophage extracellular cholesterol a 30 min incubation in 5 g/ml biotinylated goat anti-mouse IgM diluted in DPBS containing 0.1% BSA. After three rinses in DPBS, cultures were incubated 10 min with 10 g/ml streptavidin Alexa Fluor 488 diluted in DPBS. Cultures were then rinsed three times with DPBS and mounted in Vectashield hard-set mounting medium with DAPI nuclear stain in preparation for digital imaging using an Olympus IX81 fl uorescence microscope. Because macrophages were not permeabilized, MAb 58B1 staining represents cell surface or extracellular staining. No staining was observed when the control MAb was substituted for the anticholesterol microdomain MAb.

Microscopic analysis
Cells were identifi ed using phase-contrast microscopy, or by locating DAPI-stained nuclei. The pattern and intensity of MAb 58B1 staining were then analyzed for cultures from each experimental parameter, and these data were compared with one another. We considered MAb 58B1 labeling cellular if it was located within cell membrane boundaries, as identifi ed on the corresponding phase-contrast view. Labeling was considered extracellular if it was located outside the cell membrane boundaries seen on phase-contrast view. Different planes of focus were visualized before acquiring images to confi rm that only a monolayer of cells was present, thereby ensuring that labeling seen outside cell membrane boundaries did not represent cellular labeling from cells lying in a different plane of focus. As we reported before ( 15 ), MAb 58B1 labeling of mouse macrophage cultures showed extracellular rather than plasma membrane staining. The immunostained cells shown in the fi gures are representative of a minimum of fi ve microscopic fi elds viewed in one culture well.

Quantifi cation and statistical analysis of MAb 58B1 immunofl uorescence
For each condition shown in the fi gures, including additional control images where macrophages were incubated without AcLDL, we quantifi ed MAb 58B1 immunofl uorescence in three separate digital images using Image J software (version 1.37) developed by the National Institutes of Health. Control image fl uorescence values were subtracted from non-control image fl uor escence values. Statistical analysis of the obtained fl uorescence data was carried out with SigmaPlot for Windows (version 11.0). One-way ANOVA using the Holm-Sidak method was employed for comparisons of three groups ( Figs. 1-3 ), and the unpaired t -test was used for comparison of two groups ( Figs. 4-6 ). P р 0.05 was considered signifi cant.

RESULTS
In an earlier study, we observed that cholesterol-enriched ABCG1 Ϫ / Ϫ mouse bone marrow-derived macrophages, in contrast to cholesterol-enriched wild-type mouse bone marrow-derived macrophages, excreted very little cholesterol into the extracellular matrix ( 15 ). This suggested that ABCG1 mediated the extracellular cholesterol deposition process in the mouse. However, probucol inhibited cholesterol-enriched human monocyte-derived macrophage deposition of extracellular cholesterol ( 15 ). Because probucol inhibits ABCA1 ( 27,28 ), this suggested the possibility that besides ABCG1, ABCA1 may also contribute to macrophage deposition of extracellular cholesterol. Although cholesterol-enriched wild-type mouse bone marrow-derived macrophages showed extracellular cholesterol deposition (AcLDL treatment in Fig. 1 ), the Fig. 1. Cholesterol-enriched wild-type macrophages deposit extracellular cholesterol without TO9, but deposition is increased with TO9. Wild-type mouse bone marrow-derived macrophages were incubated for 4 days with either TO9 (5 M), AcLDL (50 g/ml), or AcLDL + TO9 before cultures were immunostained with anticholesterol microdomain MAb 58B1 (green fl uorescence) and DAPI nuclear stain (blue fl uorescence). Upper and lower rows are, respectively, the fl uorescence and phase photomicrographs. Scale bar = 50 m and applies to all. was partially reduced in TO9-treated cholesterol-enriched ABCA1 Ϫ / Ϫ macrophages compared with TO9-treated cholesterol-enriched wild-type macrophages ( Fig. 5 ). MAb 58B1 immunofl uorescence levels of ABCA1 Ϫ / Ϫ macrophages were 34 ± 2% of wild-type macrophages. This partial reduction would be expected in TO9-treated cholesterol-enriched deposition ( 15 ), and that TO9 stimulates extracellular cholesterol deposition by ABCG1 Ϫ / Ϫ mouse macrophages ( Fig. 2 ), we considered the possibility that with TO9 treatment, ABCA1, in addition to ABCG1, would contribute to macrophage extracellular cholesterol deposition . If ABCA1 contributes to extracellular cholesterol deposition by TO9-treated cholesterol-enriched mouse macrophages, then probucol should inhibit the component of cholesterol deposition stimulated by TO9, as probucol inhibits ABCA1 but does not inhibit ABCG1 ( 27,28,30 ). We tested this by incubating cholesterol-enriched wild-type mouse macrophages with TO9 in the presence and absence of probucol. We observed that TO9 increased MAb 58B1 immunofl uorescence 2.5-fold compared with macrophages incubated with AcLDL alone ( Fig. 3 ). However, when probucol was added to AcLDL + TO9, there was no signifi cant difference in MAb 58B1 immunofl uorescence compared with macrophages treated with AcLDL alone. Thus, probucol blocked the increment of macrophage extracellular cholesterol deposition that was stimulated by TO9, consistent with ABCA1 mediating a portion of this macrophage cholesterol deposition.
We further tested the function of ABCA1 in macrophage cholesterol deposition by incubating TO9-treated cholesterol-enriched ABCG1 Ϫ / Ϫ macrophages with probucol. Given that ABCG1 would not be contributing to cholesterol deposition in these macrophages, we expected that probucol should completely block macrophage cholesterol deposition through its inhibition of ABCA1. That is what we observed, in that probucol blocked the TO9-stimulated cholesterol deposition that occurred with cholesterolenriched ABCG1 Ϫ / Ϫ macrophages ( Fig. 4 ). Quantifi ed MAb 58B1 immunofl uorescence levels for macrophages incubated with AcLDL + TO9 + probucol were similar to macrophages incubated with AcLDL (not shown).
Next, we directly confi rmed that ABCA1 functioned in macrophage cholesterol deposition. Cholesterol deposition   Ϫ / Ϫ mouse bone marrow-derived macrophages were incubated for 4 days with AcLDL (50 g/ml) + TO9 (5 M) before cultures were immunostained with anti-cholesterol microdomain MAb 58B1 (green fl uorescence) and DAPI nuclear stain (blue fl uorescence). Upper and lower rows are, respectively, the fl uorescence and phase photomicrographs. Scale bar = 50 m and applies to all. This experiment was repeated two additional times with similar results. extracellular cholesterol, which can occur even in the absence of macrophages ( 17 ). Although both ABCA1 and ABCG1 mediated deposition of cholesterol into the extracellular matrix, they functioned independently. Absence of one or the other did not eliminate cholesterol deposition by TO9-treated cholesterol-enriched macrophages. A block in cholesterol deposition would be expected if the two proteins were functioning in a sequential fashion. Rather, elimination of either protein partially decreased the extent of cholesterol deposition compared with that occurring with TO9-treated cholesterol-enriched wild-type macrophages. Thus, there was an additive effect of ABCA1 and ABCG1 in mediating macrophage cholesterol deposition. Similarly, ABCA1 and ABCG1 produce an additive effect in their mediation of reverse cholesterol transport in vivo ( 3 ).
ABCA1 and ABCG1 induction of cholesterol microdomains that label with MAb 58B1 in the plasma membrane of human macrophages and the extracellular matrix surrounding human and mouse macrophages may occur through enrichment of the plasma membrane with cholesterol ( 8,33 ). Cholesterol microdomains form in both model and cell membranes when these membranes are enriched with cholesterol ( 23,(32)(33)(34)(35). This is due to lateral phase separation of cholesterol within the membrane as certain critical membrane cholesterol concentrations are reached. We previously showed that SU6656, a Src kinase inhibitor, causes human macrophage cholesterol microdomains to accumulate in association with the plasma membrane rather than deposit into the extracellular matrix ( 17 ). Thus, there could be a two-step process in which ABCA1 and ABCG1 mediate transport of cholesterol to the plasma membrane, and then some other process mediates shedding of these microdomains into the extracellular matrix. In support of an independently regulated two-step process, we have observed that cholesterol-enrichment of fi broblasts induces plasma membrane-associated cholesterol microdomains that do not shed ( 16 ). Furthermore, cholesterol enrichment of human macrophages grown on certain substrates also blocks the shedding process of plasma membrane cholesterol microdomains detected with MAb 58B1 (unpublished observation).
The cholesterol microdomains we detect here in the extracellular matrix could be related to previously observed plasma membrane-associated structures that form with cholesterol enrichment of cells or increased expression of cellular ABCA1. Upregulation of ABCA1 in fi broblasts and macrophages induces the formation of ApoA-I binding to plasma membrane-associated ( р 200 nm diameter generally spherical) structures ( 34,35 ). Possibly also related to the extracellular lipid particles that we have observed are the lipid-containing binding sites for ApoA-I that underlie cultured J774 mouse and THP-1 human macrophages ( 36 ).
Without liver X receptor stimulation of cholesterol effl ux ABC transporters, ABCA1 mediates extracellular cholesterol deposition by human macrophages as deposition is eliminated by probucol, an ABCA1 inhibitor ( 17 ), while ABCG1 mediates extracellular cholesterol deposition by ABCA1 Ϫ / Ϫ macrophages if both ABCA1 and ABCG1 were contributing to macrophage cholesterol deposition by TO9-treated cholesterol-enriched wild-type macrophages. This is because TO9 stimulation of ABCA1 could not occur with the ABCA1 Ϫ / Ϫ macrophages.
Lastly, we tested the effect of probucol on cholesterol deposition by TO9-treated cholesterol-enriched ABCA1 Ϫ / Ϫ macrophages. If probucol's effect of inhibiting macrophage cholesterol deposition was mediated by ABCA1, then we would expect no effect of probucol on macrophage cholesterol deposition by these macrophages. Indeed, that is what we observed ( Fig. 6 ). There was no quantitative difference between MAb 58B1 immunofl uorescence of macrophages incubated with AcLDL + TO9 and macrophages incubated with AcLDL + TO9 + probucol.

DISCUSSION
While in some studies ABCA1 mediates cholesterol effl ux to mature HDL as well as nascent HDL ( 11,31 , and references contained therein), this cholesterol effl ux has been attributed to ApoA-I that dissociates from the mature HDL and then interacts with ABCA1 generating nascent HDL. In this scenario, nascent HDL functions as the true cholesterol acceptor ( 32 ). Recently, this point of view has been challenged based on no evidence for dissociation of ApoA-I from HDL3b particles, a very effi cient acceptor of cholesterol effl uxed by ABCA1 ( 31 ). Previously we reported that ABCG1 mediates macrophage deposition of cholesterol into the extracellular matrix ( 15 ). Our new fi nding that ABCA1 as well as ABCG1 mediate macrophage deposition of cholesterol into the extracellular matrix can explain how ABCA1 and ABCG1 both mediate cholesterol effl ux to mature HDL: by mature HDL mobilizing Ϫ / Ϫ macrophages. ABCA1 Ϫ / Ϫ mouse bone marrow-derived macrophages were incubated for 4 days with AcLDL (50 g/ml) + TO9 (5 M) without or with probucol (10 M) before cultures were immunostained with anti-cholesterol microdomain MAb 58B1 (green fl uorescence) and DAPI nuclear stain (blue fl uorescence). Upper and lower rows are, respectively, the fl uorescence and phase photomicrographs. Scale bar = 50 m and applies to all. mouse macrophages ( 15 ). A similar difference in mouse and human macrophage effl ux to HDL has been reported ( 11 ( 11 )].
In conclusion, we have shown that in addition to ABCG1, ABCA1 independently mediates deposition of cholesterol into the extracellular matrix by cholesterol-enriched macrophages. Our fi ndings show a novel function for both ABCA1 and ABCG1 that results in excretion of cholesterol from the cell that is not mediated by formation of classical HDL cholesterol acceptor lipoproteins. While macrophage export of excess cholesterol into the extracellular matrix may be a protective cellular mechanism, if not mobilized through reverse cholesterol transport, buildup of this extracellular cholesterol possibly promotes atherosclerosis.